MEDICAL Volume 24, Number 1

March 2022

ULTRASONOGRAPHY
A N I N T E R N A T I O N A L J O U R N A L O F C L I N I C A L I M A G I N G

„ Safety and parents´ acceptance of ultrasound contrast agents

in children and adolescents – contrast enhanced voiding
urosonography and contrast enhanced ultrasound
„ Ultrasound of the chest and mediastinum in children,
inter ventions and artefacts. WFUMB review paper (part 3)
„ Cystic renal diseases: role of ultrasound. Part II, genetic cystic
renal diseases
LOGIQ P10 XDclear
™ ™

UȘUREAZĂ-ȚI SARCINA. IA-ȚI ȘI TU.

 Meet the ACUSON
US-20-NAM-571

family of
ultrasound systems

ACUSON Sequoia ACUSON Redwood ACUSON Juniper

Ultra-Premium
ltra-Premium ultrasound system
that delivers more image clarity
and penetration than
conventional ultrasound
ltrasound systems.
Premium ultrasound system with a High-performance ultrasound system
lightweight portable e design to with the form and function to
deliver better clinical
linical outcomes operate in a variety of clinical areas
against the restraints
estraints of a tight without sacrificing performance.
budget.

Siemens Healthineers is ushering in a new era in ultrasound, dedicated

to providing the best imaging for early detection, diagnosis, and timely
treatment for the diverse needs of medical organizations across the globe.

Welcome to a new era in ultrasound.

Ask your Siemens Healthineers representative for more information.

siemens-healthineers.com/ultrasound
 Contents

Editorial
Lung ultrasound for ever
Ch.F Dietrich .................................................................................................................................................................................5
Original papers
Feasibility of pneumoperitoneum diagnosis using point-of-care ultrasound: a pilot study using a fresh cadaver model
M.K. Herbst, E.M. Carter, S. Wu, C.F. Dietrich, B. Hoffmann .......................................................................................................7
Ultrasonographic characteristics and outcome of Type III umbilical-portal-systemic venous shunt
L. Zhu, H. Wu, X. Cong, Z. Ma, G. Tao ........................................................................................................................................14
The role of lung ultrasonography in predicting the clinical outcome of complicated community-acquired
pneumonia in hospitalized children
M.D. Ionescu, M. Balgradean, C. Filip, R. Taras, G.M. Capitanescu, F. Berghea, C.E. Berghea, C.G. Cirstoveanu ............... 19
Safety and parents´ acceptance of ultrasound contrast agents in children and adolescents – contrast enhanced
voiding urosonography and contrast enhanced ultrasound
J. Seelbach, P.C. Krüger, M. Waginger, D.M. Renz, H-J. Mentzel ...............................................................................................27
Multiparametric ultrasound in torsion of the testicular appendages: a reliable diagnostic tool?
G. Laimer, R. Müller, C. Radmayr, A.K. Lindner, A. Lebovici, F. Aigner .....................................................................................33
Effects of local anaesthetic dilution on the characteristics of ultrasound guided axillary brachial plexus block:
a randomised controlled study
A. Ranganath, O. Ahmed, G. Iohom .............................................................................................................................................38
An ultrasound study of the long posterior sacroiliac ligament in healthy volunteers and in patients
with noninflammatory sacroiliac joint pain
P. Todorov, L. Mekenjan, R. Nestorova, A. Batalov ......................................................................................................................44
Doppler ultrasonographic evaluation of radial and ulnar artery diameters and blood flow,
before and after percutaneous coronary interventions
Y. Gündüz, H. Gunduz, O.F. Ates, M. Ciner, A. Cakmak, C. Akcay., E. Ilguz., K. Cosansu .........................................................52
Comparison of the effects of adenosine, isoproterenol and their combinations on pulmonary transit time in rats
using contrast echocardiography
F. Su, Y-Y. Shi, B. Wang, X.-Z. Zheng .......................................................................................................................................... 58
Reviews
Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
C. Fang, J. Jaworska, N. Buda, I.M. Ciuca, Y. Dong, A Feldkamp, J. Jüngert, W. Kosiak, H.J. Mentzel, C. Pienar,
J.S. Rabat, V. Rafailidis, S. Schrading, D. Schreiber-Dietrich, C.F. Dietrich ..............................................................................65
Transvaginal three-dimensional ultrasound for preoperative assessment of myometrial invasion
in patients with endometrial cancer: a systematic review and meta-analysis
T. Costas, R. Belda, J.L. Alcazar ..................................................................................................................................................77
Diagnostic value of endobronchial ultrasound elastography for differentiating benign and malignant hilar and
mediastinal lymph nodes: a systematic review and meta-analysis
J. Wu, Y. Sun, Y. Wang, L. Ge, Y. Jin, Z. Wang ..............................................................................................................................85
How to perform shear wave elastography. Part I
G. Ferraioli, R.G. Barr, A. Farrokh, M. Radzina, X.W. Cui, Y. Dong, L. Rocher, V. Cantisani, E. Polito, M. D’Onofrio,
D. Roccarina, Y. Yamashita, M.K. Dighe, C.F. Dietrich ..............................................................................................................95
Pictorial essay
Cystic renal diseases: role of ultrasound. Part II, genetic cystic renal diseases
A. Kabaalioglu, N. Gunduz, A. Keven, E. Durmaz, M. Aslan, A. Aslan, S. Guneyli ..................................................................107
 Medical Ultrasonography
Official Journal of the Romanian Society for Ultrasonography in Medicine and Biology
Medical Ultrasonography (formerly Revista Româna de Ultrasonografie from 1999 to 2008) is the official publication of the
Romanian Society for Ultrasonography in Medicine and Biology (SRUMB). Starting with 2008 the entire content of Medical
Ultrasonography is published in English, quarterly. The journal aims to promote ultrasound diagnosis by publishing papers in a
variety of categories, including Original papers, Review Articles, Pictorial Essays, Technical Innovations, Case Report, or Letters to
the Editor (fundamental as well as methodological and educational papers). The published papers cover a wide variety of discipline
of ultrasound. The journal also host information regarding the society’s activities, the scheduling of accredited training courses in
ultrasound diagnosis, as well as the agenda of national and international scientific events.
Medical Ultrasonography is now listed in Science Citation Index Expanded/ ISI Thomson Master Journal List, Medline/
PubMed, Scopus, Pro Quest, Ebsco, and Index Copernicus data bases. Impact Factor 1.611 (JCR 2020); 5 year IF=1.824
Editorial Office
2nd Medical Clinic, 2-4 Clinicilor str., 400006 Cluj-Napoca, Romania
Tel.: +4 0264 591942/442, Fax: +4 0264 596912, Email: medultrasonography@gmail.com
Contact person: Daniela Fodor, email: dfodor@ymail.com
Journal web site: http://www.medultrason.ro
Editorial board
Editor in Chief Methodological adviser Editors Assistant Editors English language editors
Daniela Fodor Petru Adrian Mircea Radu Ion Badea Carolina Solomon Sally Wood-Lamont
Sorin Marian Dudea Bogdan Chis Ioana Robu
Oana Serban
Members
Mihaela Băciuţ (Cluj-Napoca, Romania) Richard Hoppmann (Columbia, South Carolina, USA Alina Popescu (Timişoara, Romania)
Boris Brkljacic (Zagreb, Croatia) Walter Grassi (Ancona, Italy) Alper Ozel (Istambul, Turkey)
Ciprian Brisc (Oradea, Romania) Lucas Greiner (Wuppertal, Germany) Adrian Săftoiu (Craiova, Romania)
Vito Cantisani (Rome, Italy) Norbert Gritzmann (Salzburg, Austria) Paul Singh Sidhu (London, UK)
Anca Ciurea (Cluj-Napoca, Romania) Zoltán Harkányi (Budapest, Hungary) Zeno Spârchez (Cluj-Napoca, Romania)
Sorin Crişan (Cluj-Napoca, Romania) Anamaria Iagnocco (Rome, Italy) Ioan Sporea (Timişoara, Romania)
Adrian Costache (Bucureşti, Romania) Adnan Kabaalioglu (Antalya, Turkey) Florin Stamatian (Cluj-Napoca)
Jarosław Czubak (Otwock, Poland) Daniel Lichtenstein (Paris, France) Dan Stănescu (Bucureşti, Romania)
Christoph Dietrich (Frankfurt am Main, Germany) Carmen Mihaela Mihu (Cluj-Napoca, Romania) Iwona Sudoł-Szopińska (Warsaw, Poland)
Dan Dumitraşcu (Cluj-Napoca) Dan Mihu (Cluj-Napoca, Romania) Kazmierz Szopinski (Warsaw, Poland)
Viorela Enăchescu (Craiova, Romania) Daniel Muresan (Cluj-Napoca, Romania) Adrian Şanta (Sibiu, Romania)
Otilia Fufezan (Cluj-Napoca, Romania) Luca Neri (Milan, Italy) Roxana Sirli (Timişoara, Romania)
Odd Helge Gilja (Bergen, Norway) Monica Platon Lupsor (Cluj-Napoca, Romania)
Tehnical staff Instruction for authors: Subscription information:
Iulia China Full instructions are available online at http://www.medultrason.ro/ Medical Ultrasonography is published quarterly.
authors-guidelines/ and at the end pages of the journal. ISSN (print) 1844–4172; ISSN (online) 2066–8643
The annual subscription:
For Romania – 150.00 lei for individuals For other countries (including postage fees) – 60 euro for individuals
– 350.00 for institutions – 120 euro for institutions
Bank account:
– in lei: SRUMB – Filiala Cluj-Napoca (CIF 10232679), BRD, sucursala Cluj-Napoca, Cont: RO62 BRDE 130S V959 5116 1300
– in euro: SRUMB – Filiala Cluj-Napoca (CIF 10232679), BRD, sucursala Cluj-Napoca, Cont: RO54 BRDE 130S V959 5132 1300
How to subscribe: Subscribers may pay in the bank account already mentioned. Orders should be sent online at www.medultrason.ro
“Subscription” or by email to the Editorial Office (dfodor@ymail.com)
Publishing House: “Iuliu Haţieganu” Medical Publishing House, Cluj-Napoca, Contact person: Ioana Robu,
8, Victor Babeş str, Cluj-Napoca, Tel.: + 40-264-592629 / 590832, Fax: + 40-264-598820, E-mail: irobu@umfcluj.ro
All right reserved to the Medical Publishing House of the “Iuliu Hatieganu” University of Medicine and Pharmacy, Cluj-Napoca;
no parts of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise without the permission of the publisher.
 Contents
(continued)

Case report
Primary splenic leiomyosarcoma – case report and literature review
E.S. Ioanițescu, M. Grasu, L. Toma ........................................................................................................................................... 114
Vomiting-induced costal cartilage fracture: a case report
E. Drakonaki, I. Karageorgiou, S. Kokkinakis, N. Maliotis, R. Spyridaki, E.K. Symvoulakis ................................................... 117
Letters to the Editor
Ultrasonographic diagnosis and guided treatment of erector spinae aponeurosis enthesopathy
I-C. Liu, M. Boudier-Revéret, M.C. Chang, M.-Y. Hsiao ...........................................................................................................120
Ultrasound guidance may be beneficial for localizing the atrophied muscles in electromyography
W.-C. Huang, Y-H. Chiu, K.-C. Wei ............................................................................................................................................121
Imaging findings of a tall cell variant of papillary breast carcinoma
M. Pang, M. Yuan, M. Yu ............................................................................................................................................................122
Imaging findings of a spindle epithelial tumour with thyroid thymoid differentiation
M. Pang, M. Yuan, M. Yu ............................................................................................................................................................124
Ultrasound and clinical findings of hyalinizing trabecular tumor of the thyroid
Y.J. Xing, J. Zhang, B.S. Yi .........................................................................................................................................................125
Comment on “Usefulness of lung ultrasound in the early identification of severe COVID-19:
results from a prospective study”
R. Mungmunpuntipantip, V. Wiwanitkit ..................................................................................................................................... 126
Comment on “Usefulness of lung ultrasound in the early identification of severe COVID-19:
results from a prospective study”
D. Di Costanzo, M. Mazza, A. Esquinas ................................................................................................................................... 127
Authors’ response
Hernández-Píriz, Y. Tung-Chen, D. Jiménez-Virumbrales, I. Ayala-Larrañaga, R. Barba-Martín, J. Canora-Lebrato,
A. Zapatero-Gaviria, G.G. De Casasola-Sánchez .....................................................................................................................128
In memoriam
Dr Mircea Leonid Stamate .........................................................................................................................................................129
Editorial
Lung ultrasound for ever
Christoph F Dietrich
Department Allgemeine Innere Medizin (DAIM), Kliniken Hirslanden, Beau Site, Salem und Permanence, Bern,
Switzerland
The World Federation for Ultrasound in Medicine
and Biology (WFUMB) is dedicated to the advancement
of ultrasound by encouraging research, promoting inter-
national cooperation, disseminating scientific informa-
tion and improving communication and understanding
in the world community using ultrasound in medicine
and biology. One mission of WFUMB is to bring sus-
tainable ultrasound programs to the underserved areas of
the world to improve global healthcare through collabo-
ration, communication and education (www.wfumb.org).
Medical Ultrasonography (formerly Revista Româna de
Ultrasonografie from 1999 to 2008) is the official publi-
cation of the Romanian Society for Ultrasonography in
Medicine and Biology (SRUMB), published in English,
quarterly. SRUMB and Med Ultrason are active members
in WFUMB and the European Federation of Societies for
Ultrasound in Medicine and Biology (EFSUMB) [1].
In the current WFUMB paper series, three consecu-
tive papers are published in Med Ultrason describing the
examination technique, applications and practical use of
chest, diaphragmatic, pleura, mediastinal and lung ultra-
sound in children with congenital and acquired diseases.
The advantages of ultrasound in pediatric patients are ob-
vious: its high spatial and temporal resolution, real-time
imaging, and lack of ionizing radiation and bedside avail-
ability at the point of care [2-4]. In addition, new ultra-
sound technologies have been introduced into chest and
lung ultrasound applications including higher resolution
transducers and harmonic imaging allowing direct visu-
alization of structures being less dependent on artifacts
[5-8]. The use and controversies of lung artefacts have
Received Accepted
Med Ultrason
2022, Vol. 24, No 1, 5-6
Corresponding author: Prof. Dr. med. Christoph F. Dietrich, MBA
Department of Internal Medicine (DAIM)
Kliniken Hirslanden Bern, Beau Site,
Salem and Permanence
Schänzlihalde 11, 3031 Bern, Switzerland
E-mail: c.f.dietrich@googlemail.com
Med Ultrason 2022, Vol. 24, no. 1, 5-6
DOI: 10.11152/mu-3616
been described in Med Ultrason [5,7]. Contrast-enhanced
ultrasound (CEUS) ([11] and also to a lesser degree elas-
tography [12] have expanded the roles of LUS. The chal-
lenges of CEUS in pediatric patients have been recently
reviewed [13-15].
The lack of superficial adipose and bone tissue pro-
vides favorable acoustic windows in children compared
to the more often more difficult situation in adults due to
the presence of a bony thorax makes ultrasound the first
line of investigation for evaluation of pleural and chest
wall abnormalities in pediatric patients. In a meta-anal-
ysis consisting of 1510 children, chest ultrasound shows
significantly better sensitivity and similar specificity in
detecting pneumonia in children compared to chest radi-
ography [16]. LUS has been proposed as a reasonable al-
ternative first-line investigation for diagnosing suspected
community-acquired pneumonia [17].
The first article covers the technical requirements, ex-
amination technique, normal sonographic appearance of
the pleural space, identifying and determining the type
and volume of pleural effusion, pleuritis and diffuse pleu-
ral thickening, solid pleural lesions including benign and
malignant pleural tumors and fibrogenic changes of the
pleura (fibrothorax) [2]. The most important pathologies
of the pleural cavity in pediatric patients are pleural effu-
sion and pneumothorax [2].
In the second paper the use of ultrasound in the lung
in pediatric patients is described including the intersti-
tial syndrome, bacterial pneumonia and viral infections,
COViD findings, atelectasis, lung consolidation, bron-
chiolitis and congenital diseases of the respiratory sys-
tem including congenital pulmonary airway malforma-
tion (CPAM) and sequester [3]. Severe acute respiratory
syndrome (SARS) coronavirus-2 disease (COViD) has
been a challenge since 2019. Lung ultrasound allows the
identification of typical sonographic signs in the course
of COVID-19 infections including the triad of ultrasound
features of the pleura and pleural space represented by
irregularities, findings of interstitial pneumonia repre-
6 Christoph F Dietrich
sented by subpleural consolidation and B-line artefacts.
In addition, contrast enhanced ultrasound depicts typical
perfusion defects in COViD-patients compared to other
entities [9,11]. Extrapulmonary manifestations can be
also detected and characterized using ultrasound [18].
In the third article, the use of ultrasound for chest
wall, mediastinum, diaphragmatic diseases, trachea, in-
terventions and artifacts in pediatric patients are summa-
rized [4].
In conclusion, the change in attitude and growing
awareness of the diagnostic possibilities has led to lung
ultrasound being accepted as a valuable point of care
method and is supported by international societies in-
cluding EFSUMB and WFUMB [19-22].
References
1. Piscaglia F, Stefanini F, Cantisani V, et al. Benefits, Open
questions and Challenges of the use of Ultrasound in the
COVID-19 pandemic era. The views of a panel of world-
wide international experts. Ultraschall Med 2020;41:228-
236.
2. Jaworska J, Buda N, Ciuca IM, et al. Ultrasound of the
pleura in children, WFUMB review paper. Med Ultrason
2021;23:339-347.
3. Dietrich CF, Buda N, Ciuca IM, et al. Lung ultrasound in
children, WFUMB review paper (part 2). Med Ultrason
2021;23:443-452.
4. Fang C, Jaworska J, Buda N, et al. Ultrasound of the chest
and mediastinum in children, interventions and artefacts.
WFUMB review paper (part 3). Med Ultrason 2022;24:
65-76.
5. Mathis G, Horn R, Morf S, et al. WFUMB position paper
on reverberation artefacts in lung ultrasound: B-lines or
comet-tails? Med Ultrason 2021;23:70-73.
6. Yue Lee FC, Jenssen C, Dietrich CF. A common misunder-
standing in lung ultrasound: the comet tail artefact. Med
Ultrason 2018;20:379-384.
7. Dietrich CF, Mathis G, Blaivas M, et al. Lung artefacts and
their use. Med Ultrason 2016;18:488-499.
8. Dietrich CF, Mathis G, Blaivas M, et al. Lung B-line arte-
facts and their use. J Thorac Dis 2016;8:1356-1365.
9. Safai Zadeh E, Beutel B, Dietrich CF, et al. Perfusion Pat-
terns of Peripheral Pulmonary Lesions in COVID-19 Pa-
tients Using Contrast-Enhanced Ultrasound (CEUS): A
Case Series. J Ultrasound Med 2021;40:2403-2411.
Lung ultrasound for ever
10. Safai Zadeh E, Görg C, Dietrich CF, Görlach J, Alhyari
A, Trenker C. Contrast-enhanced ultrasound for evalua-
tion of pleural effusion: a pictorial essay. J Ultrasound Med
2022;41:485-503.
11. Safai Zadeh E, Westhoff CC, Keber CU, et al. Perfusion
Patterns of Peripheral Organizing Pneumonia (POP) Using
Contrast-Enhanced Ultrasound (CEUS) and Their Correla-
tion with Immunohistochemically Detected Vascularization
Patterns. Diagnostics (Basel) 2021;11:1601.
12. Dietrich CF, Ferraioli G, Sirli R, et al. General advice in
ultrasound based elastography of pediatric patients. Med
Ultrason 2019;21:315-326.
13. Dietrich CF, Averkiou M, Nielsen MB, et al. How to per-
form Contrast-Enhanced Ultrasound (CEUS). Ultrasound
Int Open 2018;4:E2-E15.
14. Sidhu PS, Cantisani V, Deganello A, et al. Role of Con-
trast-Enhanced Ultrasound (CEUS) in Paediatric Prac-
tice: An EFSUMB Position Statement. Ultraschall Med
2017;38:33-43.
15. Dietrich CF, Augustiniene R, Batko T, et al. European Fed-
eration of Societies for Ultrasound in Medicine and Biology
(EFSUMB): An Update on the Pediatric CEUS Registry on
Behalf of the “EFSUMB Pediatric CEUS Registry Working
Group”. Ultraschall Med 2021;42:270-277.
16. Balk DS, Lee C, Schafer J, et al. Lung ultrasound compared
to chest X-ray for diagnosis of pediatric pneumonia: A me-
ta-analysis. Pediatr Pulmonol 2018;53:1130-1139.
17. Stadler JAM, Andronikou S, Zar HJ. Lung ultrasound for
the diagnosis of community-acquired pneumonia in chil-
dren. Pediatr Radiol 2017;47:1412-1419.
18. Dehmani S, Penkalla N, Jung EM, et al. Scoping Review:
Sonographic evidence of intraabdominal manifestations of
COVID-19. Med Ultrason 2022. doi:10.11152/mu-3538
19. Dietrich CF, Mathis G, Cui XW, Ignee A, Hocke M, Hirche
TO. Ultrasound of the pleurae and lungs. Ultrasound Med
Biol 2015;41:351-365.
20. Dietrich CF, Goudie A, Chiorean L, et al. Point of Care Ul-
trasound: A WFUMB Position Paper. Ultrasound Med Biol
2017;43:49-58.
21. Soldati G, Smargiassi A, Inchingolo R, et al. Proposal for
International Standardization of the Use of Lung Ultra-
sound for Patients With COVID-19: A Simple, Quantitative,
Reproducible Method. J Ultrasound Med 2020;39:1413-
1419.
22. Volpicelli G, Elbarbary M, Blaivas M, et al. International
evidence-based recommendations for point-of-care lung ul-
trasound. Intensive Care Med 2012;38:577-591.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 7-13
DOI: 10.11152/mu-3238

Feasibility of pneumoperitoneum diagnosis using point-of-care

ultrasound: a pilot study using a fresh cadaver model
Meghan Kelly Herbst1, Elizabeth Carter2, Shirley Wu3, Christoph F Dietrich4, Beatrice
Hoffmann5

1Department of Emergency Medicine, University of Connecticut School of Medicine, Farmington CT, USA, 2Depart-
ment of Emergency Medicine, Inova Alexandria Hospital, Alexandria VA, USA, 3Department of Emergency Medicine,
Rhode Island Hospital, Brown University, Providence, RI, USA, 4Department of Medicine, Klinik Hirslanden, Bern,
Switzerland, 5Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School,
Boston MA, USA

Abstract
Aims: To assess the accuracy of point-of-care ultrasound (PoCUS) in the hands of two trained and blinded emergency
physicians (EPs) in detecting very small amounts of free intraperitoneal air injected intra-abdominally, using a fresh human
cadaver model. Material and methods: Fifteen cadavers were injected on 3 occasions with predefined quantities of free in-
traperitoneal air ranging from 0-10 mL. Seven cadavers were injected in the mid-epigastrium (ME), while 8 were injected in
the left lower quadrant (LLQ). Each cadaver was scanned after each of the 3 injections by 2 trained and blinded EPs, resulting
in 45 scans per sonographer. Scans were performed using previously validated and standardized techniques. All scans were
recorded, time-stamped and labeled. For each scan the sonographers indicated “yes” or “no” to whether pneumoperitoneum
was detected. A chi square analysis was performed to determine the sensitivity and specificity of PoCUS utilized by each so-
nographer of pneumoperitoneum based on the location and volume of air injected. Results: Free air (0.25-10 mL) injected into
the ME was successfully diagnosed in 36/42 instances (86% sensitivity), but only detected in 10/36 instances when injected
into the LLQ (28% sensitivity). Both EPs detected all air injections of ≥2 mL into the ME. Conclusion: Detection of free air
originating from the midepigastric region may become a future PoCUS indication for adequately trained EPs.
Keywords: point-of-care ultrasound; emergency medicine; pneumoperitoneum; cadaver

Introduction in the differential diagnosis for the acute abdomen [2-

The acute abdomen, a sudden severe abdominal pain
of unclear etiology that is less than 24 hours in duration,
accounts for more than 7 million United States Emer-
gency Department (ED) visits annually and up to 40%
of emergency surgical admissions [1]. Gastrointestinal
perforation with pneumoperitoneum is often considered
Received 15.05.2021  Accepted 21.07.2021
Med Ultrason
2022, Vol. 24, No 1, 7-13
Corresponding author: Beatrice Hoffmann
Beth Israel Deaconess Medical Center Boston,
Department of Emergency Medicine,
One Deaconess Road, Rosenberg 2,
Boston 02215, USA
E-mail: bhoffma2@bidmc.harvard.edu
4]. Morbidity and mortality from secondary peritonitis
remains high (30-50%) despite advances in medical and
surgical therapies [5]. Rapid diagnosis and management
in the ED is therefore essential to reduce complications
associated with gastrointestinal perforation [5-7].
While the initial conventional imaging approach in
the ED to detect pneumoperitoneum is upright poster-
oanterior chest or left lateral decubitus radiography, it is
known to be insufficiently sensitive for small to moderate
amounts of free air and may miss the diagnosis and delay
management [8-14]. X-rays also expose patients to small
doses of ionizing radiation and should be avoided when
possible in children and women of reproductive age [15].
Computed tomography (CT) is highly accurate for even
small amounts of free intraabdominal air and is consid-
ered gold standard imaging for this diagnosis, however, it
8 Meghan Kelly Herbst et al Feasibility of pneumoperitoneum diagnosis using point-of-care ultrasound: a pilot study
is associated with even higher exposure to ionizing radia-
tion and higher health care costs [14,16-22]. Critically ill
and hemodynamically unstable patients may especially
benefit from a portable imaging modality, such as point-
of-care ultrasound (PoCUS) performed by trained emer-
gency physicians (EPs), to expedite diagnosis and man-
agement prior to definitive imaging.
Prior investigations by non-EP imaging experts have
shown that ultrasound (US) can detect small quantities
of intraperitoneal air. As early as 1982, Seitz and Reising
reported in a proof of concept study that the accuracy of
US is compatible with conventional chest x-ray imaging,
the gold standard imaging test in 1982 for free air. These
highly trained experts detected as little as 1 mL of free in-
traperitoneal air with 100% accuracy [23]. In their large-
scale follow up study of about 4,000 consecutive patients
presenting to a single ED with non-traumatic acute ab-
dominal pain, the same authors demonstrated 90% sensi-
tivity and 100% specificity of US for pneumoperitoneum
[23]. Smaller subsequent studies performed by highly
trained physicians further supported these initial find-
ings, including in blunt trauma and non-trauma patients.
All showed high accuracy and superior sensitivity when
compared to plain radiography [24-28]. However, given
the trend of increasing use of PoCUS in emergency medi-
cine education, there is little knowledge about whether
EPs who train in PoCUS can reliably identify pneumop-
eritoneum [29,30]. It would additionally be of interest to
learn if a cadaver model with induced pneumoperitone-
um can serve as a reliable teaching model for emergency
physicians learning to diagnose pneumoperitoneum with
PoCUS.
If EPs trained in PoCUS can accurately detect free
intraperitoneal air, especially small amounts of free in-
traperitoneal air, management of critically ill patients
with perforated viscous and free air could potentially be
expedited.
We conducted a pilot study to determine if two Po-
CUS trained EPs could reliably detect small amounts of
intraabdominal free air. We chose a fresh human cadav-
er model for feasibility reasons to assure a prospective
blinded study design and to reliably determine the exact
amount of free air present in the intraabdominal cavity.
Material and methods
Preparation of cadavers
Fifteen fresh human cadavers donated to the Anatomy
Institute were utilized for this prospective randomized
study, which took place at the University of Maryland,
Baltimore, Anatomical Services Division cadaver lab.
The Institutional Review Board reviewed this research
study and deemed this study exempt as per current fed-
eral guidelines.
In preparation of scanning, all cadavers were posi-
tioned in reverse Trendelenburg at 30 degrees for ten
minutes and pre-scanned by two trained separate EPs, to
rule out the presence of any artifacts or irregularities at
the peritoneal area including sonographic artifacts con-
sistent with pneumoperitoneum. All participating emer-
gency physicians had prior extensive US experience in
PoCUS including pneumoperitoneum diagnosis.
The two coordinators injected 15 fresh human ca-
davers with various predefined quantities of free intra-
peritoneal air, ranging from 0-10 mL. Each cadaver was
injected 3 times throughout the study and scanned three
times. The injection/scanning order was based upon a
pre-designed randomized order and matrix (Table I, fig
1). Figure 2 illustrates a positive and negative US finding
for free intraperitoneal air. The matrix specified the quan-
tity of air to be injected, the location of injection and the
order of the cadavers to be scanned. A cadaver was rand-
omized to either the mid-epigastric (ME) injection (1 cm
cephalad to the umbilicus) or left lower quadrant (LLQ)
abdominal injection, which was performed exactly equi-
distant between umbilicus and left anterior superior iliac
spine (ASIS). Once a cadaver was randomized to a spe-
cific injection site, all three injections for this cadaver
were administered at the same anatomical location and
resulted in cumulative amounts of free air.
Of the 15 cadavers, eight were injected x3 in the
midepigastric area, seven x3 into the left lower quadrant
as described above. This created 45 different scanning
scenarios: 24 for cadavers with ME free air and 21 for
LLQ free air (fig 3). As each cadaver was scanned by
both sonographer A and B, this resulted in a total of 48
encounters for ME cadavers and 42 observations in LLQ
Fig 1. Workflow of air injection and randomized scanning of
cadavers.
 Med Ultrason 2022; 24(1): 7-13 9
Table I. Matrix of air injected for each cadaver per round of observation and area injected.
round 1 cadaver # total mL area round 2 cadaver # total mL area round 3 cadaver # total mL area
1 1 0 ME 16 1 0.5 ME 31 12 0.75 LLQ
2 13 0.25 LLQ 17 5 1 ME 32 11 0.25 LLQ
3 10 1 LLQ 18 13 4 LLQ 33 8 4 ME
4 4 0 ME 19 5 4 ME 34 8 10 ME
5 5 0.75 ME 20 14 0.75 LLQ 35 12 2 LLQ
6 14 0 LLQ 21 4 0.75 ME 36 11 1 LLQ
7 7 0.25 ME 22 1 1 ME 37 9 2 LLQ
8 15 0 LLQ 23 13 7 LLQ 38 6 1 ME
9 2 2 ME 24 15 1 LLQ 39 3 0.25 ME
10 14 0.5 LLQ 25 2 10 ME 40 3 0.75 ME
11 2 7 ME 26 10 4 LLQ 41 9 4 LLQ
12 15 0.5 LLQ 27 7 7 ME 42 6 4 ME
13 7 0.5 ME 28 11 0 LLQ 43 6 10 ME
14 10 2 LLQ 29 8 0 ME 44 3 2 ME
15 4 0.25 ME 30 12 0.25 LLQ 45 9 10 LLQ
ME = midepigastrium, LLQ = Left lower quadrant

injected cadavers. Several of the initial injections were

sham injections with 0 mL of free air.
For each of the three rounds or observations, injec-
tions of air in increments of 0, 0.25, 0.5, 0.75, 1, 2, 4, 7 or
10 mL were performed. The assigned volume of air with
was combined with normal saline to total a volume of 12
mL. For example, if the assigned volume of air was 0.5
mL air, 11.5 mL normal saline was added to the syringe
so that the total volume of air and normal saline to be
injected was 12 mL. For the sites assigned zero mL of
air, a total of 12 mL of normal saline was injected. All in-

Fig 3. Illustration of injection sites and transducer positioning

for midepigastric (ME) and left lower quadrant (LLQ) injec-
tions of free air.

jections were performed under ultrasound guidance. The

Fig 2. Normal peritoneal line (white thick arrow) over anterior
liver (a). Typical ultrasonographic signs of pneumoperitoneum
showing normal peritoneal line (white thick arrow) and the re-
verberation artifact caused by free air (thin arrows) below a free
air collection (b).
skin was punctured with a 25-gauge needle in both the
ME and LLQ regions of all cadavers regardless of the lo-
cation of injection for blinding purposes. As noted above,
injections were repeated x2 for each of the 15 cadavers to
introduce 45 (15x3) combinations of scanning scenarios
for each sonographer (fig 4). Cadavers were stationed
in several rooms, covered with sheets and towels, and
wheeled to different locations randomly to preserve so-
nographer blinding. All cadaver anatomy was concealed
with the exception of the abdomen.
10 Meghan Kelly Herbst et al Feasibility of pneumoperitoneum diagnosis using point-of-care ultrasound: a pilot study

Fig 4. Progression of injection volume per cadaver. Each colored line represents one cadaver, and each point represents one encoun-
ter. The vertical axis represents the cumulative number of mLs injected, while the horizontal axis represents each round of injections.

Scanning in 26 of 36 instances when injected into LLQ (28% sen-

The two blinded expert sonographers (A and B) sitivity in 18 separate injections, fig 5-7). Notably, the
scanned each cadaver independently using a SonoSite sonographers correctly diagnosed and documented all 10
Turbo (Bothell, WA) high a frequency linear probe (L38) separate air injections of ≥2 mL into the ME (6 cadav-
positioned sagittally with the indicator directed cephalad ers) but were not successfull to diagnose comparable air
in three locations: the right upper quadrant (RUQ), the injections of ≥2 mL into the LLQ in 13 of 16 trials (4 ca-
ME and the left upper quadrant (LUQ). Scans were per-
formed via previously validated and standardized tech-
niques to detect typical signs of pneumoperitoneum. The
ME and LLQ were chosen as injection sites as these are
common sites of perforated viscous that can produce air
visible throughout the abdomen and detectable at per-
foration site, right upper quadrant and mid-abdomen
[23,31-33]. All scans were recorded, time-stamped and
labeled. For each scan, the sonographers indicated “yes”
or “no” to whether pneumoperitoneum was detected.
Statistical analysis
A chi square analysis was performed to determine
the sensitivity and specificity of PoCUS utilized by each Fig 5. Correctly diagnosed midepigastric (ME) and left lower
quadrant (LLQ) injected free air.
sonographer to detect accuracy and various quantities
of pneumoperitoneum based on location of air injected.
An unweighted Cohen’s Kappa value was calculated to
measure interrater agreement.

Results

The sonographers were considerably more effec-

tive at correctly diagnosing and documenting air that
had been injected (0.25-10 mL) into the ME region than
in instances where air had been injected into the LLQ
region (p<0.001 by Chi-square analysis). Air (0.25-10
mL) injected into ME was successfully diagnosed and Fig 6. Success rates of first, second and third observation for
documented with 86% sensitivity in 36 of 42 encounters. each mid-epigatsric (ME) and left lower quadrant (LLQ) loca-
Pneumoperitoneum was not detected and/or documented tion free air injections after multiple scans on a single cadaver.

Fig 7. Overview of success rates of the 2 sonographers evaluat-
ing the mid-epigastrium (ME) or left lower quadrant (LLQ).
Cadaver Number (#) shows injection side with injection vol-
ume in mL. Sonographer A and B are shown as a circle if mak-
ing an incorrect diagnosis, and if correct as full dot.
davers). The disparity in success rates for diagnosing and
documenting air injections into the two locations is fur-
ther illustrated by the number of instances in which the
sonographers had correctly diagnosed and documented
air at a lower volume of injection, only to fail to diagnose
or document a higher volume of air injected into the same
location in the same cadaver (once with injections in ME,
versus eight separate instances (four for each sonogra-
pher) with injections into the LLQ).
There was no obvious benefit of performing multiple
scans on a single cadaver, as success rates when air had
been injected were comparable after the 1st, 2nd and 3rd
injections in both the ME (90, 88 and 81% respectively)
and the LLQ (50, 14 and 29% respectively). Sonogra-
pher B more frequently provided a correct diagnosis and
documentation of air injected into the ME (95 vs. 76%)
or the LLQ (39 vs. 17%), but also more frequently gener-
ated false positive diagnoses (3 vs. 1).
Interobserver agreement between Sonographer A and
B for all 45 exams was moderate (Cohen’s unweighted
Kappa = 0.486, 95% CI [0.260, 0.711]) with agreement
in 33 of the 45 observations (73%). Agreement improved
and was good for the 18 exams with 2 mL or more of
free air injected. Numbers of observed agreements were
15/18 (83.33%) of the observations and Kappa = 0.649
(standard error of kappa = 0.173; 95% CI [0.310, 0.989].
When looking only at cadavers where air was injected
into the ME, the sensitivity and specificity of PoCUS
performed by sonographer A for pneumoperitoneum
were 76.1% (95% CI [52.4, 90.9]) and 100% (95% CI
[31, 100], respectively; Sonographer B PoCUS sensitiv-
ity and specificity were 90.4% (95% CI [68.1, 98.3] and
33.3% (95% CI [1.8, 87.4], respectively.
Med Ultrason 2022; 24(1): 7-13 11
Discussion
We found in our pilot study that two PoCUS trained
EPs detected small amounts of free air ≥ 2 mL with 100%
accuracy when injected into the midepigastrium in a hu-
man cadaver model. While prior studies and case reports
suggest that US can be a critical tool in the identifica-
tion of free air [23] and has been shown to identify small
amounts of free air by specialized imaging experts [24],
this is the first study showing that EPs trained in PoCUS
can reliably perform PoCUS to detect very small pneu-
moperitoneum on a cadaver model. Importantly, this may
translate into detecting small pneumoperitoneum on pa-
tients with undifferentiated abdominal pain or hypoten-
sion, which can help inform next steps and management
in the acute care setting.
If validated in a clinical setting, PoCUS would be a
safe, quick and potentially sensitive option for detecting
pneumoperitoneum, when performed by trained physi-
cians. In very young patients and populations at highest
risk from radiation exposure, PoCUS may even obviate
the need for CT imaging.
The study also shows a potential training module for
EPs that would allow fresh frozen cadavers for pneumo-
peritoneum training models.
Our study has several limitations. Fresh cadavers are
different than living human beings, which impact the
external validity of this study. In live humans, there are
many ongoing dynamic processes, that likely promote
movement of pneumoperitoneum to the least depend-
ent region possible. For example, bowel peristalsis, dia-
phragmatic movements, movements of the abdominal
and back musculature and ambulation all may contribute
to free air movement to the RUQ or least dependent re-
gion of the peritoneal space.
In our cadaver model, factors that may have pre-
vented air from moving to the least dependent region of
the peritoneal space could include not allowing the air
enough time to travel to the least dependent regions after
positioning the cadavers in reverse Trendelenburg posi-
tions, lack of bowel peristalsis and abdominal muscle
tone. These barriers to air migration may have especially
skewed results of pneumoperitoneum detected from the
LLQ injection cadavers.
While not a common occurrence, bowel can interpose
itself between the peritoneal line and the liver, creating a
sonographic appearance similar to that of free air. In this
situation, the presence of peristalsis may help differenti-
ate free air from air within the bowel in a living patient.
Because peristalsis does not occur in our cadaver model,
air within the bowel may appear to be just beneath the
peritoneal line, contributing to the false positive findings.
12 Meghan Kelly Herbst et al Feasibility of pneumoperitoneum diagnosis using point-of-care ultrasound: a pilot study
Despite the creation of 45 blinded encounters, this
was a small study limited by the number of cadavers
available and needs to be repeated on a larger scale, ide-
ally on live patients.
In conclusion, this study demonstrated that two
trained emergency physicians detected small amounts
of free air ≥2 mL reliably when injected into the ME in
this human cadaver model. PoCUS appeared feasible for
detecting small amounts of pneumoperitoneum in this
model with this approach, and, if clinically validated,
could expedite the diagnosis and management of patients
suspected of having pneumoperitoneum in the acute care
setting.
A future prospective study looking at PoCUS detec-
tion of pneumoperitoneum among live human beings
would be needed to improve external validity and pro-
vide clinical value.
Acknowledgements: We would like to thank
Dr. Richard Levitan and Mr. Ronn Wade from the Uni-
versity of Maryland, Baltimore Anatomical Services Di-
vision for permitting us to utilize their human cadavers
and cadaver lab facilities. We would also like to thank
Dr. Ilene Staff for her assistance with the statistical analy-
sis of the data collected. Funding for this study was from
the Hartford Hospital New Investigator Grant.
Rerefences
1. Langell JT, Mulvihill SJ. Gastrointestinal perforation and
the acute abdomen. Med Clin North Am 2008;92:599-625.
2. McCaig LF, Burt CW. National Hospital Ambulatory Medi-
cal Care Survey: 2002 emergency department summary.
Adv Data 2004;(340):1-34.
3. Court-Brown CM, McQueen MM, Patterson-Brown S,
Nixon SJ. Emergency surgical care in Scotland. Surgeon
2007;5:72-75.
4. Schietroma M, Cappelli S, Carlei F, Pescosolido A, Lygida-
kis NJ, Amicucci G. “Acute abdomen”: early laparoscopy
or active laparotomic-laparoscopic observation? Hepato-
gastroenterology 2007;54:1137-1141.
5. Ordonez CA, Puyana JC. Management of peritonitis in the
critically ill patient. Surg Clin North Am 2006;86:1323-
1349.
6. Wittmann DH, Schein M, Condon RE. Management of sec-
ondary peritonitis. Ann Surg 1996;224:10-18.
7. Feldman M, Friedman LS, Sleisenger MH. (Eds.). Sleisen-
ger and Fordtran’s gastrointestinal and liver disease: Patho-
physiology, diagnosis, management. 7th ed. Philadelphia:
WB Saunders, 2004.
8. Stoker J, van Randen A, Lameris W, Boermeester MA.
Imaging patients with acute abdominal pain. Radiology
2009;253:31-46.
9. MacKersie AB, Lane MJ, Gerhardt RT, et al. Nontrau-
matic acute abdominal pain: unenhanced helical CT com-
pared with three-view acute abdominal series. Radiology
2005;237:114-122.
10. Butler J, Martin B. Towards evidence based emergency
medicine: best BETs from the Manchester Royal Infirmary.
Detection of pneumoperitoneum on erect chest radiograph.
Emerg Med J 2002;19:46-47.
11. Gupta H, Dupuy DE. Advances in imaging of the acute ab-
domen. Surg Clin North Am 1997;77:1245-1263.
12. Billittier AJ, Abrams BJ, Brunetto A. Radiographic imag-
ing modalities for the patient in the emergency department
with abdominal complaints. Emerg Med Clin North Am
1996;14:789-850.
13. Ahn SH, Mayo-Smith WW, Murphy BL, Reinert SE, Cro-
nan JJ. Acute nontraumatic abdominal pain in adult pa-
tients: abdominal radiography compared with CT evalua-
tion. Radiology 2002;225:159-164.
14. Stapakis JC, Thickman D. Diagnosis of pneumoperitone-
um: abdominal CT vs. upright chest film. J Comput Assist
Tomogr 1992;16:713-716.
15. Greene CS. Indications for plain abdominal radiography in the
emergency department. Ann Emerg Med 1986;15:257-260.
16. National Research Council. Health risks from exposure to
low levels of ionizing radiation: BEIR VII phase 2. Nation-
al Academies Press 2006. Available from: http://www.nap.
edu/catalog/11340.html
17. Preston DL, Ron E, Tokuoka S, et al. Solid cancer inci-
dence in atomic bomb survivors: 1958-1998. Radiat Res
2007;168:1-64.
18. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure
from CT scans in childhood and subsequent risk of leukae-
mia and brain tumours: a retrospective cohort study. Lancet
2012;380:499-505.
19. Mathews JD, Forsythe AV, Brady Z, et al. Cancer risk in
680,000 people exposed to computed tomography scans in
childhood or adolescence: data linkage study of 11 million
Australians. BMJ 2013;346:f2360.
20. Lumbreras B, Donat L, Hernandez-Aguado I. Incidental
findings in imaging diagnostic tests: a systematic review.
Br J Radiol 2010;83:276-289.
21. Thompson RJ, Wojcik SM, Grant WD, Ko PY. Inciden-
tal Findings on CT Scans in the Emergency Department.
Emerg Med Int 2011;2011:624847.
22. Medicare Payment Advisory Commission MPA. Report to
the Congress: improving incentives in Medicare. Available
from: http://www.medpac.gov/docs/default-source/reports/
Jun09_EntireReport.pdf Accessed 2016.
23. Seitz K, Reising KD. Ultrasound detection of free air in the
abdominal cavity. Ultraschall Med 1982;3:4-6.
24. Chang-Chien CS, Lin HH, Yen CL, Lee CM, Lin SM. So-
nographic demonstration of free air in perforated peptic ul-
cers: comparison of sonography with radiography. J Clin
Ultrasound 1989;17:95-100.
25. Chen SC, Wang HP, Chen WJ, et al. Selective use of ultra-
sonography for the detection of pneumoperitoneum. Acad
Emerg Med 2002;9:643-645.

26. Muradali D, Wilson S, Burns PN, Shapiro H, Hope-Simp-
son D. A specific sign of pneumoperitoneum on sonog-
raphy: enhancement of the peritoneal stripe. AJR Am J
Roentgenol 1999;173:1257-1262.
27. Butcher CH, Dooley RW, Levitov AB. Detection of sub-
cutaneous and intramuscular air with sonography: a sensi-
tive and specific modality. J Ultrasound Med 2011;30:791-
795.
28. Moriwaki Y, Sugiyama M, Toyoda H, et al. Ultra-
sonography for the diagnosis of intraperitoneal free
air in chest-abdominal-pelvic blunt trauma and criti-
cal acute abdominal pain. Arch Surg 2009;144:137-
141.
29. Taylor MA, Merritt CH, Riddle PJ Jr, DeGennaro CJ, Bar-
ron KR. Diagnosis at gut point: rapid identification of pneu-
Med Ultrason 2022; 24(1): 7-13 13
moperitoneum via point-of-care ultrasound. Ultrasound J
2020;12:52.
30. Nazerian P, Tozzetti C, Vanni S, et al. Accuracy of abdomi-
nal ultrasound for the diagnosis of pneumoperitoneum in
patients with acute abdominal pain: a pilot study. Crit Ul-
trasound J 2015;7:15.
31. Hoffmann B, Nürnberg D, Westergaard MC. Focus on ab-
normal air: diagnostic ultrasonography for the acute abdo-
men. Eur J Emerg Med 2012;19:284-291.
32. Lee DH, Lim JH, Ko YT, Yoon Y. Sonographic detection
of pneumoperitoneum in patients with acute abdomen. AJR
Am J Roentgenol 1990;154:107-109.
33. Karahan OI, Kurt A, Baykara M, Coskun A. Detectability
of intraperitoneal free air by ultrasonography. Tani Girisim
Radyol 2003;9:60-62.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 14-18
DOI: 10.11152/mu-3163

Ultrasonographic characteristics and outcome of Type III

umbilical-portal-systemic venous shunt
Linlin Zhu, Haifang Wu, Xiang Cong, Zhe Ma, Guowei Tao

Department of Ultrasound, Qilu Hospital of Shandong University, Jinan, Shandong Province, China

Abstract
Aims: According to a novel in-utero classification termed “umbilical-portal-systemic venous shunt (UPSVS)” recently
proposed for an abnormal umbilical, portal and ductal venous system, the portal-systemic shunt belongs to type III UPSVS.
This study was designed to examine the ultrasonographic characteristics and outcome of type III UPSVS. Material and
methods: All cases of Type III UPSVS diagnosed at our department from April 2016 to December 2020 were retrospectively
studied. Results: Seventeen patients with type III UPSVS including 12 type IIIa and 5 IIIb cases were identified. Sonography
showed a shunt between the inferior left portal vein and the left hepatic vein in all type IIIa cases. Three cases of type IIIb had
a combination of another shunt (2 with type I and one with type IIIa). Integrate intrahepatic portal vein system was not seen
in those 2 cases of type IIIb combined with type I UPSVS, leading to termination of pregnancy (TOP). TOP occurred in 4
patients with type IIIa as requested by the parents. Two cases (type IIIa and type IIIb each) underwent surgical procedure for
the closure of the shunt. Spontaneous complete closure in 4 type IIIa cases and partial closure in one type IIIb case occurred
during a period of 3-16 months. Conclusions: The majority of patients had type IIIa UPSVS presenting a good outcome. The
lack of integrate intrahepatic portal vein system was the main reason for TOP in patients with type IIIb UPSVS. These data
suggest the UPSVS classification is a useful tool for a prognosis prediction of type III UPSVS.
Keywords: ultrasonography; umbilical-portal-systemic venous shunt; portosystemic shunt; termination of pregnancy;
prognosis prediction

Introduction case studies [5-8]. Nevertheless, the manifestations and

An abnormal communication between the portal vein
and the systemic venous system causes the portosystemic
shunt (PSS), resulting in complete or partial diversion
of the portal flow from the liver to the systemic venous
circulation [1]. This condition, also termed as “congeni-
tal portosystemic shunt”, has been described mostly in
the pediatric population [1-3]. With the development
of prenatal screening techniques, especially ultrasonog-
raphy [4], diagnosis of fetal PSS has been described in
Received 19.03.2021  Accepted 06.06.2021
Med Ultrason
2022, Vol. 24, No 1, 14-18
Corresponding author: Prof. Guowei Tao
Department of Ultrasound,
Qilu Hospital of Shandong University, 1
07 Wenhua West Road, Jinan 250012, China
E-mail: taoguowei@yahoo.com
Phone: 0086-531-82169114
evolution of fetal PSS, which are important for prenatal
counselling and perinatal management, are still poorly
understood [9].
The umbilical vein (UV), portal vein (PV) and ductus
venosus (DV) in fetus form an integral conduit, through
which the highly oxygenated blood is transported from
the placenta into the fetal heart [10]. Malformation of
UV, PV and DV can cause shunting into the systemic
circulation. In 2016, Achiron and Kivilevitch defined the
“umbilical-portal-systemic venous shunt (UPSVS)” and
proposed a novel classification of in-utero UV-PV-DV
anomalies [11]. According to the anatomical origin of the
shunt (umbilical, portal or ductal), the UPSVS is clas-
sified into 4 types [11]. The type I shunt, the umbilical-
systemic shunt, is the direct drainage of the UV into the
systemic circulation. In type II shunt (ductus venosus-
systemic shunt) and for this type of shunt, the UV, PV
and DV are intact but the DV connects to the inferior
vena cava (IVC) below the pre-diaphragmatic infun-
 Med Ultrason 2022; 24(1): 14-18 15
dibulum or the DV drains into the hepatic vein. Type patic portal venous system and hepatic veins spotted by
III shunt, (portal-systemic shunt) is further divided into the 3-plane scan was diagnosed as type IIIa UPSVS and
two subtypes: IIIa, i.e., the intrahepatic portal-systemic any shunt between the portal system and systemic veins
shunt (IHPSS), for which, a communication between the (IVC, iliac vein and renal vein) identified by the 3-plane
intrahepatic portal venous system and hepatic veins oc- scan was diagnosed as type IIIb UPSVS [11].
curs; and IIIb, i.e., the extrahepatic portal-systemic shunt Follow-ups were carried out 1 month, 3-6 months and
(EHPSS), which is defined as a shunt between the por- 12 months post discharge, and once every year thereafter.
tal system and systemic veins (IVC, iliac vein and renal Rearrangements were made for missing appointments.
vein). A more detailed description of each type of UP- Abdominal ultrasound examination was performed to
SVS can be viewed in the schematic diagrams created by check whether the shunt was spontaneously closed or not.
Achiron and Kivilevitch [11]. This in-utero classification Statistical analysis
has been shown to be valuable in prognosis prediction The student’s t test was used to compare the age of
and prenatal counseling [11,12]. To date, different post- patients in type IIIa and type IIIb groups. The chi-square
natal PSS categories have been used primarily for direct- test was performed to compare the incidence of associ-
ing the surgical repair of PSS in a largely pediatric popu- ated anomalies and the rate of termination of pregnancy
lation [13,14], which, however, are believed unsuitable between type IIIa and type IIIb patients. All statistical
for prenatal analysis of PSS [11]. Fetal PSS is the most analyses were done using the GraphPad Prism 8 (Graph-
common shunt seen in our clinical practice. Therefore, Pad Software Inc., San Diego, CA, USA). p<0.05 was
the aim of this study is to use the in-utero UPSVS clas- considered statistically significant.
sification to analyze the ultrasonographic characteristics
and the outcome of fetal PSS, which has been, until now, Results
poorly documented.
A total of 17 patients with type III UPSVS, which
Material and methods include 12 cases of type IIIa and 5 cases of type IIIb,
were identified. The maternal age was 30.1±5.5 years
This study was approved by the Research Ethics and the gestation age at diagnosis was 33.0±4.5 weeks.
Committee of our institution (approval No. 2017049). There was no significant age difference between mothers
The Research Ethics Committee waived informed con-
sent as the study is retrospective.
All cases of type III UPSVS diagnosed at our depart-
ment from April 2016 to December 2020 were included
in the present study. The demographic data, ultrasono-
graphic findings, pregnancy outcome and follow-up re-
sults were collected and analyzed following the in-utero
UPSVS classification [11].
For regular fetal monitoring, the standard 2D proce-
dure was conducted using the Philips iU22 ultrasound
system with a C5-1 transducer. When portal-systemic
venous abnormalities were suspected, a 3-plane scan was
further performed using the iU22 with a C5-1 transducer
or the EPIQ7 with a C5-1 or an eL18-4 transducer. The 3
planes are: the transverse abdominal plane; the ventral or
lateral transverse plane; and the longitudinal anteropos-
terior plane [12,15]. At the transverse abdominal plane,
malformations of following structures can be detected:
UV, portal sinus, main portal vein, left portal vein, an-
terior right portal vein, posterior right portal vein and
splenic vein [15]. The ventral or lateral transverse plane
scan can identify anomalies in the right, middle and left Fig 1. Representative ultrasonography of type IIIa UPSVS. A
shunt between the inferior left portal vein (LPVi) and the left
hepatic veins and IVC [15]. The longitudinal anteropos- hepatic vein (LHV) is shown in this figure. UV: umbilical vein;
terior plane scan can discover abnormal UV, DV, left he- LPVs: superior left portal vein; IVC: inferior vena cava; AO:
patic vein and IVC [15]. Any shunt between the intrahe- aorta; and ST: stomach.
16 Linlin Zhu et al US characteristics and outcome of Type III umbilical-portal-systemic venous shunt
Table I. Demographic information, shunt characteristics and birth outcome.
Case GA Shunt characteristics Associated anomalies Outcome
1 36 LPVi-LHV (IIIa) None FTB
2 32 LPVi-LHV, LPV-LHV (IIIa) Splenomegaly, CTR↑ FTB
3 28 LPVi-MHV (IIIa) None FTB
4 33 LPVi-LHV (IIIa) IUGR PB at 35w
5 32 LPVi-LHV (IIIa) IUGR C-section at 37w
6 25 LPVi-LHV, LPV-MHV (IIIa) IUGR TOP
7 34 LPVi-LHV (IIIa) IUGR TOP
8 28 LPVi-LHV (IIIa) PH C-section at 40w
9 38 LPVi-LHV (IIIa) IUGR, CTR↑ TOP
10 35 LPVi-LHV, LPVs-LHV (IIIa) IUGR, splenomegaly TOP
11 39 LPVi-LHV (IIIa) None FTB
12 36 LPVi-LHV (IIIa) IUGR C-section at 39w
13 28 MPV-UV, UV-IVC (IIIb and Type I) CPC, PH, HSM, RHE TOP
14 28 SplV-IVC, SMV-IVC, UV-IVC (IIIb and Type I) Multiple IHC TOP
15 30 MPV-IVC (IIIb) None C-section at 38w
16 34 RPV-IVC (IIIb) IUGR, Strephenopodia TOP
17 40 PS-IVC, LPVs-LHV, LPVm-MHV, ARPV-RHV, ARPV-MHV, RHE, PH FTB
PRPV-RHV (IIIa and IIIb)
GA: gestational age (weeks) at diagnosis; LPV: left portal vein; LPVi: inferior LPV; LHV: left hepatic vein; MHV: middle hepatic vein;
MPV: main portal vein; IVC: inferior vena cava; UV: umbilical vein; PS: portal sinus; LPVs: superior LPV; LPVm: medial LPV; RPV: right
portal vein; ARPV: anterior right portal vein; PRPV: posterior right portal vein; RHV: right hepatic vein; SplV: splenic vein; SMV: superior
mesenteric vein; IUGR: intrauterine growth retardation; PH: polyhydramnios; CTR: cardiothoracic ratio; HSM: hepatosplenomegaly; CPC:
choroid plexus cysts; IHC: intrahepatic calcifications; RHE: right heart enlargement; TOP: termination of pregnancy; FTB: full-term birth;
and PB: premature birth.

with fetus with type IIIa and type IIIb (30.8±5.1 years vs.
32.6±7.9 years, p=0.65). Ultrasonographic findings of
the shunt in each case are presented in Table I.
All the 12 type IIIa cases had a shunt between the
inferior left portal vein (LPVi) and the left hepatic vein
(LHV) and a representative sonography showing such a
shunt is presented in figure 1 (case 1). Three type IIIb
cases had a combination of another shunt (2 with type I
and one with type IIIa) (Table I). The ultrasonographic
features of the case 17 that had type IIIb combined with
type IIIa are presented in figure 2.
Several associated anomalies were detected, the
most-common one being intrauterine growth restriction
(IUGR) followed by polyhydramnios and increased car-
diothoracic ratio. There was no significant difference in
the incidence of associated anomalies between type IIIa
and type IIIb patients (p=0.82). Prenatal karyotyping was
done in 2 cases (cases 12 and 13), which showed that

Fig 2. Representative ultrasonography of type IIIa combined with type IIIb UPSVS: a) a communication between the portal sinus
(PS) and inferior vena cava (IVC) was detected; b) a shunt between the posterior right portal vein (PRPV) and the right hepatic vein
(RHV) was observed; c) shunt between superior left portal vein (LPVs) and left hepatic vein (LHV). UV: umbilical vein; ST: stom-
ach; LPVi: inferior left portal vein; and MHV: middle hepatic vein.
 Med Ultrason 2022; 24(1): 14-18 17
case 13 had trisomy 21 while the other was normal. Ter- Table II. Shunt treatment and follow-up time and results.
mination of pregnancy (TOP) occurred in 4 cases of type Case Shunt Follow Ultrasonographic findings
IIIa as requested by the parents and 3 cases of type IIIb treatment time
due to the lack of intrahepatic portal vein system. Sta- (months)
tistical analysis showed that the rate of TOP in the two 1 None 32 Spontaneous closure at 16 m
groups was not substantially different (p=0.31). Among 2 None 24 Spontaneous closure at 3 m
the 10 survivors, case 3 had surgery 12 days post birth 3 SC at 12d 22 SC
to close the shunt and case 17 had surgery 4 months post 4 None 17 Spontaneous closure at 5 m
birth to close multiple right portal vein-right hepatic vein 5 None 15 Shunt not closed
shunts, while the rest did not undergo any surgical pro- 8 None 9 Shunt not closed
cedure. The follow-up time ranged from 1 to 32 months. 11 None 2 Shunt not closed
Ultrasonography showed spontaneous complete closure 12 None 1 Spontaneous closure at 1 m
in 4 cases and spontaneous partial closure in 1 case (case
15 None 1 Shunt not closed
17: the LPVm-MHV shunt was closed while the LPVs-
17 SC for 10 LPVs-LHV shunt not closed
LHV shunt was not), while the shunt in 4 cases were not multiple at 7 m
closed during the follow-up period (Table II). RPV-RHV
shunts
Discussion at 4 m
SC: surgical closure; RPV: right portal vein; RHV: right hepatic
Postnatal PSS categories have been proposed with vein; LPVs: superior left portal vein; and LHV: left hepatic vein;
m: months
the aim to provide criteria for surgical repair of PSS in
largely pediatric patients [13,14]. However, these post- ther review of the 12 cases reported by Francois et al, we
natal classifications refer only to the survivors, do not found 2 cases belong to the type I UPSVS according to
take into account the unique feature of the fetal UV-PV- the Achiron and Kivilevitch classification.
DV structure, lacking two essential components of the It has been shown that patients with type I and II UP-
fetal venous complex, i.e, the UV and the DV and are SVS have a high incidence of trisomy 21 [11,17]. In con-
therefore believed unsuitable for the prenatal analysis trast, none of the 16 type III cases had aneuploidy [11].
of portal-systemic venous anomalies. In view of this, We identified one case with trisomy 21. Of note, this case
Achiron and Kivilevitch proposed an in-utero classifica- had combined type I UPSVS that has been reported to be
tion for fetal UPSVS and showed its value in the pre- associated with trisomy 21 [17]. The genetic testing rate
natal analysis of UV-PV-DV abnormalities for prognosis was low in our series and solid data that support genetic
prediction and prenatal counselling [11]. In the present testing for patients with type III UPSVS are still lacking.
study, we applied the in-utero classification to analyze Nevertheless, if patients have type III UPSVS combined
the ultrasonographic characteristics of 17 cases of fetal with type I or type II UPSVS, genetic testing may be nec-
type III UPSVS and reported the following findings: 1) essary.
the majority of cases had type IIIa UPSVS; 2) type IIIb In our series, TOP in 4 type IIIa cases was pursued
in combination with type I UPSVS was associated with by the parents; TOP in 3 type IIIb cases (60%) was de-
poor development of intrahepatic portal vein system, termined by the lack of intrahepatic portal vein system.
leading to dismal outcome; 3) approximately half of the A previous study showed that 2 out of a total of 4 type
cases (8/17) had IUGR; and 4) spontaneous shunt closure IIIb cases (50%) underwent TOP due to the poor devel-
(including complete and partial) occurred in half of the opment of the intrahepatic portal vein system [11], in line
survivors (5/10) during a period of 1-16 months. with our findings. However, live birth was found to oc-
Our series is the largest analyzed by the UPSVS clas- cur in all 12 type IIIa cases by Achiron and Kivilevitch
sification. Using the UPSVS classification. Achiron et [11], which is in contrast to our results. However, caution
al retrospectively analyzed 16 cases of prenatally diag- should be taken in the interpretation of our data, as TOP
nosed type III shunts and showed that 75% (12/16) had in all 4 type IIIa cases was not dictated by the nature of
type IIIa shunts [11], which is in line with our finding. the shunt but rather pursued by the parents, which artifi-
More recently, Francois et al described the application cially increased the TOP rate. Additionally, the sample
of the Park classification for the prenatal analysis of 12 size is small, which may limit the statistical power.
cases of intrahepatic PSS and revealed that only 3 cases Few authors have reported spontaneous postnatal
had IUGR [5]. Of note, the Park classification was pro- closure of type III shunts [6]. We observed spontaneous
posed for the postnatal analysis of PSS [16] and, by fur- complete closure in 4 cases. Han et al retrospectively
18 Linlin Zhu et al US characteristics and outcome of Type III umbilical-portal-systemic venous shunt
studied 6 fetuses with shunts between portal and hepatic
vein systems that are in line with the definition of the type
IIIa shunt under the Achiron and Kivilevitch classifica-
tion. Their follow up results showed 5 cases had spon-
taneous closure at approximately 1 year after birth [6].
Conclusion
This is the largest series of Type III shunt prenatally
analyzed by ultrasonography using the new in-utero UP-
SVS classification. We conclude that the majority of type
III UPSVS cases belong to type IIIa with good outcome
and the lack of integrate intrahepatic portal vein system
is the main reason for TOP in type IIIb cases. These data
suggest the UPSVS classification is a useful tool for prog-
nosis prediction of fetal PSS and prenatal counseling.
Acknowledgements: This study was supported by
the Provincial Key Research and Development Fund of
Shandong Province, China (Grant #: 2015GSF118081
and 2016GSF201141).
Conflict of interest: none.
References
1. Sokollik C, Bandsma RH, Gana JC, van den Heuvel M,
Ling SC. Congenital portosystemic shunt: characteriza-
tion of a multisystem disease. J Pediatr Gastroenterol Nutr
2013;56:675-681.
2. Franchi-Abella S, Branchereau S, Lambert V, et al. Com-
plications of congenital portosystemic shunts in children:
therapeutic options and outcomes. J Pediatr Gastroenterol
Nutr 2010;51:322-330.
3. Chocarro G, Amesty MV, Encinas JL, et al. Congenital
portosystemic shunts: clinic heterogeneity requires an in-
dividual management of the patient. Eur J Pediatr Surg
2016;26:74-80.
4. Yagel S, Kivilevitch Z, Cohen SM, et al. The fetal venous
system, Part II: ultrasound evaluation of the fetus with con-
genital venous system malformation or developing circula-
tory compromise. Ultrasound Obstet Gynecol 2010;36:93-
111.
5. Francois B, Gottrand F, Lachaux A, Boyer C, Benoit B, De
Smet S. Outcome of intrahepatic portosystemic shunt diag-
nosed prenatally. Eur J Pediatr 2017;176:1613-1618.
6. Han BH, Park SB, Song MJ, et al. Congenital portosystem-
ic shunts: prenatal manifestations with postnatal confirma-
tion and follow-up. J Ultrasound Med 2013;32:45-52.
7. Delle Chiaie L, Neuberger P, Von Kalle T. Congenital in-
trahepatic portosystemic shunt: prenatal diagnosis and pos-
sible influence on fetal growth. Ultrasound Obstet Gynecol
2008;32:233-235.
8. Gorincour G, Droulle P, Guibaud L. Prenatal diagnosis of
umbilicoportosystemic shunts: report of 11 cases and review
of the literature. AJR Am J Roentgenol 2005;184:163-168.
9. Francois B, Lachaux A, Gottrand F, De Smet S. Prenatally
diagnosed congenital portosystemic shunts. J Matern Fetal
Neonatal Med 2018;31:1364-1368.
10. Yagel S, Kivilevitch Z, Cohen SM, et al. The fetal venous
system, part I: normal embryology, anatomy, hemodynam-
ics, ultrasound evaluation and Doppler investigation. Ultra-
sound Obstet Gynecol 2010;35:741-750.
11. Achiron R, Kivilevitch Z. Fetal umbilical-portal-systemic
venous shunt: in-utero classification and clinical signifi-
cance. Ultrasound Obstet Gynecol 2016;47:739-747.
12. Wu H, Tao G, Cong X, et al. Prenatal sonographic char-
acteristics and postnatal outcomes of umbilical-portal-sys-
temic venous shunts under the new in-utero classification: A
retrospective study. Medicine (Baltimore) 2019;98:e14125.
13. Blanc T, Guerin F, Franchi-Abella S, et al. Congenital
portosystemic shunts in children: a new anatomical clas-
sification correlated with surgical strategy. Ann Surg
2014;260:188-198.
14. Matsuura T, Takahashi Y, Yanagi Y, et al. Surgical strategy
according to the anatomical types of congenital portosys-
temic shunts in children. J Pediatr Surg 2016;51:2099-2104.
15. Yagel S, Cohen SM, Valsky DV, Shen O, Lipschuetz M,
Messing B. Systematic examination of the fetal abdominal
precordial veins: a cohort study. Ultrasound Obstet Gy-
necol 2015;45:578-583.
16. Park JH, Cha SH, Han JK, Han MC. Intrahepatic portosys-
temic venous shunt. AJR Am J Roentgenol 1990;155:527-
528.
17. Dong X, Wu H, Zhu L, et al. Prenatal ultrasound analy-
sis of umbilical-portal-systemic venous shunts concurrent
with trisomy 21. J Ultrasound Med 2020. doi:10.1002/
jum.15507.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 19-26
DOI: 10.11152/mu-3124

The role of lung ultrasonography in predicting the clinical outcome

of complicated community-acquired pneumonia in hospitalized
children
Marcela Daniela Ionescu1,2, Mihaela Bălgrădean1,2, Cristina Filip2, Roxana Taraș1,2,
Georgiana Mihaela Căpitănescu2, Florian Berghea1,3, Elena Camelia Berghea1,2, Cătălin
Gabriel Cîrstoveanu1,2

1”Carol Davila” University of Medicine and Pharmacy, 2”Marie Curie” Emergency Children’s Hospital,
3”Sfânta Maria” Clinical Hospital, Bucharest, Romania

Abstract
Aims: This study’s objective was to analyze lung ultrasonography (LUS) characteristics in hospitalized pediatric patients
with complicated community-acquired pneumonia (CAP). We hypothesized that LUS could be correlated with the clinical out-
come in these cases. Materials and methods: In this retrospective study, we evaluated the LUS appearances (at admission and
five days after the beginning of the treatment) and the progression of complicated CAP. Results: We identified 45 patients who
fulfilled the inclusion criteria. Several complications occurred in these subjects during follow-up including: serofibrinous pleu-
risy (62.2%), empyema (15.6%), encapsulated pleurisy (11.1%), lung abscess (6.7%) and necrotizing pneumonia (2.2%). In
addition, 22.2% of the patients required surgical treatment: draining tube (11.1%), decortication (6.7%) and resection (4.4%).
Intensive care unit admission was needed in 8.9% of patients. The median duration of hospitalization was 14 [9.7; 19.7] days.
The thickness of pleural effusion with a cut-off value of 10 mm seen by LUS was a predictor for the need for continuous
thoracic drainage (p<0.01), segmentectomy or thoracoscopic surgery (p=0.03) and prolonged hospitalization over 10 days
(p<0.01). Hyperechogenic pleural effusion, presence of septa and fluid bronchogram on 1st LUS evaluation were independent
predictors of segmentectomy or thoracoscopic decortication (p<0.01) and of longer hospitalization (p=0.02, p<0.01, p<0.01
respectively). Conclusions: The ultrasound characteristics of complicated CAP can offer valuable information to predict the
clinical evolution of CAP and so can help the development of personalized medical management plans in these patients.
Keywords: lung; ultrasonography; paediatric; pneumonia; pleural effusion

Introduction tentially severe and considerable morbidity and mortal-

Community-acquired pneumonia (CAP) is a common
infectious disease in children, involving the alveoli and
variable proportion of pulmonary parenchyma, with po-
Received 21.02.2021  Accepted 22.06.2021
Med Ultrason
2022, Vol. 24, No 1, 19-26
Corresponding author: Elena Camelia Berghea
“Carol Davila” University of Medicine and
Pharmacy, ”Marie Curie” Emergency
Children’s Hospital, Bucharest, Romania
20 Bd Constantin Brâncoveanu,
75534 Bucharest, Romania
Phone: 0722565117
E-mail: bcamelia@gmail.com
ity. It is defined as the presence of signs and symptoms
of pneumonia in a previously healthy child who acquired
the infection outside of a healthcare setting [1,2]. The an-
nual incidence of pneumonia in developed countries is
estimated to be 40/1000 in children younger than 5 years
of age and 20/1000 in children over 5 years of age [3].
Worldwide, CAP remains the largest cause of death in
children and a major health issue. Approximately one-
half of children younger than 5 years of age with CAP
require hospitalization. [1-3]. The true incidence of pneu-
monia is difficult to define because many other clinical
entities that express low respiratory tract infections are
labelled as “pneumonia”. No single sign or symptom
is pathognomonic for CAP in children and the diagno-
20 Marcela Daniela Ionescu et al Lung US & complicated community-acquired pneumonia in hospitalized children
sis may be challenging. Fever and cough in presenting
pictures suggest pneumonia, but non-specific symptoms,
such as abdominal pain and nuchal rigidity, may create
significant difficulties [1]. The etiological diagnosis is
difficult and neither clinical nor radiologic features dis-
tinguish between bacterial, atypical bacterial and viral
pneumonia [2,4]. Pleural effusion, empyema, necrotizing
pneumonia, lung abscess and pneumatocele may compli-
cate CAP in children [5].
Although not recommended routinely for diagnosis in
children, chest radiography (chest X-ray) is the most used
investigation for detecting CAP lesions. The technique
has certain limitations as non-severe pneumonia cases
may show normal chest X-ray or just perihilar changes
while minimal pleural effusion may go undetected [1,3].
Few studies have shown that radiological findings are as-
sociated with the severity of CAP in pediatric patients.
Despite that, the prognostic role of chest radiography in
children with CAP has not been established [6,7].
Thereby, the clinicians’ interest shifted towards lung
ultrasonography (LUS), considered in time a safe and ac-
cessible option for diagnosing pneumonia and its com-
plications [8]. Recent systematic reviews have verified
the high accuracy of LUS for diagnostic of pneumonia,
concluding that it is a useful imaging alternative to chest
radiography inclusive in pediatric CAP [9,10]. Even if
ultrasound’s usefulness is limited in evaluating a well-
aerated lung due to the imperfect transmission of the
sound waves, it is an essential diagnostic tool in the pres-
ence of consolidations, allowing the evaluation of lung
parenchyma [11]. LUS can detect small pulmonary le-
sions not visible on chest X-ray, suggesting that this is
the preferred approach for assessing patients with small
consolidations or small parapneumonic pleural effusion
[6,9,12-15]. Fluid or aerial bronchogram may be well de-
scribed, suggesting a consolidation process. LUS is also
fairly useful in evaluating the pleural effusion, providing
the best detail of the fluid nature, quantity, consistency,
echogenicity. It can demonstrate the presence of internal
septations, cellular debris, honeycombing, pleural thick-
ening, or the lack of free movements with gravity, sug-
gesting complications that cannot be well characterized
on static X-ray or computer tomography images [6,7,11].
Besides, LUS can help guide any invasive diagnostic or
therapeutic procedures [11].
This study aimed at evaluating the LUS character-
istics of complicated CAP in hospitalized children at
baseline and 5 days after beginning the antibiotic treat-
ment and to assess the LUS value for predicting the clini-
cal outcome in these children. Therefore, we examined
which LUS findings were risk factors for a poor outcome
in complicated CAP, requiring specific invasive thera-
peutic procedures. The following outcomes were pur-
sued: needing continuous thoracic drainage, requiring
surgical interventions consisting of either segmentecto-
my or video-assisted thoracoscopic surgery decortication
and demanding prolonged hospitalization over 10 days.
Material and methods
Study design
We conducted a retrospective, observational, unicen-
tric study. Data were collected from electronic medical
records of pediatric patients with the diagnosis of CAP
requiring hospitalization between 2017-2019 in “Marie
S. Curie” Emergency Children’s Hospital, Bucharest,
Romania. All patients’ records and clinical information
were analyzed anonymously. The study was conducted
with the approval of the local Ethics Committee.
Patients’ medical files were studied for demographic
information, clinical and paraclinical data, treatment,
clinical course, and discharge summary. The diagnosis
of CAP was established according to the British Tho-
racic Society guidelines [1]. Clinical findings (fever,
cough, chest pain, respiratory distress, gastrointestinal
complaints and other symptoms, respiratory rate, oxygen
saturation), laboratory tests (complete blood count, C-
reactive protein - CRP and procalcitonin level) and imag-
istic investigations (chest X-ray and LUS) at admission
and during hospitalization were noted. All patients were
hospitalized and a 5th day LUS control was performed,
while control blood investigations were carried out only
when considered by the physician. The chest X-rays and
LUS clips or images were digitally archived in the hos-
pital informatic system and they were reanalyzed for this
study by an experienced senior pediatric radiologist.
Performance of LUS
LUS was performed soon after complicated CAP was
suspected, using the GE LogiqS8 ultrasound system,
with convex (C1-6) and linear (L3-12) probes. Anterior,
lateral and posterior intercostal spaces were examined in
longitudinal and transverse sections with patients in su-
pine and sitting positions.
All the enrolled patients underwent first LUS at ad-
mission and the second LUS 5 days after the beginning
of treatment. Per center’s capabilities, the second LUS
was performed by the same person that made the first ex-
amination – this approach reduced possible interobserver
variability. Findings were reported as normal or LUS
features of pneumonia were recorded including: the pres-
ence of pulmonary parenchymal lesions (consolidation/
atelectasis), number, size and localization of lesions, the
presence of bronchogram and its characteristics (air or
fluid bronchogram), the presence and aspect of pleural
 Med Ultrason 2022; 24(1): 19-26 21

Fig 1. A 2-year-old girl with severe community-acquired pneu-

monia who needed surgical treatment consisting of video-assist-
ed thoracoscopic decortication, along with complex antibiotic
and supportive therapy, requiring prolonged hospital stay over
10 days: A) Chest X-ray (1st): right basal pulmonary opacity,
with heterogeneous structure and air bronchogram, suggestive
for right lower lobe consolidation; blunting the costal phrenic
angle and right pleural homogenous opacity, with coastal inten-
sity, net delimitation representative for pleural parapneumonic
effusion; B) LUS (1st): right lateral-thoracic pleural effusion,
with lenticular shape, echogenic appearance, and small free
movements with gravity, having 3 cm length and a maximum
thickness of 7 mm; C) Chest X-ray (2nd): progression of later-
al-thoracic opacity, with coastal intensity and tendency to lock
on the lateral thoracic wall; D) LUS (2st): right lateral-thoracic
pleural effusion, with multiple cellular debris, septa and lack of
free movements with gravity.

effusion (echogenicity, homogeneity, thickness, the pres-
ence of septa).
Two examples of comparative evolutive findings on
the chest X-ray and LUS are presented in figure 1 and 2.
Inclusion criteria
We included pediatric patients (age under 18 years
old) with admission diagnosis of complicated CAP, based
on history, initial clinical examination and chest radiog-
raphy, who underwent LUS at admission and 5 days after
the beginning of antibiotic therapy.
Severe CAP was considered according to the British
Thoracic Society guidelines, including persistent fever,
tachypnea, respiratory distress, cyanosis, grunting respi-
rations, hypoxemia, tachycardia, prolonged capillary re-
fill time, signs of dehydration or intense positively acute
phase reactants (leukocytosis, CRP, procalcitonin) [1].
We defined as complicated CAP those patients presenting
with clinical, biological and chest X-ray features sugges-
tive for local development of pleural effusion or empy-
ema, necrotizing pneumonia and lung abscess.

Fig 2. A 14-year-old boy with complicated community-acquired pneumonia who needed continuous thoracic drainage and complex
antibiotic therapy: A) Chest X-ray (1st): alveolar opacity with low intensity and flue contour of the inferior right pulmonary lobe; right
basal pulmonary opacity, with homogenous structure, costal power, net delimitation and blunting of the costal phrenic angle, sug-
gestive for medium pleural effusion, with minimal passive pulmonary collapse; B) LUS (1st): right lateral-thoracic pleural effusion,
with transonic liquid, in large quantity (up to axilla level in orthostatic position), determining passive lung collapse of the inferior
right pulmonary lobe; C) Chest X-ray (2nd): slight progression of right pleurisy; D) LUS (2st): persistent right lateral-thoracic pleural
effusion, with transonic aspect, and free movements with gravity, without septa or cellular debris.
22 Marcela Daniela Ionescu et al Lung US & complicated community-acquired pneumonia in hospitalized children
Table I. Cohort characteristics of the 45 patients enrolled.
Parameter Value
Male gender 30 (66.7)
Age (months) 42 [25.7; 69.7]
Prior days of fever 5 [3; 7]
Prior days of cough 5 [4; 7.3]
Grunting 9 (20)
Chest pain 8 (17.8)
Respiratory rate (breaths/minute) 40 [30; 46.7]
Respiratory distress
Minimum 12 (26.7)
Moderate 23 (51.1)
Severe 10 (22.2)
Oxygen saturation 94 [90.6; 97]
Other associated symptoms
Fig 3. Flowchart of the study.
Sleepiness 5 (11.1)
Loss of appetite 10 (22.2)
Exclusion criteria Diarrhea 1 (2.2)
Patients with underlying disease, including respira- Vomiting 8 (17.8)
tory tract anomalies, malignancy, immunodeficiency, Abdominal pain 3 (6.7)
neurologic disorders (cerebral palsy, neuromuscular dis- None 18 (40)
eases), congenital heart disease, were excluded. White blood cell count (x103) 19000 [14566; 24083]
The flowchart of the study is presented in Fig 3. PMN 78 [64.7; 85]
Statistical analysis CRP (mg/dL) 145 [76.3; 282]
We performed statistical analysis and graphs using Procalcitonin (ng/dL)
the Analyze IT 5.5 program (Microsoft Office Excel <0.5 17 (37.8)
Add-on, Leeds, UK). Continuous variables had a non- 0.5-2 6 (13.3)
gaussian distribution and were presented as the median 2-10 10 (22.2)
and the interval between the quartiles. Categorical vari- >10 12 (26.7)
ables were presented as numbers and percentages. Dif- Severe CAP 18 (40)
ferences in quantitative parameters were tested using Local complications
nonparametric tests (Kruskal-Wallis). Qualitative data Serofibrinous pleurisy 28 (62.2)
Empyema 7 (15.6)
were compared with the chi-square test. We considered
Encapsulated pleurisy 5 (11.1)
statistical significance at a p-value lower than 0.05.
Lung abscess 3 (6.7)
Necrotizing pneumonia 1 (2.2)
Results
General complications
SIRS 6 (13.3)
Patients’ characteristics Sepsis 23 (51.1)
Table I summarizes the main features of the included Severe sepsis 2 (4.4)
patients: demographic data, clinical and laboratory char- Septic shock 1 (2.2)
acteristics on admission, local and general complications Surgery 10 (22.2)
and required major therapeutic procedures. Draining tube 5 (11.1)
The evolution of the baseline pleural and pulmo- Thoracotomy and decortication 3 (6.7)
nary lesions (localization, size and aspect of lung con- Surgical resection 2 (4.4)
solidation and pleural effusion, presence and aspect of ICU 4 (8.9)
bronchogram) was appreciated using the second LUS as- Hospital stay (days) 14 [9.7; 19.7]
sessment, performed after 5 days of treatment. Baseline Sequelae < 6 months 15 (33.3)
radiologic and LUS findings 5 days after the beginning Pahipleuritis 13 (28.9)
of treatment in all 45 patients are summarized in Table II. Segmentectomy 2 (4.4)
Relation between LUS characteristics and Continuous variables are presented as median [IQR]. Categorical
clinical outcome variables are presented as number (%). CRP = C reactive protein;
PMN = polymorphonuclear leucocytes; ICU = Intensive Care Unit;
We analyzed the relation between LUS character- IQR = Interquartile Range; SIRS = Systemic Inflammatory Re-
istics and clinical outcome. Using univariate analysis, sponse Syndrome
 Med Ultrason 2022; 24(1): 19-26 23
we examined which LUS findings were risk factors for tions: the need for continuous thoracic drainage, indi-
a poor outcome in complicated CAP. We considered the cation for surgical interventions (consisting of either
poor outcome at least one of the three following condi- segmentectomy or video-assisted thoracoscopic surgery
decortication), and requiring prolonged hospitalization
Table II. Cohort imagistic features of the 45 patients enrolled over 10 days.
Parameter Value Regarding the first outcome, we noted that LUS could
Chest radiograph (at admission) predict the need for continuous thoracic drainage in com-
Multilobed consolidation 3 (6.7) plicated CAP at a pleural effusion thickness cut-off value
Pleural effusion 40 (88.9) of 10 mm. The ultrasound identification of aerial bronch-
Lung abscess 2 (4.4) ogram had a statistically significant protective value re-
Consolidation 42 (93.3) garding the need for continuous thoracic drainage. Those
1st LUS (at admission) patients often presented a good clinical outcome and did
Consolidation 43 (95.6) not need invasive therapeutic procedures, including tho-
Pleural effusion 45 (100) racic drainage. The fluid bronchogram was dominant
Thickness (mm) 7 [5; 15] in patients that required continuous thoracic drainage,
Echogenicity alongside conservative treatment (Table III).
Isoechogenic fluid 15 (33.3) Concerning the second outcome, the need for surgi-
Hyperechogenic fluid 8 (17.8) cal interventions including segmentectomy or video-
Transonic fluid 22 (48.9)
assisted thoracoscopic surgery decortication, we found
Homogeneity 32 (71.1)
that the hyperechogenic pleural effusion, the thickness
Septa 13 (28.9)
of pleural effusion with a cut-off value of 10 mm, the
Multilobed lesions 8 (17.8)
Air bronchogram 26 (57.8)
presence of septa and the fluid bronchogram on 1st LUS
Fluid bronchogram 16 (35.6) evaluation predicted a poor outcome and a worse clinical
2nd LUS (5 days after admission) evolution, requiring surgical therapeutic procedures. On
Consolidation 23 (56.1) the other hand, the presence of aerial bronchogram and
Improvement vs. 1st 21 (46.7) the homogenous pleural effusion alongside the improved
Pleural effusion 17 (41.5) appearance of pleural effusion on the 2nd LUS evaluation
Improvement vs. 1st 25 (55.6) were associated with a better clinical course (Table IV).
Continuous variables are presented as median [IQR]. Categorical Using univariate analyses, it was also examined
variables are presented as number (%) whether LUS findings were risk factors for a poor out-

Table III. Association between LUS findings and the need for continuous thoracic drainage
Parameter Thoracic tube drainage Univariate analysis p-value
Yes (5) No (40) Odds ratio 95% confidence interval
1st LUS (at admission)
Consolidation 4 (80) 39 (97.5) 0.1 0.009-1.19 0.07
Hyperechogenic fluid 2 (40) 6 (15) 3.77 0.61-24.52 0.16
Homogeneity 2 (40) 30 (75) 0.22 0.03-1.32 0.1
Thickness>10 mm 5 (100) 12 (30) 2.66 2.66- <0.01
Septa 3 (60) 10 (25) 4.5 0.75-26.43 0.1
Multifocal lesions 2 (40) 6 (85) 0.26 0.04-1.63 0.16
Aerial Bronchogram 0 (0) 26 (65) 2.13 2.13- <0.01
Fluid Bronchogram 4 (80) 12 (30) 0.1 0.01-0.83 0.02
2nd LUS (5 days after admission)
Consolidation 4 (80) 19 (52.5) 4.42 0.57-32.06 0.17
Improvement vs. 1st 1 (20) 20 (50) 4 0.52-29.01 0.2
Pleural effusion 5 (100) 12 (30) 2.66 2.66- <0.01
Improvement vs. 1st 0 (0) 25 (62.5) 1.92 1.92- <0.01
Severe CAP 4 (80) 14 (35) 7.42 0.96-54.25 0.052
Data are expressed as numbers (%). CAP = community acquired pneumonia; LUS = lung ultrasonography
24 Marcela Daniela Ionescu et al Lung US & complicated community-acquired pneumonia in hospitalized children

come, including prolonged hospitalization over 10 days. Discussion

For this outcome, we noted a poor evolution regarding
hyperechogenic pleural effusion, the thickness of pleural
effusion with a cut-off value of 10 mm, the presence of
septa and the fluid bronchogram on 1st LUS evaluation
(Table V). The presence of aerial bronchogram and the
homogenous pleural effusion alongside the improved ap-
pearance of pleural effusion on the 2nd LUS evaluation
predicted a better course for this outcome.
Our study analyzes the LUS characteristics of com-
plicated CAP in hospitalized pediatric patients at ad-
mission (baseline) and after the beginning of antibiotic
treatment. In addition, the study highlights how specific
LUS features may predict the clinical course of these
patients by assessing the following three outcomes: the
need for continuous thoracic drainage, requiring surgi-

Table IV. Association between LUS findings and the need for surgery

Parameter Surgery Univariate analysis p-value
Yes (5) No (40) Odds ratio 95% confidence interval
1st LUS
Consolidation 5 (100) 38 (95) NS 0.6
Hyperechogenic fluid 3 (60) 5 (12.5) 10.5 1.6-68.8 <0.01
Homogeneity 1 (20) 31 (77.5) 0.07 0.01-0.58 <0.01
Thickness > 10 mm 4 (80) 13 (32.5) 8.3 1.07-60.91 0.03
Septa 4 (80) 9 (22.5) 13.77 1.72-103.17 <0.01
Multifocal lesions 0 (0) 8 (80) NS 0.27
Aerial bronchogram 0 (0) 26 (65) 2.13 2.13- <0.01
Fluid bronchogram 5 (100) 11 (27.5) 0 0-0.33 <0.01
2nd LUS
Consolidation 4 (80) 19 (47.5) 4.42 0.57-32.06 0.17
Improvement vs. 1st 1 (20) 20 (50) 4 0.52—29.01 0.2
Pleural effusion 5 (100) 12 (30) 2.66 2.66- <0.01
Improvement vs. 1st 0 (0) 25 (62.5) 1.92 1.92- <0.01
Severe CAP 4 (80) 14 (35) 7.42 0.96-54.25 0.052
Data are expressed as numbers (%). NS: not significant; CAP = community acquired pneumonia; LUS = lung ultrasonography

Table V. Association between LUS findings and more extended hospital stay

Parameter Hospital stays over 10 days Univariate analysis p-value
Yes (30) No (15) Odds ratio 95% confidence interval
1st LUS, number, (%)
Consolidation 29 (96.2) 14 (93.3) 2.07 0.19-21.72 0.6
Hyperechogenic fluid 8 (26.7) 0 (0) 1.27 1.27- 0.02
Homogeneity 17 (56.7) 0 (0) 0 0-0.37 <0.01
Thickness >10 mm 16 (53.3) 1 (6.7) 16 2.26-106.27 <0.01
Septa 13 (43.3) 0 (0) 2.68 2.68- <0.01
Multifocal lesions 6 (20) 2 (13.3) 0.61 0.12-3.2 0.58
Aerial bronchogram 14 (46.7) 12 (80) 4.57 1.1-18.38 0.038
Fluid bronchogram 16 (53.3) 0 (0) 0 0-0.25 <0.01
2nd LUS, number, (%)
Consolidation 18 (60) 5 (33.3) 3 0.81-10.74 0.091
Improvement vs. 1st 12 (40) 9 (60) 2.25 0.64-7.85 0.2
Pleural effusion 14 (46.7) 3 (20) 3.5 0.84-14.04 0.08
Improvement vs. 1st 13 (43.3) 12 (80) 5.23 1.26-21.09 0.019
Severe CAP 18 (60) 0 (0) 5.19 5.19- <0.01
Data are expressed as numbers (%). CAP = community acquired pneumonia; LUS = lung ultrasonography

cal treatment, and the need for prolonged hospitalization
over 10 days. To our knowledge, this is the first study that
evaluated the predictive potential of the LUS character-
istics in hospitalized pediatric patients with complicated
CAP. Multiple studies have mainly focused on LUS ac-
curacy compared with chest X-ray findings in diagnosing
and evaluating pediatric CAP, offering very promising
results (overall high sensitivity – up to 95% – and high
specificity – up to 96%) [1,2,9,16-18]. Also, few studies
compared LUS with chest CT, claiming the accuracy of
LUS in the evaluation of CAP and its complications in-
clusively in pediatric patients [19,20].
All patients enrolled in our study presented local
complications of CAP, including serofibrinous pleurisy,
encapsulated pleurisy, pulmonary abscess or necrotizing
pneumonia. LUS evidence of pulmonary consolidation
with fluid bronchogram, hyperechogenic pleural fluid and
septa was demonstrated to be statistically significant pre-
dictors of poor clinical outcome, expressed as prolonged
hospitalization (more than 10 days), the need for continu-
ous thoracic drainage and the need for segmentectomy or
video-assisted thoracoscopy surgery decortication.
The fluid bronchogram, generally described in ap-
proximately 20% of CAP patients, less frequently than
the air bronchogram, represents an exudate packed con-
ducting airway that occurs in the early phase of the dis-
ease because of bronchial obstruction with secretions and
edema [14,21,22]. Our study described the presence of
fluid bronchogram as a negative predictor of complicated
CAP, needing surgical treatment and prolonged hospitali-
zation (p<0.01). Therefore, this has allowed for the con-
clusion that a fluid bronchogram revealed on LUS should
indicate the need to consider CAP’s potential progres-
sion to complications requiring supplemental therapeutic
measures.
However, the most characteristic sign of CAP re-
mains the festive air bronchogram which is generally
detectable in about 70-97% of cases, reflecting the re-
sidual air within the consolidation areas [6,23,24]. While
the fluid bronchogram dominated the positive outcome
group in our study, the presence of air bronchogram on
LUS performed on admission had a statistically signifi-
cant protective value (p<0.05).
Because a liquid allows for excellent propagation of
sound waves, LUS is a particularly good modality to as-
sess the presence of parapneumonic pleural effusions.
The role of LUS in assessing pleural effusion is very
well known, as well as the prediction of poor outcomes
when complex pleural lesions are detected, based mainly
on adult studies [9,18]. Our paper confirms these find-
ings and, in addition, it highlights the importance of LUS
in assessing CAP-associated pleural effusion in pediat-
Med Ultrason 2022; 24(1): 19-26 25
ric patients. Also, the study evaluates ultrasound pleural
effusion features in the early phase of CAP (admission
time) and after initiation of antibiotic therapy. We iden-
tified hyperechogenic pleural effusion and presence of
septa on 1st LUS evaluation as independent predictors of
segmentectomy or thoracoscopic decortication (p<0.01)
and of prolonged hospitalization, over 10 days (p=0.02,
p<0.01, respectively). The thickness of pleural effusion
with a cut-off value of 10 mm seen by LUS was also
associated with requiring continuous thoracic drain-
age (p<0.01), segmentectomy or thoracoscopic surgery
(p=0.03) and prolonged hospitalization over 10 days
(p<0.01). In the second LUS evaluation, we found that
the alleviated pleurisy predicted a better outcome with
a statistically significant relationship between the LUS
characteristics and clinical outcome.
Based on these findings, our study supports the con-
sideration of LUS for diagnosis and monitoring of com-
plicated CAP in hospitalized children and also suggests
that LUS may be a good early predictor of poor clinical
outcome in these patients Taken all together, our results
may impact daily clinical practice, by providing new
data to monitor treatment response, alongside with cur-
rent follow-up clinical (fever remission) and biological
(leukocytosis, CRP and procalcitonin level reduction)
parameters.
Study limits and strengths
The study included a relatively small number of chil-
dren, although it represents the entire patient population
(satisfying inclusion and exclusion criteria) from one of
the largest children’s hospitals. The unicentric design of-
fered the chance of a better-standardized work; simulta-
neously, the execution of LUS by a single sonographer
might raise the question of generalizability in the case of
less experienced hands. The retrospective design reduced
the control of the quality of available data. Although,
our study strengthens the role of LUS for early detec-
tion, monitoring and reliable prediction of severe out-
come in complicated pediatric CAP. LUS has proven to
be a useful tool to avoid more invasive interventions and
complex investigations, which could involve consuming
more significant material and human resources.
Conclusions
In conclusion, our study provides a detailed LUS anal-
ysis as an advantageous imaging predicting and monitor-
ing tool in pediatric patients hospitalized for complicated
CAP. It highlights early LUS features that may predict
poor clinical outcome in these patients. The results can
help physicians better manage a child with CAP and offer
a personalized approach, from diagnosis to treatment and
26 Marcela Daniela Ionescu et al Lung US & complicated community-acquired pneumonia in hospitalized children
follow-up. However, more extensive prospective studies
are necessary to support our findings regarding the as-
sociation between LUS and clinical outcome in pediatric
patients with complicated CAP.
Conflict of interests: none
References
1. Harris M, Clark J, Coote N, et al. British Thoracic Soci-
ety guidelines for the management of community-acquired
pneumonia in children: update 2011. Thorax 2011;66:ii1-
ii23.
2. Man SC, Fufezan O, Sas V, Schnell C. Performance of
lung ultrasonography for the diagnosis of community-ac-
quired pneumonia in hospitalised children. Med Ultrason
2017;19:276-281.
3. Everard ML. Paediatric respiratory infections. Eur Respir
Rev 2016;25:36-40.
4. Heiskanen-Kosma T, Korppi M, Jokinen C, et al. Etiology
of childhood pneumonia: serologic results of a prospective,
population-based study. Pediatr Infect Dis J 1998;17:986-
991.
5. Mbata G, Chukwuka C, Onyedum C, Onwubere B, Aguwa
E. The Role of Complications of Community-Acquired
Pneumonia on the Outcome of the Illness: A Prospective
Observational Study in a Tertiary Institution in Eastern Ni-
geria. Ann Med Health Sci Res 2013;3:365-369.
6. Reissig A, Gramegna A, Aliberti S. The role of lung ul-
trasound in the diagnosis and follow-up of community-
acquired pneumonia. Eur J Intern Med 2012;23:391-397.
7. Chen IC, Hsu JH, Wu JR, Dai ZK. Updated Guidelines for
Childhood Pneumonia Management: A Promising Role for
Lung Ultrasound. Pediatr Neonatol 2015;56:363-364.
8. Weinberg B, Diakoumakis EE, Kass EG, Seife B, Zvi ZB.
The air bronchogram: sonographic demonstration. AJR Am
J Roentgenol 1986;147:593-595.
9. Heuvelings CC, Bélard S, Familusi MA, Spijker R, Gro-
busch MP, Zar HJ. Chest ultrasound for the diagnosis of
paediatric pulmonary diseases: a systematic review and
meta-analysis of diagnostic test accuracy. Br Med Bull
2019;129:35-51.
10. Xin H, Li J, Hu HY. Is Lung Ultrasound Useful for Diag-
nosing Pneumonia in Children? A Meta-Analysis and Sys-
tematic Review. Ultrasound Q 2018;34:3-10.
11. Light M, Blaisdell CJ, Homnick DN, Schechter MS, Wein-
berger MW. (Eds.). Paediatric Pulmonology. American
Academy of Pediatrics, 2011.
12. Claes AS, Clapuyt P, Menten R, Michaux N, Dumitriu D.
Performance of chest ultrasound in paediatric pneumonia.
Eur J Radiol 2017;88:82-87.
13. Esposito S, Papa SS, Barzani I, et al. Performance of lung
ultrasonography in children with community-acquired
pneumonia. Ital J Pediatr 2014;40:37.
14. Shah VP, Tunik MG, Tsung JW. Prospective Evaluation
of Point-of-Care Ultrasonography for the Diagnosis of
Pneumonia in Children and Young Adults. JAMA Pediatr
2013;167:119-125.
15. Reali F, Sferrazza Papa GF, Carlucci P, et al. Can lung ultra-
sound replace chest radiography for the diagnosis of pneu-
monia in hospitalised children? Respiration 2014;88:112-
115.
16. Musolino AM, Tomà P, Supino MC, et al. Lung ultrasound
features of children with complicated and noncomplicated
community-acquired pneumonia: A prospective study.
Pediatr Pulmonol 2019;54:1479-1486.
17. Pereda MA, Chavez MA, Hooper-Miele CC, et al. Lung
Ultrasound for the Diagnosis of Pneumonia in Children: A
Meta-analysis. Paediatrics 2015;135:714-722.
18. Ambroggio L, Sucharew H, Rattan MS, et al. Lung Ul-
trasonography: A Viable Alternative to Chest Radiogra-
phy in Children with Suspected Pneumonia? J Pediatr
2016;176:93-98.e7.
19. Saraya S, El Bakry R. Ultrasound: Can it replace CT in the
evaluation of pneumonia in paediatric age group? Egypt J
Radiol Nucl Med 2017;48:687-694.
20. Kurian J, Levin TL, Han BK, Taragin BH, Weinstein
S. Comparison of Ultrasound and CT in the Evaluation
of Pneumonia Complicated by Parapneumonic Effu-
sion in Children. AJR Am J Roentgenol 2009;193:1648-
1654.
21. Chen IC, Lin MY, Liu YC, et al. The role of transthoracic
ultrasonography in predicting the outcome of community-
acquired pneumonia in hospitalised children. PLoS One
2017;12:e0173343.
22. Haggag YI, Mashhour K, Ahmed K, Samir N, Radwan
W. Effectiveness of Lung Ultrasound in Comparison with
Chest X-Ray in Diagnosis of Lung Consolidation. Open
Access Maced J Med Sci 2019;7:2457-2461.
23. Sperandeo M, Carnevale V, Muscarella S, et al. Clinical ap-
plication of transthoracic ultrasonography in patients with
pneumonia. Eur J Clin Invest 2011;41:1-7.
24. Byington CL, Spencer LY, Johnson TA, et al. An epidemio-
logical investigation of a sustained high rate of paediatric
parapneumonic empyema: risk factors and microbiological
associations. Clin Infect Dis 2002;34:434-440.
CeVUS and CEUS in children and adolescents – safety and parents´
acceptance
Original papers Med Ultrason 2022, Vol. 24, no. 1, 27-32
DOI: 10.11152/mu-3196

Safety and parents´ acceptance of ultrasound contrast agents in

children and adolescents – contrast enhanced voiding urosonography
and contrast enhanced ultrasound
Josefina Seelbach1, Paul C. Krüger1, Matthias Waginger1, Diane M. Renz2, Hans-Joachim
Mentzel1

1Section
of Pediatric Radiology, Institute of Diagnostic and Interventional Radiology, University Hospital Jena, Jena,
2Section
of Pediatric Radiology, Institute of Diagnostic and Interventional Radiology, Hannover Medical School,
Hannover, Germany

Abstract
Aims: To evaluate the safety of the contrast enhanced voiding urosonography (ceVUS) and contrast enhanced ultrasound
(CEUS) in children and adolescence and to receive data about parents’ acceptance of intravesical and intravenous application
of sulfur hexafluoride. Material and methods: In this prospective, single centre study conducted over a 1 year study period,
parents of 56 children (f/m=32/24; mean age 3.1 years; range 3 weeks - 15.9 years) with ceVUS and of 30 children (f/m=15/15;
mean age 10.5 years; range 2 months - 17.7 years) with CEUS agreed to be included. A standardized telephone survey about
the acceptance of the parents during and after the procedure as well as the adverse events (AE) were conducted within three
days of the examination. Results: The parents would agree with the use of both ceVUS and CEUS as a diagnostic tool again
in 96% (54/56) or 100% (30/30) of the cases, respectively and 92.9% (52/56) would prefer ceVUS to voiding cystourethrogra-
phy (VCUG). In addition, 83.3% (25/30) would prefer CEUS to CT and 73.3% (22/30) would prefer CEUS to MRI. AE were
reported in 3.6% after ceVUS (2/56; skin rash, mild fever) and in 3.3% after CEUS (1/30; vomiting). AE were subacute and
self‑limited. Conclusions: The vast majority of parents prefer ceVUS and CEUS to VCUG, CT or MRI because of the safety
profile of the contrast agent and diagnostic accuracy.
Keywords: child; Sulphur Hexafluoride; contrast-enhanced ultrasonography; contrast-enhanced voiding urosonography;
survey

Introduction
Contrast enhanced voiding urosonography (ceVUS)
and contrast enhanced ultrasound (CEUS) are adjunctive
techniques to conventional ultrasound. These techniques
are radiation free imaging modalities, which are impor-
tant especially for children [1]. Using ceVUS and CEUS
modalities as alternatives to radiography, fluoroscopy or
Received 08.04.2021  Accepted 14.07.2021
Med Ultrason
2022, Vol. 24, No 1, 27-32
Corresponding author: Hans-Joachim Mentzel
Institute of Diagnostic and Interventional
Radiology, University Hospital Jena
Am Klinikum 1, 07740 Jena, Germany
E-mail: hans-joachim.mentzel@med.uni-jena.de
Phone: 00493641 - 9 328 501
CT, one can reduce the radiation exposure during child-
hood or the need for sedation which is necessary in time
consuming MRI examinations [2-4].
CeVUS and CEUS have a diagnostic efficiency
comparable to conventional imaging methods. Studies
demonstrated that ceVUS has a sensitivity of 57-100 %
and specificity of 85-100 % in comparison to voiding
cystourethrography (VCUG) [5]. In addition, a study
showed that 96.2 % of the parents would prefer ceVUS
for further examinations [6]. CEUS has almost equal sen-
sitivity and specificity in comparison to MRI or CT and is
superior to fundamental B-mode sonography dependent
on the indication [7,8]. Until now there is no study evalu-
ating the acceptance of parents for CEUS examinations
in their children.
For ceVUS the ultrasound contrast agent (UCA) is
applied intravesically and for the CEUS examination the
28 Josefina Seelbach et al CeVUS and CEUS in children and adolescents – safety and parents´ acceptance
UCA is applied intravenously [9]. The UCA SonoVue®
(Bracco Imaging, Italy) is approved in Europe for the
intravesical application in children since 2017, but the
intravenous application is only possible with off-label
use. Whereas, in the USA Lumason® (=SonoVue) is also
approved for intravenous use since 2016 [1]. SonoVue®
consists of stabilized sulfur-hexafluoride microbubbles
[10] and is well tolerated [11,12]. Intravesically applied
SonoVue® is eliminated by micturition whereas intrave-
nously applied SonoVue® is eliminated by the lungs [10].
In consequence, SonoVue® can also be used in case of
renal failure [9].
There are few studies evaluating the risks of intra-
vesical or intravenous use of SonoVue® in children. In
intravesical use the minority of children shows adverse
events caused by catheterisation, mostly minor or moder-
ate and non-serious events [6,13,14]. In intravenous use
few paediatric cases showed adverse events, mostly mi-
nor or moderate. There were only two cases of serious
adverse events in children reported in the literature up to
now [15,16].
The aim of this study was to evaluate parents’ ac-
ceptance of contrast enhanced sonography in their chil-
dren and to ask if they would prefer ceVUS to VCUG
or CEUS to CT or MRI for a possible next examination.
Furthermore, this study was designed to evaluate the
safety profile of SonoVue®.
Materials and methods
The prospective study was approved by the local in-
ternal Ethics Review Board.
Patient Selection
Over a one-year study period, 55.4 % (56/101) of the
parents of all ceVUS examinations and 54.5 % (30/55)
of the parents of all CEUS examinations could be inter-
viewed. The other parents or legal guardians rejected the
participation in the survey because of time constraints or
no interest.
Before ceVUS and also before CEUS examinations
the parents were informed about the aim and asked to
participate in the study. All legal guardians were in-
formed about the off-label use of SonoVue®, proce-
dures and alternative imaging modalities for ceVUS and
CEUS. Alternative procedures, such as VCUG, CT and/
or MRI were explained in detail including information
about advantages and disadvantages of these methods.
Some parents were familiar with the alternative imag-
ing modalities due to prior examinations. Their informed
written consent was obtained prior to the examination.
Their knowledge about the procedures as well as the ad-
vantages and disadvantages were not analysed.
Exclusion criteria included patients aged >18 years,
a known sensitivity of sulfur hexafluoride or other com-
ponents of SonoVue®, cardiopulmonary disorders [17],
acute urinary tract infection for ceVUS or the lack of in-
formed consent of the legal guardians.
Ultrasound was performed by two certified paediatric
radiologists with experience in CEUS for more than ten
years (each >300 ceVUS, >100 CEUS). The indication
for ceVUS or CEUS was made in an interdisciplinary
consultation in coordination with the legal guardians.
ceVUS examination
CeVUS was performed using a 9-3 MHz convex
probe on a ZS3 ultrasound machine (Mindray, China).
Baseline pre-contrast ultrasound of the urinary tract
(bladder, ureter, kidneys) was conducted in supine and
prone position. Afterwards, the bladder was catheterized
under standardized aseptic conditions using a 6 CH feed-
ing tube with two lateral eyes (B. Braun, Germany). One
urine tube was sampled for laboratory examination and
the bladder was emptied. Then, the catheter was linked
to a three-way stopcock; one line (direct way) was con-
nected to the UCA and the other to the saline solution
bag. SonoVue® was always prepared in accordance with
the manufacturer’s recommendations and applied in a
sterile manner. 0.1 ml of SonoVue® was applied into the
bladder. After the administration of the UCA the bladder
was filled with prewarmed saline solution by drop infu-
sion (70 cm table height) until the estimated age-related
maximum bladder capacity was reached or the child
started to micturate. Under real-time ultrasound guidance
the distribution of the UCA in the bladder was observed
and the retrovesical and proximal ureters as well as the
ureteropelvine junction and the kidneys were examined
continuously during filling and voiding. During voiding
the urethra was explored by perineal positioned probe.
Filling and micturition were repeated up to four times in
each patient (minimum two times). To minimize destruc-
tion of the bubbles, the mechanical index was turned to a
level maximum of 0.10.
CEUS examination
For UCA administration application in the cubital
vein was preferred. In small infants, other positions (e.g.
scalp veins) were used. Baseline pre-contrast ultrasound
examination was tailored to the specific clinical query
9-3 MHz probe on a ZS3 ultrasound machine (Mindray,
China) or a 6-1 MHz convex abdominal probe ACUSON
S2000 (Siemens Healthineers, Germany). The amount
of SonoVue® was calculated using the formula 0.1 ml x
age in years as mentioned in the ESPR guidelines [18].
SonoVue® was always prepared according to the manu-
facturer’s recommendation and applied in a sterile man-
ner in the direct way of a three‑way cock. Single UCA

dose was administered with a repeated dose if necessary.
Each contrast bolus was followed by a saline bolus of
10  ml. The average amount of administered SonoVue®
was 1.4 ml ± 0.9 per patient. During the contrast-specific
examination mode the mechanical index was turned to
a low level (0.04 - 0.10) to minimize ultrasound related
disruptions of microbubbles.
Adverse event monitoring
During the examination and until 30 min thereafter,
all children with ceVUS and CEUS were observed for
any perineal skin or mucosal tissue reaction, generalized
hypersensitivity or anaphylactoid reactions. Parents were
instructed to monitor their children for three days and to
inform their family doctor or paediatrician in the case of
any adverse events (AE).
Telephone survey
Three days after the examination the parents were
contacted for a standardized telephone survey. At first,
parents were asked if they would be willing to repeat
the examination with the same procedure (ceVUS or
CEUS) if necessary. Secondly, they were asked if they
would prefer VCUG to ceVUS for the next time or if they
would prefer CT or MRI to CEUS. Furthermore, the par-
ents were asked about a list of possible AE. For ceVUS
these were skin rash, pruritus, wheals, fever, urinary tract
infection, respiratory problems and for CEUS these were
skin rash, pruritus, wheals, fever, respiratory problems,
taste disorders, vomiting and pain.
Statistical analysis
Descriptive statistics was used to report the results.
Results
CeVUS
All ceVUS examinations were of diagnostic quality
and answered all diagnostic questions. No child needed
further imaging diagnostics such as VCUG following
ceVUS. In the study, the indications for ceVUS were as-
sessment of vesicoureteric reflux (VUR) (fig 1), bladder
rupture or urogenital malformation as recommended in
the guidelines of the ESPR/ESUR [19]. Patient charac-
teristics for both groups are shown in Table I.
Table I. Patient characteristics and study results
  ceVUS (n=56)
Gender (f/m) 32/24
Age (mean) 3.1 year
Age range 3 week – 15.9 year
Weight (mean) ± SD
Dose (mean) of SonoVue® ± SD 0.1 ml
ceVUS, contrast-enhanced voiding urosonography; CEUS contrast-enhanced ultrasonography; SD, standard deviation
Med Ultrason 2022; 24(1): 27-32 29
Fig 1. In a 2-month-old female baby, diagnosed with hydro-
nephrosis during fetal ultrasonography and postnatal positive
urothel sign, the ceVUS showed high grade reflux (IV) with
dilated and elongated ureter
For ceVUS 96 % (54/56) of the parents would repeat
ceVUS if necessary and 4 % (2/56) of the parents refused
ceVUS for further examinations. The main reason report-
ed for rejection was the stress children are put under due
to catheterisation which is needed also in the alternative
examination VCUG. Concerning CEUS examination,
100% (30/30) of parents would agree with performing
CEUS again. ceVUS was favour to VCUG by 92.9%
(52/56) parents. The main reported reason was in 84.6%
(44/52) of cases the lack of radiation exposure. Two par-
ents preferred VCUG to ceVUS because they felt VCUG
is less complicated and less stressful for their child. In
one case, every further examination with catheterization
was refused and in another case the parents did not give
any statement.
AE after ceVUS were reported by the parents in 3.6
% (2/56) of the cases. These events were mild fever and
skin rash, which each occurred once and were both self-
limited.
CEUS
After CEUS the therapy of one child could be finished,
one child needed a change in the therapy and 86.7 %
(26/30) needed a follow-up. For 93.3% (28/30) of the
children CEUS was adequate. Two children needed MRI
and an endoscopic retrograde cholangiopancreatography
CEUS (n=30)
15/15
10.5 year
2 month – 17.7 year
40.8 ± 24.8 kg
1.4 ± 0.9 ml
30 Josefina Seelbach et al CeVUS and CEUS in children and adolescents – safety and parents´ acceptance
(ERCP) for further imaging diagnostics, because the im-
aging of the questioned structure was not high quality.
One structure was a case with necrotizing areas in the
pancreas. Within the CEUS examination it was possible
to detect a homogenous distribution of the microbubbles
in the organ though the imaging of the entire pancreas
was superimposed by air. Consequently, necrotizing
pancreatitis could not be excluded by CEUS, and MRI
was indicated for further strategies by the clinicians. The
other case was a tumour within the main biliary duct
which was suspected in an outside MRI, performed with-
out diffusion weighted imaging and contrast application.
During the CEUS examination, no suspect tumour could
be identified and ERCP was necessary which excluded
intraluminal pathology.
Indication for intravenous CEUS were trauma in
13.3% (4/30), inflammation in 30% (9/30) and focal le-
sions in 56.7% (17/30) (fig 2) of the liver (n=10), spleen
(n=4), kidney (n=6) and pancreas (n=4) as recommended
in the EFSUMB guidelines and position statements [1].
CT as an alternative to CEUS was refused by 83.3
% (25/30) of the parents; 32% (8/25) of the parents an-
swered that the reason for refusal was the radiation ex-
posure. No one favoured CT over CEUS and 17% (5/30)
did not give any statement. MRI as an alternative to
CEUS was refused in 73.3% (22/30) of the cases; 45.5%
(10/22) of the parents answered that CEUS was a more
comfortable examination procedure than that of the MRI.
Four parents preferred MRI and four parents did not give
any statement. In these cases, no causes were reported.
After CEUS, a 13-year-old boy had a single event of
vomiting 30 min after the examination (Table II). AE after
CEUS occurred in 3.3% (1/30) of the cases.
Discussion
The majority of parents in this study agreed with the
use of ceVUS and CEUS examination the next time if re-
quired. In addition, the parents preferred ceVUS or CEUS
to VCUG, MRI or CT. One reason for that is presumably
the comfortable imaging modality and the reduction of
radiation exposure. Furthermore, the rate of observed AE
was low with 3.6 % in the intravesical group and just
one child in the intravenous group. AE were minor or
Table II. Details of adverse events
Adverse events Examination Total (n) Age
Mild fever ceVUS 1 14 months
Skin rash ceVUS 1 3 months
Vomiting CEUS 1 13 years
ceVUS, contrast-enhanced voiding urosonography; CEUS, contrast-enhanced ultrasonography; f, female; m, male
Fig 2. Hemangioma in a 4-month-old female preterm baby
with native sonographic hypoechogenic lesion in the liver. Af-
ter 0.3 ml of SonoVue® i.v. the lesion showed typical features
of hemangioma with early peripheral nodular enhancement (a),
followed by centripetal fill in (b). In the portalvenous phase the
surrounding liver tissue shows hyperperfusion in comparison to
liver tissue besides, 20 sec after contrast application the whole
liver tissue was isoechogen (c). There was no wash out during
5 min of investigation.
moderate and no severe AE occurred. In the present study
the most frequent indications for CEUS are oncological,
traumatic or inflammatory questions. All children who
had received a ceVUS did not need further diagnostic
imaging (VCUG) and were therefore less exposed to ion-
izing radiation. Only in two cases a CEUS was followed
by a diagnostic contrast enhanced MRI examination and
an ERCP. All other children could be treated or followed
up in a more gentle and radiation-free way.
To the best of our knowledge, there is no study evalu-
ating the acceptance of parents after a CEUS examination
of their children and there is only one study evaluating
the acceptance of parents after ceVUS. This survey re-
vealed that 96.2% of parents favour further radiation-free
ceVUS examinations [6]. The acceptance in our study
is comparably high (96%). All parents would do CEUS
again, 83.3% refused the CT as alternative and 73.3%
refused the MRI.
CeVUS has an equal or higher diagnostic sensitiv-
ity and specificity for VUR in comparison to VCUG
Gender Dose of SonoVue® Severity Onset
f 0.1 ml minor subacute
f 0.1 ml minor subacute
m 1.3 ml minor subacute

[13,20,21]. In addition, CEUS has a sensitivity and speci-
ficity of up to 100% in low energy abdominal trauma in
children [8]. In focal liver lesions, CEUS has a specificity
of 98% and the interpretations correspond with the refer-
ence imaging in 85.3% of the cases [22]. Another study
showed that CEUS has a diagnostic detection rate of 93%
for the characterization of liver lesions and portal vein
anomalies [7].
To date, there are only few studies concerning the rate
of AE in intravenous use of SonoVue® in children. There
are some reports about complications during ceVUS. Our
study showed AE after ceVUS in 3.6% of the children.
Equally, a study with 1010 children showed minor AE
in the intravesical use of SonoVue® in 3.7% of the cases
and, as in our study, no severe AE occurred [14]. Further
studies using ceVUS as an alternative method to VCUG
showed no or few minor‑to‑moderate AE [6,13,21]. All
the events were not caused by the contrast media itself
but by the catheterization procedure.
In our study, only in one case self-limited vomiting
was reported as a possible side effect after intravenous
use of SonoVue®. A study of 40 paediatric patients and
young adults showed comparable results: in 2.5% (2/79)
applications AE occurred [23]. There are further studies
which indicate that the intravenous use of SonoVue® in
children has a low rate of AE [7,8,12,22,24]. There are
only two cases of serious AE mentioned in the literature
[15,16]. Our study showed a good safety profile of So-
noVue® in children in intravesical and intravenous use
in line with the existing literature. There are no reports
about a possible correlation between the volume of Sono-
Vue® and the observed rate of AE [15]. The recommend-
ed dose for SonoVue® for intravesical administration is
0.2 - 1.0% of the actual bladder filling [19]. Since there
are no dose finding studies for the appropriate intrave-
nous use of SonoVue® in children, we use 0.1 ml/year of
life. Furthermore, a weight adapted dose of SonoVue® is
often used (e.g. for infants 0.08 - 0.1 ml/kg body weight)
[18]. The FDA recommended a dosage of 0.03 ml/kg of
body weight up to a maximum of 2.4 ml per injection
followed by intravenous flush of 0.9% sodium chloride
injection. Repeated application during a single examina-
tion is possible.
The advantages of the ceVUS and CEUS examina-
tion are equal or higher sensitivity and specificity in com-
parison to the standard routine imaging modalities, low
level of AE, reduction of radiation and the acceptance
of the parents for these ultrasound modalities. Further-
more, SonoVue® is not nephrotoxic, has no interaction
with the thyroid gland, is not accumulating in the blood
and is eliminated by the lung after intravenous applica-
tion [10,25,26]. The main disadvantage of CEUS is the
Med Ultrason 2022; 24(1): 27-32 31
off-label use in children in most indications. AE such
as hypersensitivity, headache, paraesthesia, dizziness,
dysgeusia, blurred vision, flushing, hypotension, nausea,
abdominal pain, rash, pruritus, back pain, chest discom-
fort, injection side reaction, feeling hot, chest pain, pain
and fatigue are reported in adults; in children the list of
AE is shorter (e.g. headache, disturbance of taste), most
of them mild and temporary [11]. Considering all advan-
tages, disadvantages and the acceptance by the parents in
comparison to methods which need ionizing radiation or
sedation one can see that there is a need for the approval
of the intravenous use of SonoVue® in children.
The main limitation of the present study is the small
sample size. That is why it is important to validate the
obtained findings in a greater multicentre trial. Further-
more, the participation in the survey was optional. In this
survey 44.6% of the parents in the ceVUS and 55.5%
of the parents in the CEUS group refused participation.
There is a likelihood of selection bias in the respond-
ents. Parents supporting ceVUS or CEUS have possibly
a higher propensity to respond. To underline the role of
ceVUS and CEUS as an imaging modality for therapeu-
tic decisions, a survey of the opinions of paediatricians
would be helpful.
Conclusion
This prospective study demonstrates that ceVUS and
CEUS are imaging modalities that are supported and pre-
ferred by parents. The survey underlines that AE are rare
and mild. The parents’ acceptance is the consequence of
the UCA safety and the diagnostic accuracy of ceVUS
and CEUS in children.
Conflict of interest: none
References
1. Sidhu PS, Cantisani V, Deganello A, et al. Role of Con-
trast-Enhanced Ultrasound (CEUS) in Paediatric Prac-
tice: An EFSUMB Position Statement. Ultraschall Med
2017;38:33-43.
2. Coleman JL, Navid F, Furman WL, McCarville MB.
Safety of ultrasound contrast agents in the pediatric onco-
logic population: a single-institution experience. AJR Am J
Roentgenol 2014;202:966-970.
3. Riccabona M, Avni FE, Damasio MB, et al. ESPR Urora-
diology Task Force and ESUR Paediatric Working Group
– Imaging recommendations in paediatric uroradiology,
part V: childhood cystic kidney disease, childhood renal
transplantation and contrast-enhanced ultrasonography in
children. Pediatr Radiol 2012;42:1275-1283.
4. Darge K, Ghods S, Zieger B, Rohrschneider W, Troeger J.
Reduction in voiding cystourethrographies after the intro-
32 Josefina Seelbach et al CeVUS and CEUS in children and adolescents – safety and parents´ acceptance
duction of contrast enhanced sonographic reflux diagnosis.
Pediatr Radiol 2001;31:790-795.
5. Darge K. Voiding urosonography with US contrast agents
for the diagnosis of vesicoureteric reflux in children. II.
Comparison with radiological examinations. Pediatr Radiol
2008;38:54-63.
6. Sauer A, Wirth C, Platzer I, et al. Off-label-use of sul-
fur-hexafluoride in voiding urosonography for diagnosis
of vesicoureteral reflux in children: A survey on adverse
events. World J Clin Pediatr 2017;6:52-59.
7. Pschierer K, Grothues D, Rennert J, et al. Evaluation of
the diagnostic accuracy of CEUS in children with benign
and malignant liver lesions and portal vein anomalies. Clin
Hemorheol Microcirc 2015;61:333-345.
8. Menichini G, Sessa B, Trinci M, Galluzzo M, Miele V.
Accuracy of contrast-enhanced ultrasound (CEUS) in the
identification and characterization of traumatic solid organ
lesions in children: a retrospective comparison with base-
line US and CE-MDCT. Radiol Med 2015;120:989-1001.
9. Riccabona M. Pediatric Urogenital Radiology. Springer, 2018.
10. Schneider M. SonoVue, a new ultrasound contrast agent.
Eur Radiol 1999;9 Suppl 3:S347-5348.
11. Rosado E, Riccabona M. Off-Label Use of Ultrasound
Contrast Agents for Intravenous Applications in Children:
Analysis of the Existing Literature. J Ultrasound Med
2016;35:487-496.
12. Riccabona M. Application of a second-generation US con-
trast agent in infants and children--a European question-
naire-based survey. Pediatr Radiol 2012;42:1471-1480.
13. Kis E, Nyitrai A, Varkonyi I, et al. Voiding urosonography
with second-generation contrast agent versus voiding cys-
tourethrography. Pediatr Nephrol 2010;25:2289-2293.
14. Papadopoulou F, Ntoulia A, Siomou E, Darge K. Con-
trast-enhanced voiding urosonography with intravesical
administration of a second-generation ultrasound contrast
agent for diagnosis of vesicoureteral reflux: prospective
evaluation of contrast safety in 1,010 children. Pediatr Ra-
diol 2014;44:719-728.
15. Piskunowicz M, Kosiak W, Batko T, Piankowski A, Pol-
czynska K, Adamkiewicz-Drozynska E. Safety of intrave-
nous application of second-generation ultrasound contrast
agent in children: prospective analysis. Ultrasound Med
Biol 2015;41:1095-1099.
16. Stenzel M, Mentzel HJ. Ultrasound elastography and con-
trast-enhanced ultrasound in infants, children and adoles-
cents. Eur J Radiol 2014;83:1560-1569.
17. Damasio MB, Ording Muller LS, Augdal TA, et al. Europe-
an Society of Paediatric Radiology abdominal imaging task
force: recommendations for contrast-enhanced ultrasound
and diffusion-weighted imaging in focal renal lesions in
children. Pediatr Radiol 2020;50:297-304.
18. Riccabona M, Lobo ML, Augdal TA, et al. European So-
ciety of Paediatric Radiology Abdominal Imaging Task
Force recommendations in paediatric uroradiology, part
X: how to perform paediatric gastrointestinal ultrasonog-
raphy, use gadolinium as a contrast agent in children, fol-
low up paediatric testicular microlithiasis, and an update
on paediatric contrast-enhanced ultrasound. Pediatr Radiol
2018;48:1528-1536.
19. Riccabona M, Vivier PH, Ntoulia A, et al. ESPR urora-
diology task force imaging recommendations in paediatric
uroradiology, part VII: standardised terminology, impact
of existing recommendations, and update on contrast-en-
hanced ultrasound of the paediatric urogenital tract. Pediatr
Radiol 2014;44:1478-1484.
20. Kljucevsek D, Battelino N, Tomazic M, Kersnik Levart T.
A comparison of echo-enhanced voiding urosonography
with X-ray voiding cystourethrography in the first year of
life. Acta Paediatr 2012;101:e235-e239.
21. Papadopoulou F, Anthopoulou A, Siomou E, Efremidis S,
Tsamboulas C, Darge K. Harmonic voiding urosonography
with a second-generation contrast agent for the diagno-
sis of vesicoureteral reflux. Pediatr Radiol 2009;39:239-
244.
22. Jacob J, Deganello A, Sellars ME, Hadzic N, Sidhu PS.
Contrast enhanced ultrasound (CEUS) characterization of
grey-scale sonographic indeterminate focal liver lesions in
pediatric practice. Ultraschall Med 2013;34:529-540.
23. Knieling F, Strobel D, Rompel O, et al. Spectrum, Applica-
bility and Diagnostic Capacity of Contrast-Enhanced Ultra-
sound in Pediatric Patients and Young Adults after Intrave-
nous Application - A Retrospective Trial. Ultraschall Med
2016;37:619-626.
24. Valentino M, Serra C, Pavlica P, et al. Blunt abdominal
trauma: diagnostic performance of contrast-enhanced US in
children--initial experience. Radiology 2008;246:903-909.
25. ter Haar G. Safety and bio-effects of ultrasound contrast
agents. Med Biol Eng Comput 2009;47:893-900.
26. Piscaglia F, Nolsoe C, Dietrich CF, et al. The EFSUMB
Guidelines and Recommendations on the Clinical Practice
of Contrast Enhanced Ultrasound (CEUS): update 2011 on
non-hepatic applications. Ultraschall Med 2012;33:33-59.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 33-37
DOI: 10.11152/mu-3206

Multiparametric ultrasound in torsion of the testicular appendages:

a reliable diagnostic tool?
Gregor Laimer1, Raphael Müller2, Christian Radmayr2, Andrea Katharina Lindner2, Andrei
Lebovici3, Friedrich Aigner1

1Department of Radiology, Medical University Innsbruck, Innsbruck, Austria, 2Department of Urology, Medical
University Innsbruck, Innsbruck, Austria, 3Department of Radiology, County Emergency Hospital Cluj-Napoca,
Cluj-Napoca, Romania

Abstract
Aim: Torsion of the testicular appendages represents the most common cause of an acute scrotum in prepubertal boys. Its
sonographic appearances on gray-scale US and color Doppler US have already been presented in several studies. The aim of
this analysis was to expand those already established techniques with strain elastography and thus present typical features of
this entity on multiparametric US. Material and methods: Retrospective analysis of all patients presented to the urological
department with an acute scrotum between January 2018 and July 2020 identified eleven patients 6-17 years old (mean, 11.1
years), discharged with the diagnosis torsion of the testicular appendages that were examined with a high-end ultrasound de-
vice. Results: On gray-scale US all patients showed a round lesion with heterogenous echotexture adjacent to the upper pole
of the testis/epididymis with a diameter of 4 to 11.1 mm (mean, 7.7 mm). Scrotal skin thickening and a concomitant hydrocele
were found in 9 (81.8%) and 7 (63.6%) cases, respectively. On color Doppler images, all torsed appendages were avascular
and in 9 (81.8%) patients we observed hyperemia of the adjacent epididymis. Strain elastography showed increased tissue
stiffness in all documented images. Conclusion: Torsion of the testicular appendages has a set of features on multiparametric
US. Awareness of this features can facilitate diagnosis of torsion of the testicular appendages and reduce unnecessary surgical
scrotal exploration or unwarranted antibiotic treatment.
Keywords: torsion; testicle; appendix; color Doppler ultrasonography; elastography

Introduction (Wolffian) duct. The appendix testis is found more fre-

Torsion of the testicular appendages represents
the most common underlying cause of an acute scro-
tum in prepubertal boys, accounting for 40-60% of the
cases [1,2]. The testicular and epididymal appendages,
found at the upper pole of the testis and at the head of
the epididymis respectively, are remnants of the degen-
erating paramesonephric (Müllerian) and mesonephric
Received 15.04.2021  Accepted 21.07.2021
Med Ultrason
2022, Vol. 24, No 1, 33-37
Corresponding author: Dr. Gregor Laimer
Department of Radiology
Medical University Innsbruck
Anichstraße 35 Innsbruck 6020, Austria
E-mail: gregor.laimer@i-med.ac.at
quently than the appendix epididymis in 83.3-100% (vs.
20-24%) of pediatric testicles [3]. Both appendages are
pedunculated and therefore prone to torsion [4]. The ac-
tual cause of torsion remains unknown, but may be re-
lated to trauma or prepubertal enlargement [5].
Nevertheless, torsion of the testicular appendages is
an important differential diagnosis of testicular torsion
and based on history and physical examination alone [6],
it can be very difficult to distinguish because of a simi-
lar clinical presentation. In contrast to testicular torsion,
which is a surgical emergency and the affected testicle
should be detorsed within hours after onset of symptoms,
torsion of the testicular appendages can be treated con-
servatively. Surgery is rarely indicated and limited to cas-
es with severe, by analgesics uncontrollable prolonged or
recurrent pain [7].
34 Gregor Laimer et al Multiparametric ultrasound in torsion of the testicular appendages: a reliable diagnostic tool?
Torsion of the testicular appendages is often over-
looked by the clinician, leading to unnecessary surgical
scrotal exploration because of erroneously suspected
testicular torsion, thus increasing operation-related com-
plications and causing additional costs for the hospital.
Ultrasound (US) represents a widely available and rec-
ommended diagnostic tool for the evaluation of acute
scrotal pain in all age groups [7]. Sonographic appear-
ances on gray-scale US and color Doppler US of torsion
of the testicular appendages have already been presented
in several studies [8-13].
The aim of this retrospective analysis was to expand
those already established techniques with strain elastog-
raphy and thus present typical appearances of torsion of
the testicular appendages on multiparametric US (gray-
scale US, color Doppler US and strain elastography) to
be able to reliably confirm the diagnosis of torsion of the
testicular appendages in order to prevent unnecessary
surgical scrotal exploration.
Material and methods
Patients
This study was approved by the Institutional Review
Board of the Medical University Innsbruck. All patients’
data included in this retrospective study were handled ac-
cording to the norms of the Declaration of Helsinki and
its amendments.
Retrospective analysis of all patients presented to the
urological department with an acute scrotum between
January 2018 and July 2020 identified 32 patients dis-
charged with the diagnosis of torsion of testicular ap-
pendages. Of these, eleven patients 6-17 years old (mean,
11.1 years) were examined with a high-end ultrasound
device at the radiological department of our institution.
US evaluation
At our institution, US examinations in patients with
acute scrotal pain are performed by the Department of
Radiology during daytime working hours and by the
urologist on duty himself during off hours. All examina-
tions of our study cohort were performed at the Depart-
ment of Radiology using a high-end ultrasound device
and conducted or supervised by a radiologist (A.F.) with
more than 10 years of experience in scrotal ultrasound.
Standardized presets were used for gray-scale US
examination. Color Doppler US examination was per-
formed with the highest signal gain setting possible with-
out the appearance of background noise to maximize sen-
sitivity to slow flow velocities. Strain elastography was
performed with repeated compression and decompres-
sion of the testis, with the pressure applied to the testicles
adjusted according to the visual indicator for compres-
sion seen on the ultrasound device. Tissue elasticity was
calculated in real time and displayed as a color-coded
overlay on the B-mode. Red was encoded as soft, green
as intermediate and blue as hard tissue stiffness.
The standard protocol of pediatric scrotal US ex-
amination at our institution includes: 1) transverse
gray-scale and Doppler evaluation of both testicles for
initial comparison; 2) gray-scale scans of each testicle
and epididymis in both transverse and sagittal planes;
3) color Doppler scans of all structures; 4) derivation of
Doppler shift for arterial and venous intratesticular blood
flow; and 5) strain elastography of suspect findings.
All multiparametric US examinations were per-
formed using a Logic E9 unit with a linear probe (ML,
6–15 MHz; GE Healthcare) or a HI Vision Ascen-
dus unit with a linear probe (EUP-L74M, 5–13 MHz;
Hitachi).
Torsion of testicular appendages
on multiparametric US
Retrospective viewing of the patients’ images was
performed by two radiologists (A.F., L.G.) and evaluated
in consensus. Based on previous publications [8-13] and
our experience, we defined the following findings as the
typical appearance of torsion of the testicular appendages
on multiparametric US: Gray-scale US - 1) heterogenous
round lesion with close proximity to the upper pole of
the testis/epididymis (fig 1a,b), 2) scrotal skin thicken-
ing (fig 1c) and 3) concomitant hydrocele (fig 1d); Color
Doppler US - 1) avascularity of the round lesion found
on gray-scale US (fig 2a,b) and 3) reactive hyperemia
of the associated epididymis (fig 2c); Strain Elastogra-
phy - increased tissue stiffness (encoded as blue on strain
elastography; fig 3b,d) of the round lesion found on gray
scale US (fig 3a,c).
Statistical analysis
Descriptive statistics were performed using SPSS
Version 22 (SPSS Inc., Chicago, Illinois). Data are ex-
pressed as total numbers, mean and range.
Results
Results are shown in Table I. In all eleven cases of
our study cohort the torsed appendix was adjacent or in
close proximity to the upper pole of the testis and to the
epididymis. A differentiation between testicular/epididy-
mal appendix was not possible. Seven torsions (63.3%)
occurred on the right side. A heterogenous round lesion
was observed in all cases and had a predominantly hyper-
echogenic texture in 9 (81.8%), and a predominantly hy-
poechogenic texture in 2 (18.2%) cases. These round le-
sions had a maximal diameter ranging from 4 to 11.1 mm
(mean, 7.7 mm). Scrotal skin thin thickening was found
 Med Ultrason 2022; 24(1): 33-37 35

Fig 1. Gray-scale US showing heterogenous round lesion with close proximity to the upper pole of the testis/epididymis (a, b); scro-
tal skin thickening (c); and concomitant hydrocele (d).

in 9 (81.8%) and a concomitant hydrocele in 7 (63.6%)

cases. On color Doppler images all torsed appendages
showed total avascularity and in 9 (81.8%) patients we
observed hyperemia of the adjacent epididymis. Strain
elastography was documented in 7 (63.6%) patients dem-
onstrating increased tissue stiffness of the round lesion in
all images. Findings were limited to the affected side of
the testicle in all patients.
In the initial radiological report, torsion of the tes-
ticular appendages was diagnosed in 9 (81.8%) cases
of our study cohort. Of those, six patients were treated
conservatively and three had undergone surgical scrotal Fig 2. Color Doppler US with avascularity of the round lesion
exploration due to prolonged pain. In two (18.2%) cases, (a, b) and reactive hyperemia of the associated epididymis (c).
torsion of the testicular appendages was suspected in the
radiological report, but final diagnosis was made only af-
ter scrotal exploration.

Table I. Multiparametric US findings of torsion of the testicular

appendages in eleven patients.
Finding n (%)
Gray-scale US
Round lesion with heterogenous echotexture 11 (100)
- predominantly hyperechogenic 9 (81.8)
- predominantly hypoechogenic 2 (18.2)
Diameter of round lesion in mm, mean (range) 7.7 (4 - 11.1)
Scrotal skin thickening 9 (81.8)
Concomitant hydrocele 7 (63.6)
Color Doppler US
Avascularity of torsed appendix 11 (100)
Hyperemia of associated epididymis 9 (81.8)
Strain elastography
Increased tissue stiffness 7 (63.6)
No documentation 4 (36.4) Fig 3. Strain elastography with increased tissue stiffness (en-
US: Ultrasound coded as blue) of the round lesion.
36 Gregor Laimer et al Multiparametric ultrasound in torsion of the testicular appendages: a reliable diagnostic tool?
Discussion
This retrospective analysis confirms US as a valid
diagnostic tool to verify the diagnosis of torsion of the
testicular appendages.
The normal testicular appendages are oval or pe-
dunculated in shape and up to 7-8 mm in length, with
the appendix epididymis being slightly larger than the
appendix testis [12,14,15]. Several studies confirm the
torsed testicular appendages as round, enlarged ex-
tratesticular structures with heterogenous echotexture.
Hesser et al [11] reported diameters ranging from 3 to
17 mm. Two other studies [8,12] compared the size of
normal vs. torsed testicular appendages and described
thresholds of 5 and 5.6 mm, respectively as suggestive
for torsion of the testicular appendages. In our analysis
the mean size of the torsed testicular appendage was 7.7
mm and only one had a diameter less than 5 mm. There-
fore, the described thresholds seem to be a viable indica-
tor to suggest torsion of the testicular appendages. The
echogenicity of the torsed appendix has been described
as heterogenous, low and increased [11-13]. One study
[10] reports, that the echogenicity of the torsed appen-
dix changes according to the time of onset, associating
a hyperechoic texture with later sonography (>24 hours
after onset of symptoms). In our study almost all torsed
appendages were hyperechoic (81.8%). Thus, we cannot
confirm this thesis as it does not seem to apply to our
cases, since most of the patients received a sonographic
evaluation of the scrotum very promptly after onset of
symptoms. Overall, we agree with Yang et al [12] and
claim that the echogenicity of the testicular appendages
should not be used as a viable indicator in differentiating
between torsed and normal testicular appendages. It is
reported that torsion of the testicular appendages is often
accompanied by scrotal skin thickening and a concomi-
tant hydrocele [9,11,14]. This is in line with our findings,
as we observed it in 81.8% and in 63.6% cases of our
study cohort, respectively.
On color Doppler US, the normal and the torsed ap-
pendix have no blood flow [12]. In fact, all round le-
sions of our study cohort were avascular. One study [9]
described the avascular appendages as “an island in a
sea of color”, due to the reactive hyperemia of the sur-
rounding structure. We can confirm this finding, as we
observed reactive hyperemia of the adjacent epididymis
and surrounding structures in 81.8% of our cases. Our
study was the first to assess findings of torsion of the
testicular appendages on strain elastography. All docu-
mented images showed increased tissue stiffness of the
round lesion found on gray-scale US. Therefore, strain
elastography can be used to reliably confirm the diagno-
sis. The increased tissue stiffness can be explained by the
swelling of the torsed appendix and the consequent sta-
sis. We strongly believe that strain elastography should
be given more attention in the management of cases with
acute scrotum as it seems to be an additional, reliable US
technique for the diagnosis of torsion of the testicular ap-
pendages. Thus, further prospective studies are required
to establish multiparametric US as a valid tool for the
assessment of an acute scrotum.
Major limitations of our study lie in its retrospec-
tive design and the single center bias. Moreover, the
number of patients included was relatively small. The
small number of cases can be attributed to the fact that
many patients with an acute scrotum received initial ex-
amination during off hours at our institution. These US
examinations could not be evaluated as they were per-
formed by the urologist on duty not using a high-end
ultrasound device and with no standard documentation
protocol. This underlines the importance of standardized
documentation, the need of a high-end ultrasound device
and a careful device introduction, especially for young
trainees in order to guarantee a reliable diagnostic evalu-
ation. Another limitation we encountered is the inability
to differentiate between torsion of the appendix testis and
torsion of the appendix epididymis. However, since the
treatment of both entities is the same, a differentiation
has no clinical impact.
In conclusion, our review revealed a set of features on
multiparametric US, which can help to reliably diagnose
torsion of the testicular appendages. These features in-
clude 1) a round, extratesticular lesion with heterogenous
echogenicity 2) scrotal skin thickening and concomitant
hydrocele 3) an avascular torsed appendix with surround-
ing hyperemia and 4) increased tissue stiffness of the
torsed appendix on strain elastography. Overall, we think
awareness of the multiparametric US features presented
in this study can facilitate diagnosis of torsion of the
testicular appendages and reduce unnecessary surgical
scrotal exploration or unwarranted antibiotic treatment.
Conflict of interest: none
References
1. Mäkelä E, Lahdes-Vasama T, Rajakorpi H, Wikström S. A
19-year review of paediatric patients with acute scrotum.
Scand J Surg 2007;96:62-66.
2. Kim JS, Shin YS, Park JK. Clinical features of acute
scrotum in childhood and adolescence: Based on 17years
experiences in primary care clinic. Am J Emerg Med
2018;36:1302-1303.
3. Sahni D, Jit I, Joshi K, Sanjeev. Incidence and structure
of the appendices of the testis and epididymis. J Anat
1996;189:341-348.

4. Jones P. Torsion of the testis and its appendages
during childhood. Arch Dis Child 1962;37:214-
226.
5. Barthold JS. Abnormalities of the Testis and Scrotum and
Their Surgical Management. Campbell-Walsh Urology.
10th ed. Philadelphia. 2012:3557-3596.e13.
6. Varga J, Zivkovic D, Grebeldinger S, Somer D. Acute
scrotal pain in children--ten years’ experience. Urol Int
2007;78:73-77.
7. Pomajzl AJ, Leslie SW. Appendix Testes Torsion. In:
StatPearls. StatPearls Publishing, Treasure Island (FL) ;
2020.
8. Baldisserotto M, de Souza JC, Pertence AP, Dora MD.
Color Doppler sonography of normal and torsed tes-
ticular appendages in children. AJR Am J Roentgenol
2005;184:1287-1292.
9. Lev M, Ramon J, Mor Y, Jacobson JM, Soudack M. So-
nographic appearances of torsion of the appendix testis
and appendix epididymis in children. J Clin Ultrasound
2015;43:485-489.
Med Ultrason 2022; 24(1): 33-37 37
10. Park SJ, Kim HL, Yi BH. Sonography of intrascrotal ap-
pendage torsion: varying echogenicity of the torsed ap-
pendage according to the time from onset. J Ultrasound
Med 2011;30:1391-1396.
11. Hesser U, Rosenborg M, Gierup J, Karpe B, Nyström A,
Hedenborg L. Gray-scale sonography in torsion of the tes-
ticular appendages. Pediatr Radiol 1993;23:529-532.
12. Yang DM, Lim JW, Kim JE, Kim JH, Cho H. Torsed appen-
dix testis: gray scale and color Doppler sonographic find-
ings compared with normal appendix testis. J Ultrasound
Med 2005;24:87-91.
13. Strauss S, Faingold R, Manor H. Torsion of the testicular
appendages: sonographic appearance. J Ultrasound Med
1997;16:189-192.
14. Sellars ME, Sidhu PS. Ultrasound appearances of the
testicular appendages: pictorial review. Eur Radiol
2003;13:127-135.
15. Johnson KA, Dewbury KC. Ultrasound imaging of the
appendix testis and appendix epididymis. Clin Radiol
1996;51:335-337.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 38-43
DOI: 10.11152/mu-3069

Effects of local anaesthetic dilution on the characteristics of

ultrasound guided axillary brachial plexus block:
a randomised controlled study
Anil Ranganath, Osman Ahmed, Gabriella Iohom

Department of Anaesthesia and Intensive Care Medicine, Cork University Hospital, Cork, Ireland

Abstract
Aims: Ultrasound guidance has led to marked improvement in the success rate and characteristics of peripheral nerve
blocks. However, effects of varying the volume or concentration of a fixed local anaesthetic dose on nerve block remains un-
clear. The purpose of our study was to evaluate whether at a fixed dose of lidocaine, altering the volume and concentration will
have any effect on the onset time of ultrasound-guided axillary brachial plexus block. Material and methods: Twenty patients
were randomised to receive an ultrasound-guided axillary brachial plexus block with either lidocaine 2% with epinephrine (20
ml, Group 2%) or lidocaine 1% with epinephrine (40 ml, Group 1%). The primary endpoint was block onset time. Secondary
outcomes included duration of the block, performance time, number of needle passes, incidence of paraesthesia and vascular
puncture. Results: The median [IQR] onset time of surgical anaesthesia was shorter in Group 1% when compared to Group
2% (6.25 [5-7.5] min vs 8.75 [7.5-10] min; p=0.03). The mean (SD) overall duration of surgical anaesthesia was significantly
shorter in Group 1% compared to Group 2% (150.9±17.2 min vs 165.1±5.9 min; p=0.02). Group 1% had a shorter performance
time with fewer needle passes. The incidence of vascular puncture and paraesthesia was similar in the two groups. Conclu-
sion: Ultrasound-guided axillary brachial plexus blocks performed using a higher volume of lower concentration lidocaine
was associated with shorter onset time and duration of surgical anaesthesia.
Keywords: brachial plexus block; axillary; ultrasound; local anaesthetic; lidocaine

Introduction tration can affect the characteristics of the nerve block.

Ultrasound-guided axillary brachial plexus block
(USgABPB) is an effective and reliable technique for the
provision of surgical anaesthesia for forearm and hand
surgeries [1-4]. Previously, numerous studies have com-
pared efficacy of brachial plexus block using different lo-
cal anaesthetic solutions of varying concentrations and
volumes [5-7]. However, only few studies have shown
that, at constant dose, altering the volume or concen-
Received 02.02.2021  Accepted 05.06.2021
Med Ultrason
2022, Vol. 24, No 1, 38-43
Corresponding author: Dr Gabriella Iohom PhD
Department of Anaesthesia and Intensive Care
Medicine, Cork University Hospital,
Wilton Road, Cork, Ireland
Phone: +353214922135
Fax: +353214643454
E-mail: giohom@ucc.ie
Historically, it has been reported that higher concentra-
tion/lower volume yielded a shorter onset time when
compared to higher volume/lower concentration solu-
tion using a single injection nerve stimulation technique
for sciatic nerve block [8,9]. In contrast, in perivascular
axillary blocks with a fixed dose of local anaesthetic,
larger volumes provided a better quality sensory [10] and
quicker onset motor block [11] when compared to lower
volumes. The results from these studies were inconsist-
ent with respect to onset time, success rate and duration
of the block. In addition, it is unknown whether they can
be replicated with ultrasound guidance. We have chosen
to focus on the USgABPB with lidocaine plus epineph-
rine which is offered preferentially to ambulatory upper
limb trauma patients at our institution [12].
In this prospective, randomised, double-blind study,
we examined whether two different volumes and con-
centrations of a fixed dose of lidocaine with epinephrine

influenced the characteristics of USgABPB. We hypothe-
sised that 40 mL of lidocaine 1% with epinephrine would
result in a shorter onset time when compared to 20 mL of
lidocaine 2% with epinephrine.
Material and methods
This single centre study was approved by the Clini-
cal Research Ethics Committee of Cork Teaching Hospi-
tals, Cork, Ireland [ECM 4(mm) 01/07/14; 01 July 2014,
Chairperson Professor Michael G Molloy], registered at
https://clinicaltrials.gov (NCT03207035), and carried
out at Cork University Hospital. Having obtained written
informed consent from each, patients aged 18 years or
older, ASA grade I-III scheduled to undergo minor uni-
lateral upper limb trauma surgery of the hand or fore-
arm, were enrolled in the study. Exclusion criteria were
contraindication to regional anaesthesia, hypersensitivity
to amide local anaesthetics, intolerance, or contraindica-
tion to non-steroidal anti-inflammatory drugs, BMI >35,
pregnancy, cardiac conduction abnormalities, history of
hepatic and renal impairment, chronic pain, neuromuscu-
lar disease, and psychiatric disorder.
Patients were randomised using computer-generat-
ed sequence of random numbers and sealed envelope
technique, prepared by an investigator with no clinical
involvement in the trial. They were subsequently allo-
cated to receive USgABPB with either 20 mL lidocaine
2% with 1:200,000 epinephrine (Group 2%) or 40 mL
lidocaine 1% with 1:400,000 epinephrine (Group 1%)
(diluted up to the study volume with 0.9% saline). Intra-
venous access was established in the contralateral upper
limb and standard monitoring was employed through-
out the procedure. The operative arm was abducted and
externally rotated with the elbow flexed at 90˚. Under
aseptic precautions the axillary brachial plexus block
was performed under ultrasound guidance alone using
a SonoSite Titan unit (SonoSite®, Bothwell, WA) with
a 38 mm linear array 5–10 MHz transducer (L38). Fol-
lowing the identification of the median, ulnar, radial, and
Table I. Motor and Sensory Testing
Motor test
Median nerve Flexion of radial 3 fingers
Radial nerve Extension of wrist
Ulnar nerve Abduction of fingers
Musculocutaneous nerve Elbow flexion
Modified Bromage scale [4].
4 Full strength in relevant muscle.
3 Reduced strength but ability to move muscle against resistance
2 Ability to move relevant muscle group against gravity but not against resistance
1 Flicker of movement in relevant muscle group
0 No movement in relevant muscle group
Med Ultrason 2022; 24(1): 38-43 39
musculocutaneous nerves in the axillary region, a 50
mm 24-gauge insulated short bevel needle (Stimuplex®
B. Braun, Melsungen, Germany) was advanced in-plane
towards each nerve with the aim of surrounding it with
either 5 mL (Group 2%) or 10 mL (Group 1%) of local
anaesthetic solution. Dynamic manipulation of the nee-
dle was sought to facilitate the circumferential perineural
spread of local anaesthetic. All blocks were performed by
an operator experienced in USgABPB.
Block assessment
Upon completion of the block, a blinded observer
not aware of the injectate volume, assessed the onset
of sensory and motor block in the innervation area of
each nerve (median, ulnar, radial, and musculocutane-
ous nerve) every 2.5 mins, until surgical anaesthesia was
achieved or 30 mins have elapsed. Sensory function was
scored as being present or absent and motor function was
graded using the modified Bromage scale (Table I) [4].
Surgical anaesthesia was defined as a motor score ≤2,
with absent sensation to cold (tested with ethyl chloride
BP, Criogesic®, Dr Georg Friedrich Henning, Chemis-
che Fabrik Walldorf GmbH, Walldorf, Germany). Each
nerve distribution area was individually assessed, and
the sensory and motor onset time was measured sepa-
rately from conclusion of the block (removal of block
needle, T0) to attainment of absent sensation to cold and
a motor score <2, respectively. Overall sensory and mo-
tor block onset time was taken from T0 to attainment of
surgical anaesthesia in all innervation territories. The
block was considered a failure if surgical anaesthesia had
not been achieved at 30 mins in one or more of the four
nerve distribution areas. In case of block failure, an ad-
ditional rescue block or conversion to general anaesthe-
sia was planned together with separate analysis of data
from those patients. All patients received paracetamol
1 g and diclofenac sodium 75 mg iv intraoperatively. In
case of patient discomfort or upon request, sedation with
midazolam to a maximum of 3 mg and/or supplemental
analgesia with up to 100 ug fentanyl was provided at the
discretion of the attending anaesthesiologist.
Sensor test
Thenar eminence
Dorsum of hand
Hypothenar eminence
Over base first metacarpal
40 Anil Ranganath et al Effects of local anaesthetic dilution on the characteristics of US guided axillary brachial plexus block
Postoperative analgesia was prescribed around the
clock in the form of paracetamol 1 g po 6 hourly and
diclofenac 75 mg po 12 hourly. Oxycodone 10 mg orally
4-6 hourly was administered as rescue analgesia. Post-
operatively, sensory and motor function of each nerve
was assessed every 15 mins. Sensory and motor dura-
tion was measured separately for each nerve from T0 to
return of sensation to cold and motor power to >3, re-
spectively. Overall sensory and motor block offset was
defined as return of sensation to cold and motor power
(score ≥3) respectively, in any one nerve distribution
area.
The primary outcome was overall surgical anaesthe-
sia onset time, which was defined as the time elapsed
from conclusion of block (T0) until attainment of sur-
gical anaesthesia in all nerves distribution areas. Sec-
ondary outcome measures included overall duration of
sensory and motor block, as well as sensory and motor
onset times and durations of individual blocks. Over-
all duration of surgical anaesthesia was defined as time
elapsed from T0 to return of sensation and motor power
(score ≥3) respectively in any one nerve distribution
area.
Block performance parameters were recorded such as
imaging time (defined as time elapsed from placement of
US probe on the patient to acquisition of a satisfactory
image of the axillary artery and surrounding nerves) and
needling time (defined as the time interval between in-
sertion and removal of block needle). Thus, performance
time was defined as the sum of imaging and needling
times. The number of needle passes were recorded. The
initial needle pass was considered as the first pass and
any subsequent needle advancement preceded by retrac-
tion of 1 cm counted as an additional pass. Incidences of
vascular puncture and paraesthesia were also noted.
Sample size and statistical analysis
In the absence of data from previous studies using
20 ml of lidocaine 2% with epinephrine for ultrasound
guided axillary brachial plexus block, sample size was
calculated based on our pilot study of 10 patients. We
found a mean ± SD onset time of 11.25±2.3 min. The
Table II. Patient characteristics
Group 2% (n=10)
Age, y 46.8±18.2
Sex, M/F, n 7/3
BMI, Kg/m2 25.3±3.2
ASA grade (I/II/III), n 7/3/0
Duration of surgery, min 62±10.8
Site of surgery (wrist/hand), n 5/5
Continuous variables are presented as means ± SD, categorical variables as counts
minimum sample size required to have an 80% prob-
ability of detecting a 30% decrease in onset time (level
of significance 0.05) was 7 patients per group. We re-
cruited 10 patients per group to account for potential
dropouts.
Statistical analysis was performed using SPSS ver-
sion 24 (IBM, Armonk, New York). The Shapiro-Wilk
test was used for normality testing. Continuous, normally
distributed data are presented as mean (SD), and non-
normally distributed data as median (interquartile range
[IQR]). Comparison between groups were analysed us-
ing the unpaired Student’s t test for normally distribut-
ed data and the Mann-Whitney U test for nonparamet-
ric data. Categorical variables were compared between
groups using Pearson’s or Fischer’s exact test. All tests
were two-tailed, and P < 0.05 was considered statistically
significant.
Results
Twenty patients (10 in each group) were recruited to
the study from September 2014 to August 2015. All pa-
tients completed the study (fig 1) and none of the patients
required rescue block, conversion to general anaesthesia
or intraoperative opioid analgesia. There were no adverse
events noted in either group. The patient demographic
characteristics were similar between the groups (Ta-
ble II). Table III details onset times. The median [IQR]
overall onset time of surgical anaesthesia was shorter
in Group 1% compared to Group 2%. The overall onset
time of sensory but not motor block was also shorter in
Group 1%. Onset times of individual nerves were similar
in the two groups, with the exception of median sensory
onset time which was shorter in Group 1%. Table IV de-
picts block durations. The mean (SD) overall duration of
surgical anaesthesia was shorter in Group 1% compared
to Group 2%, reflective of overall motor block duration.
Individual sensory and motor block durations were simi-
lar, with median motor block duration shorter in Group
1%. Figure 2 shows the primary outcome measure, over-
all onset of surgical anaesthesia.
Group 1% (n=10) p value
48±15.7 0.88
8/2 0.60
24.5±3.6 0.62
4/6/0 0.18
58.5±14.9 0.56
6/4

Group 1% had a shorter needling time, performance
time and fewer needle passes when compared to Group
2%. No difference was found between the groups with
respect to imaging time, incidence of vascular puncture
or paraesthesia (Table V).
Med Ultrason 2022; 24(1): 38-43 41
Discussion
In this single centre randomised controlled trial, we
observed that when using 400 mg of lidocaine with epi-
nephrine, increasing the volume of injectate by dilution

Table III. Sensory and Motor Block Onset Times

20 ml Lidocaine 2% (n=10) 40 mL Lidocaine 1% (n=10) p value
Overall sensory onset 8.75 [5-10] 5 [5-7.5] 0.046
Overall motor onset 6.25 [5-7.5] 5 [2.5-7.5] 0.41
Overall onset of surgical anaesthesia 8.75 [7.5-10] 6.25 [5-7.5] 0.03
Radial
Sensory 5 [5-10] 5 [2.5-7.5] 0.12
Motor 5 [2.5-7.5] 3.75 [2.5-5] 0.55
Ulnar
Sensory 6.25 [5-7.5] 5 [2.5-5] 0.12
Motor 5 [2.5-7.5] 5 [2.5-7.5] 0.87
Median
Sensory 6.25 [5-10] 5 [2.5-5] 0.03
Motor 5 [2.5-7.5] 5 [2.5-5] 0.93
Musculocutaneous
Sensory 5 [5-10] 3.75 [2.5-5] 0.10
Motor 3.75 [2.5-5] 5 [2.5-5] 0.87
Values are Median [IQR] expressed in min

Table IV. Sensory and Motor Block Duration

Group 2% (n=10) Group 1% (n=10) p value
Overall sensory duration 171.6±7.1 158.4±21.7 0.08
Overall motor duration 165.1±5.9 150.9±17.2 0.02
Overall duration of surgical anaesthesia 165.1±5.9 150.9±17.2 0.02
Radial nerve
Sensory 176.1±3.7 167.40±20.4 0.20
Motor 170.1±7.2 158.4±18.6 0.08
Ulnar nerve
Sensory 174.6±3.6 168.9±22.4 0.44
Motor 168.6±8.0 158.4±19.7 0.15
Median nerve
Sensory 173.1±5.9 162.9±19.9 0.14
Motor 168.1±7.4 152.4±15.4 0.01
Musculocutaneous nerve
Sensory 173.1±7.8 161.4±19.9 0.10
Motor 166.6 ± 7.3 152.4 ± 20.4 0.05
Values are Mean±SD, expressed in min

Table V. Block performance data

Group 2% (n=10) Group 1% (n=10) p value
Imaging time, min (A) 2.5 ± 0.5 2.4 ± 0.6 0.70
Needling time, min (B) 8.5 ± 1.2 6.7 ± 0.9 0.002
Performance time, mins (A+B) 10.9 ± 1.3 9.1 ± 0.5 0.001
No. needle passes 10.5 ± 2.3 7.2 ± 1.0 0.001
Vascular puncture, n (%) 2 (20) 1 (10) 0.53
Paraesthesia, n (%) 4 (40) 2 (20) 0.48
Continuous variables are presented as mean ± SD, categorical variables as count/or percentage.
42 Anil Ranganath et al Effects of local anaesthetic dilution on the characteristics of US guided axillary brachial plexus block
Fig 1. CONSORT patient flow diagram. n = number
Fig 2. Overall onset of block. The horizontal black line repre-
sents the median, the box represents the interquartile range, and
the vertical lines show the lowest and highest values. *p = 0.02
resulted in shorter overall onset time and subsequent
shorter duration of ultrasound guided axillary brachial
plexus block. While the shorter onset may be advanta-
geous and desirable in clinical settings with high volume
activity and quick turnover, it appears to come at the ex-
pense of a shorter duration of block which should be both
anticipated and managed appropriately.
In theory, both concentration and volume of the peri-
neural injectate can influence the characteristics of the
nerve block. Higher concentrations may shorten the on-
set time by facilitating the diffusion of local anaesthetic
molecules into the nerve [9], while larger volumes may
influence the block onset time by promoting injectate
spread around neural structures [11]. However, how the
volume/concentration ratio at a fixed local anaesthetic
dose affects the characteristics of a nerve block, remains
unclear. Previous studies yielded inconsistent results
with respect to success rate, onset time and duration of
the block [8-17]. Several factors such as local anaesthetic
volume/concentration ratio, anatomical site of injection
and the nerve locating technique used in the study might
have contributed to the variable results.
For the Labat approach to the sciatic nerve block us-
ing a single injection nerve stimulation technique, Ta-
boada et al observed that 20 mL of mepivacaine 1.5%
(vs 30 mL of mepivacaine 1%) improved the success rate
and shortened the onset time of both sensory and mo-
tor block [9]. The authors speculated that, because of the
size of sciatic nerve and the thickness of epineurium it
would require a large concentration gradient to facilitate
the diffusion of local anaesthetic molecules. In contrast,
Cappelleri et al, using a double injection nerve stimula-
tor technique for sciatic nerve block, found no difference
with respect to success rate, onset time and duration of
the block between 12 mL of mepivacaine 2% and 24 mL
of mepivacaine 1% [15]. They hypothesized that com-
pared to a single injection technique, the double injection
resulted in better distribution of local anaesthetic around
each component of the peripheral nerve and with this,
the effect of local anaesthetic volume/concentration ra-
tio become secondary to the regional nerve localisation
technique.
Similarly, few studies have evaluated the effect of al-
tering the volume and concentration of a fixed local an-
aesthetic dose for the brachial plexus block. Krenn et al
suggested that higher volume of ropivacaine resulted in
faster onset of motor block for a single injection axillary
block, where loose connective tissue surrounds the bra-
chial plexus [11]. In contrast, studies where the axillary
block was performed using the multiple injection nerve
stimulator technique [13] and infraclavicular block using
ultrasound [14], did not show any difference with respect
to block success rate and onset time. In our study, overall
onset of surgical anaesthesia was faster using a higher
volume when compared to a lower volume (identical
dose), and this was mainly reflective of the onset of senso-
ry but not motor component of the block. The difference
in the result could be explained by the technique used
to locate the target nerves. We performed the ultrasound
guided axillary brachial plexus block having identified
all four terminal nerves with the precise endpoint consist-
ing of circumferential perineural spread of local anaes-
thetic, and not using a single or multiple nerve stimula-
tion, or ultrasound guided perivascular approach [3,18].
Interestingly, and perhaps counterintuitively, the in-
jection of the lower volume resulted in a longer block
performance time. This is likely due to the requirement
for a more precise needle tip positioning and subsequent
adjustment in order to achieve circumferential spread
around each of the four terminal nerves while having a
limited injectate volume at disposal.

Our study is limited inter alia by the small sample
size. Although we found differences between groups in
terms of both onset time and duration of block, these re-
sults cannot be generalised to other local anaesthetics,
techniques and peripheral injection sites due to varia-
tion in the anatomical architecture surrounding nerves
[19-22]. It has been demonstrated that, using a multiple
injection technique for a humeral canal block, higher
volume and lower concentration of levobupivacaine im-
proved the sensory block quality and success rate [16].
In contrast, ultrasound guided interscalene block resulted
in faster onset of block using lower volume and higher
concentration of ropivacaine [17].
In conclusion, when compared to 20 mL of lido-
caine 2% with epinephrine, 40 mL of lidocaine 1% with
epinephrine resulted in faster overall onset and shorter
duration of surgical anaesthesia following an ultrasound
guided axillary brachial plexus block. Further studies
are required to determine whether these results can be
extrapolated to other local anaesthetics and anatomical
injection sites.
Acknowledgement: Assistance with this manuscript:
Mr Bahman Honsari provided statistical expertise.
Conflicts of interest: none.
References
1. Sites BD, Beach ML, Spence BC, et al. Ultrasound guid-
ance improves the success rate of a perivascular axillary
plexus block. Acta Anaesthesiol Scand 2006;50:678-684.
2. Chan VW, Perlas A, McCartney CJ, et al. Ultrasound guid-
ance improves success rate of axillary brachial plexus
block. Can J Anaesth 2007;54:176-182.
3. Casati A, Danelli G, Baciarello M, et al. A prospective,
randomized comparison between ultrasound and nerve
stimulation guidance for multiple injection axillary brachial
plexus block. Anesthesiology 2007;106:992-996.
4. O’Donnell BD, Iohom G. An estimation of the minimum
effective anesthetic volume of 2% lidocaine in ultrasound-
guided axillary brachial plexus block. Anesthesiology
2009;111:25-29.
5. Vester-Andersen T, Eriksen C, Christiansen C. Perivascular
axillary block III: blockade following 40 ml of 0.5%, 1%
or 1.5% mepivacaine with adrenaline. Acta Anaesthesiol
Scand 1984;28:95-98.
6. Serradell A, Herrero R, Villanueva JA, et al. Comparison
of three different volumes of mepivacaine in axillary plex-
us block using multiple nerve stimulation. Br J Anaesth
2003;91:519-524.
7. Casati A, Fanelli G, Aldegheri G, et al. Interscalene bra-
chial plexus anaesthesia with 0.5%, 0.75% or 1% ropiv-
acaine: a double-blind comparison with 2% mepivacaine.
Br J Anaesth 1999;83:872-875.
8. Smith BE, Siggins D. Low volume, high concentration
block of the sciatic nerve. Anaesthesia 1988;43:8-11.
Med Ultrason 2022; 24(1): 38-43 43
9. Taboada Muñiz M, Rodríguez J, Bermúdez M, et al. Low
volume and high concentration of local anesthetic is more
efficacious than high volume and low concentration in La-
bat’s sciatic nerve block: a prospective, randomized com-
parison. Anesth Analg 2008;107:2085-2088.
10. Vester-Andersen T, Christiansen C, Sørensen M, et al.
Perivascular axillary block II: influence of injected volume
of local anaesthetic on neural blockade. Acta Anaesthesiol
Scand 1983;27:95-98.
11. Krenn H, Deusch E, Balogh B, et al. Increasing the injec-
tion volume by dilution improves the onset of motor block-
ade, but not sensory blockade of ropivacaine for brachial
plexus block. Eur J Anaesthesiol 2003;20:21-25.
12. Brenner D, Iohom G, Mahon P, Shorten G. Efficacy of axil-
lary versus infraclavicular brachial plexus block in prevent-
ing tourniquet pain. Eur J Anaesthesiol 2019;36:48-54.
13. Bertini L, Palmisani S, Mancini S, et al. Does local anes-
thetic dilution influence the clinical effectiveness of mul-
tiple-injection axillary brachial plexus block?: a prospec-
tive, double-blind, randomized clinical trial in patients
undergoing upper limb surgery. Reg Anesth Pain Med
2009;34:408-413.
14. González AP, Bernucci F, Techasuk W, et al. A randomized
comparison between 3 combinations of volume and con-
centration of lidocaine for ultrasound-guided infraclavicu-
lar block. Reg Anesth Pain Med 2013;38:206-211.
15. Cappelleri G, Ambrosoli AL, Turconi S, et al. Effect of local
anesthetic dilution on the onset time and duration of double-
injection sciatic nerve block: a prospective, randomized,
blinded evaluation. Anesth Analg 2014;119:489-493.
16. Nuñez Aguado D, López Alvarez S, Salamanca Montaña
ME, et al. [Brachial plexus block with levobupivacaine at
the humeral canal: comparison of a small volume at high
concentration with a large volume at low concentration].
Rev Esp Anestesiol Reanim 2005;52:529-535.
17. Zhai W, Wang X, Rong Y, Li M, Wang H. Effects of a fixed
low-dose ropivacaine with different volume and concentra-
tions on interscalene brachial plexus block: a randomized
controlled trial. BMC Anesthesiol 2016;16:80.
18. Bernucci F, Gonzalez AP, Finlayson RJ, Tran DQ. A pro-
spective, randomized comparison between perivascular
and perineural ultrasound-guided axillary brachial plexus
block. Reg Anesth Pain Med 2012;37:473-477.
19. Thompson GE, Rorie DK. Functional anatomy of the bra-
chial plexus sheaths. Anesthesiology 1983;59:117-122.
20. Areeruk P, Karmakar MK, Reina MA, Mok LYH, Sivaku-
mar RK, Sala-Blanch X. High-definition ultrasound imag-
ing defines the paraneural sheath and fascial compartments
surrounding the cords of the brachial plexus at the costocla-
vicular space and lateral infraclavicular fossa. Reg Anesth
Pain Med 2021, DOI:10.1136/rapm-2020-102304.
21. Cornish PB, Leaper CJ, Hahn JL. The “axillary tun-
nel”: an anatomic reappraisal of the limits and dynamics
of spread during brachial plexus blockade. Anesth Analg
2007;104:1288-1291.
22. Cornish PB, Leaper C. The sheath of the brachial plexus:
fact or fiction? Anesthesiology 2006;105:563-565.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 44-51
DOI: 10.11152/mu-3162

An ultrasound study of the long posterior sacroiliac ligament

in healthy volunteers and in patients with noninflammatory
sacroiliac joint pain
Plamen Todorov1, Lili Mekenjan1, Rodina Nestorova2, Anastas Batalov1

1Medical University of Plovdiv, Rheumatology Clinic, Kaspela University Hospital, Plovdiv, 2Rheumatology Centre
“St Irina”, Sofia, Bulgaria

Abstract
Aim: To describe the sonoanatomy of the long posterior sacroiliac ligament (LPSL) in healthy volunteers and to assess
by ultrasound the LPSL in patients with noninflammatory sacroiliac joint pain (SIP). Material and methods: We assessed 64
LPSLs of 32 healthy controls and 40 LPSLs of 40 patients with unilateral noninflammatory SIP and a positive Fortin finger
test. LPSLs in both groups were assessed for the presence of alterations in their structure, continuity and echogenicity and
their thickness was measured in three predefined points. All patients were examined in prone position following a strict scan-
ning protocol. Results: Detailed sonoanatomy description and measurement of the LPSL in healthy volunteers are provided
(length: 31.32±4.79 mm, width: 8.14±1.28 mm, thickness: 2.05±0.55 mm; 1.64±0.41 mm and 1.51±0.42 mm at the iliac and
sacral entheses and in its middle part, respectively). The LPSLs were found to be significantly thicker in the SIP group, with
an optimum criterion value of >2.0 mm in its middle part to identify pathologically thickened ligaments. In addition, LPSLs in
the SIP group presented significantly more often hypoechogenicity/altered fibrillar structure (57.5% vs.16%) and/or periliga-
mentous edema (72.8% vs 28%). The combination of either altered structure or periligamentous edema, with thickening of the
ligament’s body showed the best diagnostic accuracy (sensitivity and specificity 83.9% and 94.7% for the first combination
and 100% and 84.6% for the second combination) to identify LPSL pathology in noninflammatory SIP. Conclusions: LPSL
could be assessed by ultrasound and sonopathological lesions could be identified in patients with SIP.
Keywords: ultrasonography; long posterior sacroiliac ligament; sacroiliac joint

Introduction crum during walking and standing [3]. It also serves as an

The long posterior sacroiliac ligament (LPSL) is the
most superficial ligament that supports the sacroiliac
joint (SIJ) [1]. It spans between the inferior pole of the
posterior superior iliac spine (PSIS) and the lateral crest
and transverse tubercula of the third and fourth sacral
segments. [2]. The main function of this ligament is to
stabilize the SIJ, resisting the counternutation of the sa-
Received 18.03.2021  Accepted 14.07.2021
Med Ultrason
2022, Vol. 24, No 1, 44-51
Corresponding author: Plamen Todorov
Rheumatology Clinic, UMBAL Kaspela
64 Sofia street, Plovdiv 4000, Bulgaria
Phone: +359888566478
E-mail: drtodorovplamen@gmail.com
important linking point between the myofascial systems
of the lower back and the gluteal regions, as both the
erector spinae aponeurosis (ESA) and the gluteus max-
imus aponeurosis (GMA) have attachments on LPSL [4].
It was suggested that, in addition to the SIJ, the LPSL
could, by itself, represent a discrete and frequent source
of pain in the sacroiliac region [1,3,5]. Anatomical and
biomechanical studies reveal several potential mecha-
nisms for this situation: 1. as LPSL resists counternuta-
tion in the SIJ, the increased laxity and instability of this
joint in a condition known as SIJ dysfunction, can lead
to an increased tension and eventual damage of the liga-
ment [3]; 2. the LPSL is pierced by the lateral branches of
the dorsal rami of S2 and S3 spinal nerves, which could
be compressed in the thickened, slugged or inflamed
ligament giving rise to pain due to entrapment neurop-

athy [6,7]; and 3. histological data shows that both ESA
and GMA aponeuroses are in fact an integral part of the
LPSL, forming its layered structure (fig 1) [4]. Thus, an
unproper or discordant pull from these two big muscles
could potentially lead to internal PSIL injury or dehis-
cence of its layers.
Clinical assessment of the LPSL can be challenging.
The ligament can be palpated just caudal to PSIS but is
difficult to differentiate due to its bone-like texture [3].
On the other hand, pain in the same area (as pointed out
by the patient) is considered a good predictor of SIJ pa-
thology and was named Fortin finger test [8,9].
Being a relatively superficial structure, the LPSL
should be well suited for ultrasound (US) examination.
Indeed, Moore et al and Le Goff et al were able to vi-
sualize the ligament with sufficient details in healthy
volunteers [10,11]. These authors described a scanning
protocol and the basic normal sonoanatomy of the LPSL,
thus providing the background for the assessment of this
ligament in patients.
While such an application of US seems logical and
practical, to the best of our knowledge, there are no pub-
lished studies on the subject. The aims of our study were
to describe in detail the sonoanatomy of the LPSL in
healthy subjects, to assess the LPSL in patients suffer-
ing from subacute or chronic noninflammatory sacroiliac
joint pain (SIP) and to elaborate diagnostic models to iden-
tify accurately LPSL pathology in the clinical practice.
Material and methods
The study was approved by the Ethical Commission
of the Medical University of Plovdiv and all participants
signed an informed consent form at enrollment.
In the first stage of the study, 32 healthy individuals
with no current or over the last year pain in the lower
back, pelvic girdle or hips were recruited. In this group,
the normal sonoanatomy of LPSL was assessed and de-
scribed.
For the second stage of the study, participants were
selected from patients referred to a tertiary rheumatology
center due to unilateral subacute or chronic SIP. Patients
were offered to participate if their SIP did not have the
character of inflammatory back pain according to the
Assessment of Spondyloarthritis International Society
(ASAS) criteria [12], had a recent X-ray of the pelvic
girdle with no radiographic signs of sacroiliitis and clini-
cally had a positive Fortin finger test [8]. In total, 40 pa-
tients were enrolled according to the outlined inclusion
criteria.
Exclusion criteria were a diagnosis of Spondyloarthri-
tis (SpA), age under 18, BMI ˃30, radiculitis or neuro-
Med Ultrason 2022; 24(1): 44-51 45
Fig 1. The normal long posterior sacroiliac ligament (LPSL).
Transverse section across the posterior sacroiliac region (H&E):
The erector spinae aponeurosis (ESA) medially, the deep fas-
cial layer (DFL), inferio-medially and the gluteal aponeurosis
(GA) laterally, all attached to the long posterior sacroiliac liga-
ment (LPSL). Deep to the LPSL and posterior to the sacroiliac
joint (SIJ), a region of adipose and loose connective tissue (Ad)
extended laterally, deep to the GA over the ilium and medially,
deep to the DFL over the posterior sacrum. I: Ilium; S: Sacrum;
GMx: Gluteus maximus, M: multifidus, C: sacral canal, SF: sa-
cral foramina (Reprinted with permission from: McGrath et al.
Joint Bone Spine 2009;76:57-62).
pathic pain, fibromyalgia or myofascial pain syndromes,
recent osteoporotic fractures, alloplastic hip joint(s), sur-
gical procedures in the lumbar spine, SIJ cortisone injec-
tions in the last six months.
Ultrasound examination
Esaote My Lab 7 machine with a linear multifrequen-
cy transducer (4-12 MHz) was used. All subjects were
examined in prone position by the same sonographer
(PT) with 10 years of experience in musculoskeletal US.
A strict scanning protocol was followed for all subjects.
First, the probe was placed in the transverse plane over
the midline of the sacrum. After the characteristic contour
of the sacral spinous processes was identified, the probe
was moved laterally, first to the right side. When the SIJ
was reached, the probe, kept in the transverse plane, was
moved upwards following the contour of the iliac bone
until the PSIS was identified. At this point the medial part
of the transducer was rotated counterclockwise, while the
lateral part was kept at the PSIS, until a well-defined fi-
brillar structure, spanning between the inferior pole of
the PSIS and a tuberculum at the lateral border of the sa-
crum, was identified (fig 2a). This structure, visible in the
described oblique position, is the LPSL. To eliminate the
anisotropy, a slight pressure was applied with the inferior
pole of the probe. Then the transducer was slowly rotated
90 degrees to stand perpendicular to the LPSL, examin-
ing it in the transverse axis. The same protocol was car-
ried out for the left sided ligament (here the rotation of the
transducer, once PSIS is identified, should be clockwise).
46 Plamen Todorov et al Long posterior sacroiliac ligament: healthy volunteers vs. patients with NI sacroiliac joint pain
In still images the LPSL thickness was measured in
its long axis at three reference points: at its iliac enthesis
(at a point where the deep margin of the ligament meets
the iliac bone), at its sacral enthesis (at a point where the
deep margin of the ligament meets the sacrum) and at
its middle part (ligament body). In addition, the ligament
body thickness and width were measured in the short
axis. As a final value at a given point, the mean of three
consecutive measurements was recorded, as studies show
that repetitive measurements provide better reliability
[13]. In addition, LPSL was assessed for possible so-
nopathological features: 1. hypoechogenicity or altered
fibrillar structure; 2. presence of well-defined anechoic
zones within the ligament (partial-thickness tears); 3. an
anechoic rim along the entire or at least one third of the
ligament length (periligamentous edema). All of these
sono-markers were recorded as dichotomous variables,
i.e. present/absent.
Reference images of the LPSL were saved in a JPG
format. They were used to test the intra-reader reliability
of the ultrasonographer (PT) and an inter-reader agree-
ment with a novice sonographer with only a several
months experience (LM). Before testing, the novice so-
nographer received a special training in LPSL anatomy
and US appearance. For the reliability testing, 10 ran-
domly selected sets of images, each set consisting of two
reference images of a single ligament, were used. The
intra-reader testing was performed about 6 months after
the initial scan, blinded to the previous results and the
identity of patients/controls. The sonopathological le-
sions reported are relative to the first analysis.
Statistical analysis
The statistical software used to perform the analysis
included IBM SPSS version 26 (2018), Minitab version
19 (2019) and MedCalc version 19.4.1 (2020). We exam-
ined the data for normality through the Shapiro-Wilk’s
test and took into consideration the values of skewness.
Continuously measured and normally distributed vari-
ables were described through the mean values and stand-
ard deviations (±SD). Categorical data were processed
in frequencies and percentages. An independent-samples
t-test was used for two-group comparisons on normally
distributed continuous variables. When the assumption
of equal variances was not observed (Levene’s statis-
tic p<0.05), we reported the results for equal variances
not observed (uv). For multiple comparisons, Bonfer-
onni correction was applied to control for Type I error.
In such cases, statistical significance was accepted if
p<0.0125.
ROC curve analysis was used to establish the di-
agnostic potential, optimal criterion values and the as-
sociated sensitivity and specificity of LPSL thickness.
Associations between categorical variables (individual
sono-lesions) were examined through Chi-square test
and/or Fisher’s exact test. Odds ratios were calculated to
establish the odds for the occurrence of a certain sono-
lesion in patients with SIP. Pearson correlation r was
employed to examine the relationship between continu-
ously measured and normally distributed variables and
Spearman rank-order correlation was performed when
those conditions were not observed. The intra-rater and
inter-rater agreement were established through Cohen’s
kappa (95% confidence intervals).
Results
LPSL in healthy individuals
The LPSL was assessed bilaterally in 32 healthy in-
dividuals (64 ligaments), 22 females, 10 males, mean
age of 41.97±12.87 years and mean BMI of 23.21±3.52.
We were able to visualize the ligament with the above-
described protocol in all subjects.
In the longitudinal axis, the LPSL was seen as a hy-
perechoic, slightly concave fibrillar structure, with well-
defined margins and predominantly uniform thickness
that spans over the SIJ and attaches to the ilium (just in-
ferior to the PSIS) and the lateral sacrum (to the sacral
tubercle, a structure of variable size in different subjects)
(fig 2b). In the short axis, the LPSL exhibited a flat to
oval shape, layered structure, well-defined upper margin,
and less conspicuous inferior margin (fig 2c). Medially,
the ligament merged with the lateral portion of ESA, a
finding also shown previously [14]. The gluteus maxi-
mus muscle was seen to override the ligament from a
lateral direction, while the inferior gluteal aponeurosis
attached to the LPSL’s lateral part. Thus, the LPSL ap-
peared as a link between the thicker and more stretched
ESA medially and the thinner and looser GMA laterally.
Beneath the ligament, a fibro-adipose tissue of variable
amount and echogenicity was seen. This tissue occupied
the posterior part of the SIJ and expanded to the sacral
foramina and over the ilium.
The mean length of the LPSL was 32.32±4.29 mm,
and its thickness at the iliac entheses, the middle part
(ligament body) and the sacral entheses was 2.05±0.55
mm, 1.51±0.42 mm and 1.64±0.41 mm, respectively. The
width of the LPSL body measured in the transverse plane
was 8.14±1.28 mm.
No significant associations were observed between
the subjects’ age, sex, the side of the body and LPSL
measurements.
LPSL in patients with SIP
For the second part of the study, 40 SIP patients were
enrolled according to the above-pointed criteria and com-
 Med Ultrason 2022; 24(1): 44-51 47

Fig 2. a) Position of the transducer to assess the left LPSL in its longitudinal axis; b) ultrasound image of the normal LPSL in the
longitudinal plane. The ligament spans from the posterior superior iliac spine (PSIL) of the ilium (I) to the sacral tuberculum (ST)
of the lateral sacrum (S). Deep to it are laying the thinner and less conspicuous interosseous ligaments (iol) that occupy the posterior
part of the sacroiliac joint (SIJ) cleft; c) ultrasound image of the middle part (body) of a normal LPSL in the transverse plane, the
left side of the image is medial. The erector spinae aponeurosis (ESA) and the deep fascial layer (dfl) attach to the LPSL from the
medial side, while the inferior gluteal aponeurosis (GA) attaches to the lateral side. Deep to LPSL, the thinner and less conspicuous
interosseous ligaments (iol) in cross section might be seen. Posterior to the sacroiliac joint (SIJ), a region of adipose tissue is seen
(Ad). GMx: Gluteus maximus, M: multifidus, SF: sacral foramina.

pared to the healthy controls. The demographic data is

detailed in Table I.
The thickness of the LPSL in SIP patients measured
at the three reference points, i.e. the iliac enthesis, the
middle part (ligament body) and the sacral entheses was
3.23±1.00 mm, 2.55±0.70 mm and 2.52±0.66 mm, re-
spectively. These measurements were compared through
independent-samples t-tests with an adjusted alpha level
of 0.0125 to the thickness of the 64 ligaments (32 right
and 32 left sided) of the 32 control subjects described in
the first part of the study.
A strong to very strong significant trend associated
with higher LPSL thickness values in the SIP patients Fig 3. Individual value plot of the iliac enthesis, ligament body
group as compared to the healthy controls was estab- and sacrum enthesis thickness measurements. ***: p<0.001.
lished at all three measurement points (<0.001) (fig 3)
A ROC curve analysis revealed that among the three Among the assessed sonopathological lesions, the
measurements, the ligament body had the highest level most frequently detected one was the periligamentous
of diagnostic potential in distinguishing LPSL in SIP edema (fig 4a), followed by the hypoechogenicity/altered
from those in controls (AUC = 0.910, p<0.001) with op- fibrillar structure of the ligament (fig 4b,c), while par-
timum criterion value of >2.0 mm (79.49% sensitivity tial thickness tears (fig 4d) were rare. The frequency of
and 91.53% specificity). observation of these lesions in the LPSLs of both study
groups are summarized in Table II.
Table I. Demographic data of the sacroiliac pain (SIP) patients The majority of the LPSLs in the SIP patients (84%)
and healthy controls exhibited ≥2 of the assessed sonopathological markers
Variables Groups p (fig 4e-h). No significant association was found between
SIP patients Healthy controls the cumulative number of presented sonopathological
(N = 40) (N = 32) features and the age and sex of the participants.
Age 42.90±11.35 41.97±12.87 0.708t The intra-reader agreement was perfect for hypoecho-
(20-70) (18-69) genicity (kappa = 1.0, 95 CI:1.0 to 1.0) and tears (kappa
Female gender 28(70) 22 (69) 1.00f = 1.0, 95% CI:1.0 to 1.0) and good for periligamentous
Height 169.78±6.62 167.50±9.07 0.143t(uv) edema (kappa = 0.800, 95% CI: 0.43 to 1.0). The inter-
reader agreement with a novice sonographer was perfect
BMI 24.56±3.26 23.21±3.52 0.054t
for periligamentous edema (kappa = 1.0, 95% CI:1.0 to
The results are expressed as mean±standard deviation, (minim- 1.0), acceptable for tears (kappa = 0.782, 95%CI: 0.38
maxim) or number(%). t - independent-samples t-test; f - Fisher’s
exact test; t(uv) - independent-samples t-test with unequal vari- to 1.00), but low for hypoechogenicity/altered fibrillar
ances structure (kappa = 0.600, 95% CI: 0.14 to 1.00).
48 Plamen Todorov et al Long posterior sacroiliac ligament: healthy volunteers vs. patients with NI sacroiliac joint pain

Fig 4. The spectrum of LPSL sonopathological lesions: a) Periligamentous edema (*); b) thicken and hypoechoic LPSL (*) in the
longitudinal plane and c) in the transverse plane; d) LPSL with a partial thickness tear (*); e) a thicken LPSL with a partial tear (*) in
the longitudinal plane and f) in the transverse plane; g) a thicken LPSL with altered, hypoechoic structure (*); h) LPSL with a para-
ligamentous edema and an altered fibrillar structure (*). I-ilium, S-sacrum, ST-sacral tuberculum, SIJ-sacroiliac joint; GMx – gluteus
maximus. iol – interosseus ligaments, M-multifidus, FT – fatty tissue.

Table II. Sonopathology of long posterior sacroiliac ligament in sacroiliac pain (SIP) patients and healthy controls
Parameters SIP patients Healthy controls OR (95% CI) p
Hypoechogenicity
Positive 23 (57.5) 10 (16) 7.20 (2.80 to 18.50) <0.001
Negative 15 (37.5) 47 (73)
Undetermined 2 (5) 7 (11)
Periligamentous edema
Positive 29 (72.5) 18 (28) 7.25 (2.94 to 17.82)
Negative 11 (27.5) 46 (72) <0.001
Undetermined 0 (0) 0 (0)
Intraligamentous tear
Positive 7 (17.5) 0 (0) 26.64 (1.47 to 481.24)
Negative 33 (82.5) 59 (87.5) 0.026
Undetermined 0 (0) 8 (12.5)
The results are expressed in absolute numbers(%).

Diagnostic US models for identification of LPSL
pathology
Thickening of the ligament body was found to be
associated with the largest AUC among all three points
of measurement. Periligamentous edema and hypoecho-
genicity/altered fibrillar structure were the two most
frequently observed sonopathological markers. Accord-
ingly, we tested the cumulative diagnostic accuracy of th
ligament body thickening in conjunction with either hy-
poechogenicity or periligamentous edema through binary
logistic regression and ROC curve analysis.
The first model (ligament body thickening plus hy-
poechogenicity/altered structure) was significant (Chi-
square = 76.217, p<0.001), with AUC = 0.953 (95%CI:
0.911 to 0.995), sensitivity = 83.78%, specificity =
94.74%, positive predictive value = 91.19%, negative
predictive value = 89.98%.
The second model (ligament body thickening plus
periligamentous edema) was also significant (Chi-square
= 80.49, p<0.001) with AUC = 0.964 (95%CI: 0.931 to
0.996), sensitivity = 100%, specificity = 84.75%, positive
predictive value = 81.26%, negative predictive value =
100%.
Discussion
There is growing evidence that soft tissue structures
posterior to the SIJ (including the LPSL) could be respon-
sible for a significant proportion of SIP [1,3,5,15,16]. For
example, it is frequently reported that there is no major

difference in the efficacy of SIJ injections, regardless of
whether they are delivered strictly intraarticularly or the
infiltration is done periarticularly [17,18]. Indeed, the US
evaluation of the LPSL, performed in our study, revealed
ligamentous sonopathological lesions in more than two
thirds of the patients with SIP. This is in line with the data
of Murakami et al, who advocate that the first interven-
tion in SIP should be an injection in its ligamentous area,
as four of five SIP patients would have benefit from this
approach [19]. The role of sacroiliac ligaments as pain
generators was further confirmed empirically by Saun-
ders et al. The authors evaluated the effect of US guided
injections in the LPSLs in patients with SIJ dysfunction
and pain [20] and reported significant improvement in
all scores and performance indicators that lasted through-
out the entire 12-month follow-up period. In addition, a
recent study, performed by SPECT-CT, shows increased
uptake in the region of LPSL in patients complaining of
SIP and a clinical diagnosis of SIJ dysfunction [21]. Fur-
thermore, the intensity of the uptake could be well quan-
tified and possibly used as a follow-up tool. However,
this imaging method utilizes an ionizing radiation, which
would limit its use in the daily clinical practice.
In the present study we used US, a safe and widely
available imaging method. US was first used to assess
the LPSL in healthy volunteers by Moore et al and later
by LeGoff et al [10,11]. In addition to these studies, we
provide data on the normal thickness and width of LPSL
as well as more detailed description of its sonoanatomy
and relations to the surrounding structures. We have
measured the LPSL thickness at three easily reproduc-
ible points, namely at its iliac and sacral entheses and
at its middle part. These sites were chosen on the basis
of the morphology data, showing a different structure
of the ligament’s body in comparison to its entheses
[4].
Thickening is one of the major US features of liga-
mentous injury [22]. Our results show that thickness was
increased at all three sites, the ligament body being most
pronounced. Comparing to its entheses, the LPSL body
has a more complex structure, consisting of layers and
incorporating the ESA and GMA. Thus, ligament body
thickening could in fact reflect the thickening of either
or both of these aponeuroses – a finding similar to the
thickening of the thoracolumbar fascia in patients with
chronic low back pain, described by Langevin et al [23].
However, this suggestion should be confirmed by his-
tological proof. This ligament thickening could also be
the morphological basis for the compression of the lat-
eral branches of S2 and S3 dorsal rami and entrapment
neuropathy which was proposed by different authors as a
potential reason for SIP [6,7,11].
Med Ultrason 2022; 24(1): 44-51 49
In the majority of ligaments in the SIP study group
(72.5%), we observed a periligamentous edema. It was
seen as well-defined anechoic rim superiorly or, less
frequently, inferiorly along the LPSL body. This phe-
nomenon has been previously reported in traumatic and
degenerative ligamentous pathology [24,25], but its his-
tological nature remains unclear. In the case of LPSL
it may represent a pathological process similar to para-
tenonitis that involves the ESA and/or GMA.
In assessing the fibrillar structure and the echogenici-
ty of the LPSL, we found diffuse or local hypoechogenic-
ity/altered fibrillar structure significantly more often in
ligaments from the SIP patients group. Hypoechogenic-
ity of ligaments is considered to be caused by the disor-
ganization of the collagen fibrils and edema of the clefts
between the fibrils and the ligament matrix [22]. Thus,
hypoechogenicity may indicate a pathological transfor-
mation of the ligamentous tissue, especially when this
finding is accompanied by a thickening. In addition to
this, the altered fibrillar structure may represent focal or
more global dehiscence between the ligament’s layers.
Less frequently, partial intrasubstance tears of the
LPSL were detected. This might be a rare injury in a tight
ligament such as the LPSL that supports an intrinsically
stable joint as the sacroiliac [1,2], but nevertheless, no
tears were detected among the 64 ligaments in the control
group.
The analysis of the cumulative occurrence of patho-
logical sono-markers (other than thickening) in the LPSL
of patients vs. controls revealed that the majority of the
painful ligaments (84%) exhibited two or all three patho-
logical sono-markers. The combination of ligament’s
body thickening and the presence of hypoechogenic-
ity/altered fibrillar structure or periligamentous edema
yielded the highest values for sensitivity and specificity
in identifying ligamentous injury in SIP patients.
In our study we used a single clinical test to diag-
nose patients as having probable SIP (Fortin finger test
[8]), while most of the studies on SIJ pathologies use a
composite of tests [9,16,20]. Anyway, our target was the
PSIL, which is an extraarticular and relatively superficial
structure [3]. Such extraarticular and superficial source
of pain would be expected to be associated with a well-
described locus of pain [1,6] that the patents would be
able to point out – like in the Fortin finger test. In addi-
tion, a similar test was used by other authors under the
name of a one-finger test resulting in a more accurate
identification of the site of pain in the sacroiliac region
[26].
Our study has certain limitations. First and foremost,
findings were not compared to a “gold standard” or an-
other imaging modality. However, there is no approved
50 Plamen Todorov et al Long posterior sacroiliac ligament: healthy volunteers vs. patients with NI sacroiliac joint pain
gold standard for the imaging of LPSL and people with
SIP would not normally need an operation. Thus, a con-
firmation of the sonographic findings at surgery as a gold
standard is virtually impossible. Two other possible im-
aging modalities to assess the LPSL would be MRI and
SPECT-CT. MRI has certain limitations as ligaments
are relatively avascular structures. Besides MRI is still
an expensive option. SPECT-CT seems promising for
depicting LPSL pathology, but it involves a consider-
able amount of ionizing radiation, and the images have a
lower anatomical resolution. Secondly, patients and con-
trol subjects in the study were with normal BMI while in
more obese patients, the examination of the LPSL could
be more difficult. However, as the ligament presents a
well-defined hyperechoic fibrillar structure, it should be
possible to assess it even in these cases. Thirdly, through-
out the course of data collection, the researcher undertak-
ing the US assessment was not blinded to the group of
each participant, and as such, there was a potential for
observer bias. However, to minimize subjectivity, a strict
scanning protocol was followed in each subject and the
sonopathological lesions were well- and predefined and
repeated measurements undertaken. Lastly, we did not
use diagnostic periarticular SIJ injections to differenti-
ate patients with ligamentous pain. Yet, periarticular SIJ
injections are still not that well standardized (as the in-
traarticular ones) and one problem to their diagnostic use
might be the unpredictable spread of the injectate in the
periarticular tissues.
Conclusion
Our study confirms that the LPSL could be visualized
by US in details and its structure and relations - accurate-
ly assessed. In addition, in patients with noninflamma-
tory SIP, the US evaluation of the LPSL could reveal its
thickening as well as sonopathological transformation:
hypoechogenicity/altered fibrillar structure and periliga-
mentous edema. Further studies are required in order to
confirm the value of these findings for the clinical practice.
Conflict of interest: none
References
1. Bogduk N. Clinical anatomy of the lumbar spine and sa-
crum. 4th Edition. Elsevier, 2005.
2. Vleeming A, Schuenke MD, Masi AT, Carreiro JE, Dan-
neels L, Willard FH. The sacroiliac joint: an overview of
its anatomy, function, and potential clinical implications. J
Anat 2012;221:537-567.
3. Vleeming A, Pool-Goudzwaard AL, Hammudoghlu D,
Stoeckart R, Snijders CJ, Mens JM. The function of the long
dorsal sacroiliac ligament: its implication for understanding
low back pain. Spine (Phila Pa 1976) 1996;21:556-562.
4. McGrath C, Nicholson H, Hurst P. The long posterior sac-
roiliac ligament: a histological study of morphological rela-
tions in the posterior sacroiliac region. Joint Bone Spine
2009;76:57-62.
5. Borowsky CD, Fagen G. Sources of sacroiliac region pain:
insights gained from a study comparing standard intra-ar-
ticular injection with a technique combining intra- and peri-
articular injection. Arch Phys Med Rehabil 2008;89:2048-
2056.
6. McGrath MC, Jeffery R, Stringer MD. The dorsal sacral
rami and branches: Sonographic visualization of their vas-
cular signature. Int J Osteopath Med 2012;15:3-12.
7. Zhou L, Schneck CD, Shao Z. The Anatomy of Dorsal Ra-
mus Nerves and Its Implications in Lower Back Pain, Neu-
rosci Med 2012;3:192-201.
8. Fortin JD, Falco FJ. The Fortin finger test: an indicator of
sacroiliac pain. Am J Orthop (Belle Mead NJ) 1997;26:477-
480.
9. 9. Bogduk N. Pain provocation tests for the assessment of
sacroiliac joint dysfunction. J Spinal Disord 1999;12:357-
358.
10. Moore AE, Jeffery R, Gray A, Stringer MD. An anatomical
ultrasound study of the long posterior sacro-iliac ligament.
Clin Anat 2010;23:971-977.
11. Le Goff B, Berthelot JM, Maugars Y. Ultrasound assess-
ment of the posterior sacroiliac ligaments. Clin Exp Rheu-
matol 2011;29:1014-1047.
12. Arnbak B, Hendricks O, Hørslev-Petersen K, et al. The
discriminative value of inflammatory back pain in pa-
tients with persistent low back pain. Scand J Rheumatol
2016;45:321-328.
13. Fede C, Gaudreault N, Fan C, Macchi V, De Caro R, Stecco
C. Morphometric and dynamic measurements of muscular
fascia in healthy individuals using ultrasound imaging: a
summary of the discrepancies and gaps in the current litera-
ture. Surg Radiol Anat 2018;40:1329-1341.
14. Todorov P, Nestorova R, Batalov A. The sonoanatomy of
lumbar erector spinae and its iliac attachment - the poten-
tial substrate of the iliac crest pain syndrome, an ultrasound
study in healthy subjects. J Ultrason 2018;18:16-21.
15. Kurosawa D, Murakami E, Ozawa H, et al. A Diagnostic
Scoring System for Sacroiliac Joint Pain Originating from
the Posterior Ligament. Pain Med 2017;18:228-238. 
16. Vleeming A, Mooney V, Stoeckart R. Movement, stability
and lumbopelvic pain. Integration of research and therapy.
2nd Edition. Churchill Livingstone Elsevier, 2007.
17. Nacey NC, Patrie JT, Fox MG. Fluoroscopically Guided
Sacroiliac Joint Injections: Comparison of the Effects
of Intraarticular and Periarticular Injections on Immedi-
ate and Short-Term Pain Relief. AJR Am J Roentgenol
2016;207:1055-1061.
18. Murakami E, Tanaka Y, Aizawa T, Ishizuka M, Kokubun S.
Effect of periarticular and intraarticular lidocaine injections
for sacroiliac joint pain: prospective comparative study. J
Orthop Sci 2007;12:274-280.

19. Murakami E, Kurosawa D, Aizawa T. Treatment strategy
for sacroiliac joint-related pain at or around the posterior su-
perior iliac spine. Clin Neurol Neurosurg 2018;165:43-46.
20. Saunders J, Cusi M, Hackett L, Van der Wall H. An explora-
tion of ultrasound-guided therapeutic injection of the dorsal
interosseous ligaments of the sacroiliac joint for mechanical
dysfunction of the joint. JSM Pain Manag 2016;1:1003-1007.
21. Cusi M, Saunders J, Van der Wall H, Fogelman I. Meta-
bolic disturbances identified by SPECT-CT in patients with
a clinical diagnosis of sacroiliac joint incompetence. Eur
Spine J 2013;22:1674-1682.
22. Hodgson RJ, O’Connor PJ, Grainger AJ. Tendon and liga-
ment imaging. Br J Radiol 2012;85:1157-1172.  26. Murakami E, Aizawa T, Noguchi K, Kanno H, Okuno H,
23. Langevin HM, Stevens-Tuttle D, Fox JR, et al Ultrasound
evidence of altered lumbar connective tissue structure in
Med Ultrason 2022; 24(1): 44-51 51
human subjects with chronic low back pain. BMC Muscu-
loskelet Disord 2009;10:151.
24. Shahabpour M, Staelens B, Van Overstraeten L, et al. Ad-
vanced imaging of the scapholunate ligamentous complex.
Skeletal Radiol 2015;44:1709-1725.
25. Ward R, Mendoza J, Wong E, et al. MCL Periligamen-
tous Edema: Sprain or Nontraumatic Meniscal Pathology?
Radiological Society of North America 2010 Scientific
Assembly and Annual Meeting, November 28 - Decem-
ber 3, 2010, Chicago IL. Available from: http://archive.
rsna.org/2010/9012278.html, Accessed September 28,
2020.
Uozumi H. Diagram specific to sacroiliac joint pain site in-
dicated by one-finger test. J Orthop Sci 2008;13:492-497.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 52-57
DOI: 10.11152/mu-2996

Doppler ultrasonographic evaluation of radial and ulnar artery

diameters and blood flow, before and after percutaneous coronary
interventions
Yasemin Gunduz1, Huseyin Gunduz2, Omer Faruk Ates1, Mahmut Ciner1, Ahmet Can
Cakmak2, Cagla Akcay2, Ersin Ilguz2, Kahraman Cosansu2

1Radiology Department, 2Cardiology Department, Sakarya University Medical Faculty, Sakarya, Turkey

Abstract
Aim: Although the transforearm approach is considered a safe and effective option for percutaneous coronary intervention,
the different characteristics of the radial and ulnar arteries deserve attention. This study aimed to evaluate radial (RA) and ulnar
artery (UA) diameter and blood flow parameters changes after catheterization. Material and method: A total of 328 patients
were enrolled. Their artery (171 RA and 157 UA) diameter and flow parameters [peak systolic velocity (PSV), end-diastolic
volume (EDV) and pulsatility index (PI)] were evaluated before and after catheterisation. Results: After RA catheterization,
the diameters and PSV decreased in the RA (from 2.71±0.66 to 2.47±0.51, p=0.007; from 44.7±8.3 to 33.9±9.5, p=0.021) and
increased in the UA (from 2.49±0.83 to 2.59±0.58, p=0.033; from 48.3±11.9 to 59.6±11.0, p<0.001). After UA catheteriza-
tion, the diameters and PSV decreased in UA (from 2.53±0.65 to 2.49±0.77 mm, p=0.173; from 51.2±13.1 to 44.3±10.7 cm/s,
p=0.081) and increased in RA (from 2.67±0.54 to 2.76±0.43 mm, p=0.040; from 41.8±10.3 to 48.6±7.9 cm/s, p=0.054). Con-
clusions: The marked increase in the diameter and PSV of the non-catheterized artery may indicate compensatory changes
in the hand arterial network. Acute wall changes may have led to an increase in total wall thickness and a modest decrease in
lumen size and blood flow velocity in the catheterized artery.
Keywords: Doppler ultrasonography; radial artery; ulnar artery; coronary angiography; percutaneous coronary interven-
tion

Introduction fective option for PCI, the different anatomical and flow
characteristics of the RA and UA deserve attention [1,2].
The radial artery (RA) and ulnar artery (UA) are com- Both arteries form dense anastomotic arches that provide
monly the preferred access sites for coronary angiogra- sufficient blood flow to the hand. This relationship is less
phy (CAG) and percutaneous intervention (PCI). While clear at the level of the wrist because the UA gives off
the transforearm approach is considered a safe and ef- multiple branches in the forearm, whereas the RA serves
mainly as an arterial conduit to the hand [3]. In addition,
the closed-loop nature of the radio-ulnar circulation also
Received 12.12.2020  Accepted 05.06.2021 provides an opportunity for the potential alteration of
Med Ultrason flow characteristics in one artery by altering flow in the
2022, Vol. 24, No 1, 52-57
Corresponding author: Yasemin Gunduz, MD, Associate Professor
other artery [4].
Sakarya University Medical Faculty, The arterial velocity increases in response to com-
Radiology Department, pression or occlusion of the opposite artery at the wrist
Istiklal mah, 342 sokak, in normal subjects. Thus, when the RA is compressed at
Hakkı Bey konakları,
54100, Serdivan, Sakarya, Turkey
the level of the wrist, UA blood flow increases and an
Phone: 0905322769390 increase in UA velocity after RA compression is a sign
E-mail: dryasemingunduz@yahoo.com of adequate collateral perfusion. Similarly, when the UA

is compressed, RA blood flow increases, which indicates
a functional continuity between radial and ulnar circula-
tion in the hand [5].
The first invasive stage of PCI procedures is the cath-
eterization of the RA or UA. In many studies, despite
conflicting results [6], changes in blood flow parameters
and diameters of the RA and UA after reciprocal com-
pression/occlusion have been investigated in the early or
late periods of the procedure using Doppler ultrasonog-
raphy (DUS) [7-10]. However, blood flow characteristics
and diameter changes in both arteries (catheterized and
non-catheterized) before and after RA or UA catheteriza-
tion for CAG and/or PCIs have not been examined or
compared during the early period.
Our study hypothesis focusing on RA and UA cath-
eterization (including cannulation, decannulation, sheath
insertion, sheath removal and 4 hours of gradually de-
creasing compression) was that this procedure will cause
changes in the structure and hemodynamics of both arter-
ies. For this reason, the aim of our study was to evaluate
the structural and hemodynamic changes in RA and UA
after catheterization by using DUS.
Materials and methods
Study population
Between January 2018 and August 2020, a total of
328 patients who underwent CAG and/or PCI were in-
cluded in the study: 171 patients via a transradial ap-
proach (TRA), 157 patients via a transulnar approach
(TUA). All procedures were performed through the right
UA and RA. DUS was carried out before procedure
(maximum 48 hours before catheterisation) and 5-48
hours after compression bandage was removed.
The main indication for TUA or TRA was a wider
and easily palpable pulse. The demographic and clinical
characteristics (age, gender, systolic and diastolic blood
pressure, clinical diagnoses, fasting blood glucose, lipid
profile, history of diabetes mellitus, hypertension, smok-
ing, chronic kidney disease, and drug use) were recor-
ded.
Study exclusion criteria were as follows: <18 or >80
years of age, previous CAG and/or PCI by TRA or TUA,
severe heart failure, hemodynamic impairment, uncon-
trolled hypertension, acute or chronic renal failure or
hepatic diseases, chronic pulmonary disease, bleeding
diathesis, severe skeletal forearm deformities, peripheral
artery disease (i.e. Raynaud’s disease), or previous coro-
nary artery bypass surgery using an RA graft.
Sakarya University Medical Faculty Ethics Commit-
tee approved the study protocol and each patient signed a
written informed consent form.
Med Ultrason 2022; 24(1): 52-57 53
Catheterization technique
The right forearm was abducted and placed on the
catheterization table with hyperextension of the wrist.
After a local anesthetic injection, arterial puncture was
achieved. Following palpation of the site of maximal
pulse prominence, using the Seldinger technique, a 0.025-
inch straight guidewire was passed through a needle. The
needle was removed and a 5F or 6F hydrophilic sheath
was placed over the guidewire. A spasmolytic cocktail
(100-200 µg of Perlinganit and 12.5 µg of diltiazem, if
required) and heparin (5000 IU bolus in case of diagnos-
tic catheterization or 70-100 IU/kg in case of PCI) were
infused intra-arterially through the sheath. For diagnostic
purposes, conventional 5F Tiger catheters (Terumo Corp.,
Tokyo, Japan), and for PCI 5F or 6F (rarely, in complex
PCI) extra back-up, Judkins, or Amplatz guiding cath-
eters were used. After CAG or PCI, the arterial sheath
was removed immediately following cardiac catheteriza-
tion completion. Local haemostasis was achieved using a
pressure bandage (Terumo, Tokyo, Japan) inflated at the
puncture site with 15 to 20 mL of air. The pressure used
for UA compression was higher than that for the RA, as
there is no bone structure to compress the UA at the bot-
tom and the UA is a deeper vessel. These bandages were
removed after 4 hours and the patients were discharged
within 48 hours of procedure completion. Catheterization
procedures for both arteries were performed by the same
operator (H.G.) with more than 10 years of experience
and over 7000 catheterizations.
Ultrasonographic evaluation
All DUS examinations were performed by the same
radiologist (Y.G) with more than 20 years of experience,
in DUS using an Aplio 400 Doppler USG system (Toshi-
ba, Tokyo, Japan) with a 7.2-14 MHz multi-frequency
linear array transducer.
The lumen diameters, peak systolic velocity (PSV),
end-diastolic velocity (EDV), pulsatility index (PI) and
resistive index (RI) were measured at the wrist level. The
sagittal diameter of the RA and UA were measured at
the diastolic phase (from intima to intima) in a short-axis
section. The ultrasound examination was performed at
room temperature (22 and 23°C), the examination room
temperature being recorded before each DUS.
Statistical analysis
An independent Student’s t-test was used to compare
data for continuous variables which were expressed as
mean and standard deviation [age, blood pressure, blood
biochemistry measurements (fasting glucose, LDL cho-
lesterol, and triglycerides)]. The diameter and Doppler
flow of the RA and UA before and after the procedure
were compared using a paired t-test. Categorical varia-
bles expressed as numbers and percentages (gender, clin-
54 Yasemin Gunduz et al Doppler ultrasonographic evaluation of radial and ulnar artery diameters and blood flow

ical diagnosis, diabetes mellitus, hypertension, smoking,
dyslipidaemia, and chronic kidney disease) were com-
pared by Pearson’s chi-squared test or Fisher’s exact test,
when appropriate.
All analyses were performed using the SPSS Statis-
tics software package for Windows, version 22.0 (SPSS
Inc., Chicago, IL, USA). A value of P <0.05 was consid-
ered statistically significant.
Considering the RA and UA catheterizations as two
different access methods (from 2020 Medcalc statisti-
cal software, Ostend, Belgium), the measurements were
plotted using the Bland-Altman method to determine
whether the differences showed a systematic distribution
around zero and its prevalence.
Results
A total of 328 consecutive adult patients undergoing
selective transradial (171) or transulnar (157) catheteri-
zation procedures for CAG and PCI were included in this
study. The demographic, clinical, laboratory and proce-
dural time data of these groups are detailed in Table I.
Ultrasonographic measurements of radial and ulnar
arteries before and after catheterization are specified in
table II.
Comparing the right and left RA and UA preprocedur-
al ultrasound parameters, no significant differences were
found concerning the diameters (left RA and 2.64±0.74
mm, left UA 2.47±0.98 mm) or Doppler flow measure-
ments (all p>0.05). The RA was the dominant forearm
artery in most cases (198 patients, 62.2%).
Due to the detection of significant changes in the
diameter and PSV of both arteries after catheterization,
the compliance levels examined using the Bland-Altman
method showed no difference. The plots showed a sys-
tematic distribution of the differences around zero from
the distribution graphs and a clear relationship between
the differences and the means (fig 1).

Table I. Demographic, clinical and laboratory findings of study patients

All patients RA catheterization UA catheterization p values
(n=328) (n=171) (n=157)
Age (years) 61±17 63±11 59±20 0.077
Gender (F) 174 (53) 89 (52) 85 (54) 0.701
Diagnosis
Acute coronary syndrome 121 (37) 60 (35) 61 (39) 0.312
Stable angina pectoris 177 (54) 94 (54.9) 83 (52.8) 0.879
Miscellaneous 30 (9.1) 14 (8.1) 16 (10.2) 0.183
Hypertension 131 (40) 70 (40.9) 61 (38.8) 0.619
Right arm preprocedure systolic/diastolic BP (mmHg) 134/84 132/83 136/86 0.416
Right arm postprocedure systolic/diastolic BP (mmHg) 135/82 133/81 138/83 0.097
Diabetes mellitus 59 (17.9) 33 (19.2) 26 (16.5) 0.115
Dyslipidemia 128 (39) 71 (41.5) 57 (36.3) 0,092
Smoking 121 (36) 68 (39.7) 53 (33.7) 0.053
Chronic kidney disease 30 (9.1) 13 (7.6) 17 (10.8) 0.182
Procedural time parameters (min)
Arterial access time 4.6 ±2.3 4.4±3.1 4.7±2.9 >0.05
Coronary angiography time 9.3±5.8 8.8±4.3 9.7±6.1 >0.05
Elective PCI time 21.8±.9.1 20.9±11.3 22.7±10.1 >0.05
Total procedural time 31.6±.14.7 31.2±10.9 32.1±13.1 >0.05
Medications (n)
ACE inhibitors/ARBs 135 75 60 0.553
Amlodipin 62 33 29 0.817
Metoprolol 85 47 38 0.376
Diuretics 52 30 22 0.408
Digoxin 19 12 7 0.279
Statin 109 56 53 0.911
The results are expressed as mean±standard deviation or number (%). BP: blood pressure; ACE: angiotensin-converting enzyme; ARBs:
angiotensin II receptor blockers; LDL: low-density lipoprotein; Diuretics; PCI: percutaneous coronary intervention.
 Med Ultrason 2022; 24(1): 52-57 55
Table II. Ultrasonographic measurements of radial and ulnar arteries before and after catheterization
RA catheterization UA catheterization
Pre Post p values Pre Post p values
RA diameter (mm) 2.71±0.66 2.47±0.51 0.007 2.67±0.54 2.76±0.43 0.040
UA diameter (mm) 2.49±0.83 2.59±0.58 0.033 2.53±0.65 2.49±0.77 0.173
RA PSV 44.7±8.3 33.9±9.5 0.021 41.8±10.3 48.6±7.9 0.054
UA PSV 48.3±11.9 59.6±11.0 <0.001 51.2±13.1 44.3±10.7 0.081
UA EDV 9.1±3.0 9.4±2.0 0.481 8.8±1.6 8.5±2.2 0.169
RA PI 4.1±1.9 3.9±1.4 0.713 4.4±0.7 4.3±1.2 0.791
UA PI 4.4±2.7 4.0±2.3 0.059 4.3±1.4 4.7±1.9 0.674
RA RI 0.78±0.09 0.75±0.06 0.177 0.77±0.23 0.80±0.19 0.328
UA RI 0.80±0.11 0.83±0.17 0.269 0.82±0.12 0.79±0.24 0.552
The results are expressed as mean±standard deviation. UA: ulnar artery; RA: radial artery; PSV: peak systolic velocity; EDV: end-diastolic
velocity; PI: pulsatility index; RI: Resistive index

Discussion

In our study, we demonstrated the appearance of

changes in the arterial diameter and blood flow veloc-
ity in both RA and UA after catheterization. In addition,
while the diameter and PSV decreased in the catheterized
artery, the diameter and PSV of the non-catheterized ar-
tery of the same arm increased.
Although the catheterization process is the same for
both arteries, the effect of compression on the catheter-
ized and ipsilateral non-catheterized artery may be dif-
ferent. Compression of the deeply located UA is more
difficult; therefore, the pressure of the bandage is kept
higher for better compression of the UA. The flow veloc-
ity of the adjacent artery increased less (compared to RA
catheterization) due to the less reduction in UA diameter
and flow velocity. In addition, it may be thought that the
reduction in the diameter and flow velocity of the UA,
which has fewer alpha receptors and less sympathetic
activity and therefore less spasm tendency, may contrib-
ute to this result. The effect of haemostasis attempts after
catheterization may also be less pronounced in the UA
for these reasons. In the RA, which is more easily com-
pressed due to its superficial course, it has more alpha
receptors and a higher tendency to spasm, the diameter
and velocity decreased significantly after catheteriza-
tion. Thus the velocity and diameter of the adjacent UA
increased significantly, similar to previous study reports
[3,11,12].
It was known that RA flow is significantly lower than
UA flow and peripheral resistance at the radial portion of
the vascular bed is significantly higher comparing to the
ulnar portion [5,13,14]. Doscher et al reported that the
mean diameters of the RA and UA did not differ signifi-
cantly; however, PSV of the RA was lower than that of
the UA [15].
Fig 1. The Bland-Altman plots showing the compliance levels
after RA and UA catheterization. The mean and standard devia-
tion of the differences between the diameter (A) and PSV (B)
changes in the RA and the diameter (C) and PSV (D) changes
in the UA are given as ±1.96 SD and the distribution is seen to
be around the mean.
Indeed, as determined in our study, the increase in
UA flow velocity after RA catheterization was more pro-
nounced than that of the RA after UA catheterization.
The higher basal UA flow velocity compared to that of
the RA may contribute to the more pronounced increase
in flow velocity of the UA. In addition, although a di-
ameter decrease in the UA after UA catheterization was
determined, it was not as apparent as the difference after
RA catheterization in the RA. This result may be another
reason why the increase in RA PSV after UA catheteri-
zation was not as marked and significant as the increase
in UA PSV after RA catheterization. Our results were in
line with those of previous studies [2,16], indicating an
increase in the RA diameter and blood flow after tran-
sient UA compression. In our study, the fact that the
56 Yasemin Gunduz et al Doppler ultrasonographic evaluation of radial and ulnar artery diameters and blood flow
RA diameter was greater than that of the UA may have
affected these results. The fact that the basal UA flow
velocity was higher than the basal RA flow velocity (as
in the left arm) should not be ignored since the results
obtained in our study may be related to the diameter and
flow characteristics found during basal measurements (in
other words, already existing).
Considering the data obtained, our study showed a
significant increase in blood flow rates in the UA in cases
with RA catheterization, indicating a functional continu-
ity between the radial and ulnar circulation in both hands.
Changes in the inner diameter of an artery can affect arte-
rial blood flow rates by causing changes in arterial wall
properties. This equation shows a strong and inverse re-
lationship between the blood flow velocity of one artery
and the internal diameter of the other artery in the same
arm. Therefore, the decrease in vessel lumen diameter
of the catheterized artery may be associated with the in-
crease in the flow velocity of the other artery in the same
arm. With RA catheterization, the distal perfusion pres-
sure of the capillary bed decreases, causing an increase
in the pressure gradient between the UA and the capillary
bed, resulting in increase of UA blood flow [17,18].
Arterial cannulation/compression or catheterization
results in a decrease in blood flow and sometimes oc-
clusion of the catheterized artery in the early or late pe-
riods. If the radial/ulnar artery is narrowed or occluded,
distal perfusion of the capillary bed will be decreased and
blood flow in the ipsilateral artery will be increased. In-
creases in the diameter and flow rates of the other arter-
ies in patients developing stenosis or RA occlusion and
stenosis or UA occlusion ensure sufficient blood supply
to the hand and protect the hand against ischemic events.
Significant increases in the diameter, velocity, and vol-
ume flow of the arteries during reciprocal compression/
cannulation or catheterization of the two arteries in dif-
ferent studies, including our study, support this assump-
tion [19,20].
It has been demonstrated that RA cannulation would
decrease RA blood flow immediately after catheter place-
ment and the compensatory increase in UA blood flow
might quickly commence. Significant changes continued
after 24 hours from the beginning of monitoring [18,21].
Zhenxian et al investigated the impact of transradial
coronary procedures on the RA by ultrasonography. The
PSV and diameter of the right RA were measured before,
the first day and the first month after transradial coro-
nary interventions. They found that the diameter of the
RA decreased on the first day after the procedure and
this decrease continued in the control performed 1 month
later [22]. Abe et al found that a slight decrease in RA
diameter continued even 3 months after the transradial
intervention; however, they reported that this decrease
did not reach statistical significance [8]. Madssen et al
found a significant reduction in the right RA diameter
compared to left RA diameter at a control USG examina-
tion 12 months after CAG via the right RA [9].
Our study has some limitations. This was a single-
centre study with a nonrandomized design. Larger, rand-
omized, multi-centre and prospective studies are needed
to identify changes in diameter and blood flow parameters
using DUS in patients undergoing coronary angiography/
PCI to confirm our results. Since all DUS examinations
were realized by the same radiologist, and the examina-
tions were carried out in a short period before and after
the procedure, the intraobserver variation analysis was
not performed in this study. In addition, in our study, only
the early effects of catheterization in patients who un-
derwent PCI (patients are generally discharged within 48
hours after the procedure and it is thought that the effects
of catheterization in the long period may be reversible)
were evaluated.
Conclusion
The data obtained from our study suggest that cath-
eterization (starting with cannulation, ending with the re-
moval of the compression bandage after 4 hours) causes
vascular wall changes in the catheterized artery. In our
opinion, while the procedure causes a decrease in diame-
ter and blood flow velocity through vascular wall chang-
es and arterial compression in the catheterized artery, it
protects the arm against ischemia by causing an increase
in the compensatory diameter and blood flow in the other
artery of the same arm. In addition, the UA can be safely
used as an alternative to the RA catheterization.
Conflict of interest: none
References
1. Kar S. Transulnar cardiac catheterization and percutaneous
coronary intervention: techniques, transradial comparisons,
anatomical considerations, and comprehensive literature
review. Catheter Cardiovasc Interv 2017;90:1126-1134.
2. Kaplanoglu H, Dinc E. Diameters and Flow Characteristics
of Forearm Arteries in the Preoperative Period Using Dop-
pler Ultrasonography in Healthy Individuals. Iran J Radiol
2018;15:e12904.
3. Marchese RM, Geiger Z. Anatomy, Shoulder and Upper
Limb, Forearm Radial Artery. StatPearls Publishing, 2021.
4. Pancholy SB, Heck LA, Patel T. Forearm arterial anatomy
and flow characteristics: a prospective observational study.
J Invasive Cardiol 2015;27:218-221.
5. Kim SY, Lee JS, Kim WO, Sun JM, Kwon MK, Kil HK.
Evaluation of radial and ulnar blood flow after radial artery

cannulation with 20- and 22-gauge cannulae using duplex
Doppler ultrasound. Anaesthesia 2012;67:1138-1145.
6. Tonks AM, Lawrence J, Lovie MJ. Comparison of ulnar
and radial arterial blood-flow at the wrist. J Hand Surg Br
1995;20:240–242.
7. Ruengsakulrach P, Brooks M, Hare DL, Gordon I, Buxton
BF. Preoperative assessment of hand circulation by means
of Doppler ultrasonography and the modified Allen test. J
Thorac Cardiovasc Surg 2001;121:526–531.
8. Abe S, Meguro T, Naganuma T, Kikuchi Y. Change in the
diameter of the radial artery transradial intervention using
a 6 French system in Japanese patients. J Invasive Cardiol
2001;13:573-575.
9. Madssen E, Haere P, Wiseth R. Radial artery diameter and
vasodilatory properties after transradial coronary angiogra-
phy. Ann Thorac Surg 2006;82:1698–1702.
10. Peruga JP, Peruga JZ, Kasprzak JD, et al. Ultrasound evalu-
ation of forearm arteries in patients undergoing percutane-
ous coronary intervention via radial artery access: results of
one-year follow-up. Kardiol Pol 2015;73:502-510.
11. Mason PJ, Shah B, Tamis-Holland JE, et al; American Heart
Association Interventional Cardiovascular Care Committee
of the Council on Clinical Cardiology; Council on Cardio-
vascular and Stroke Nursing; Council on Peripheral Vascu-
lar Disease; Council on Genomic and Precision Medicine.
An Update on Radial Artery Access and Best Practices for
Transradial Coronary Angiography and Intervention in
Acute Coronary Syndrome: A Scientific Statement From
the American Heart Association. Circ Cardiovasc Interv
2018;11:e000035.
12. He GW, Yang CQ. Characteristics of adrenoceptors in the
human radial artery: clinical implications. J Thorac Cardio-
vasc Surg 1998;115:1136-1141.
13. Brzezinski M, Luisetti T, London MJ. Radial artery
cannulation:a comprehensive review of recent anatomic and
physio-logic investigations. Anesth Analg 2009;109:1763–
1781.
14. Zabad MN, Janzer S, George JC. Radial and Ulnar Artery
Occlusion Following Transradial and Transulnar Coronary
Intervention Requiring Endovascular Revascularization.
Cath Lab Digest 2019;27:6.
Med Ultrason 2022; 24(1): 52-57 57
15. Doscher W, Viswanathan B, Stein T, Margolis IB. Hemo-
dynamic assessment of the circulation in 200 normal hands.
Ann Surg 1983;198:776–779.
16. Yilmaztepe MA, Yilmaz E. Effect of transient ulnar ar-
tery compression on radial artery diameter. Exp Ther Med
2018;16:3735-3739.
17. Beniwal S, Bhargava K, Kausik SK. Size of distal radial
and distal ulnar arteries in adults of southern Rajasthan and
their implications for percutaneous coronary interventions.
Indian Heart J 2014;66:506-509.
18. Kim EJ, Soh S, Kim SY, et al. Impact of Diabetes Mellitus
on Radial and Ulnar Arterial Vasoreactivity after Radial Ar-
tery Cannulation: A Randomized Controlled trial. Int J Med
Sci 2016;13:701-707.
19. Saito S, Ikei H, Hosokawa G, Tanaka S. Influence of the
ratio between radial artery inner diameter and sheath outer
diameter on radial artery flow after transradial coronary in-
tervention. Catheter Cardiovasc Interv 1999;46:173-178.
20. Hahalis G, Aznaouridis K, Tsigkas G, et al. Radial Artery
and Ulnar Artery Occlusions Following Coronary Pro-
cedures and the Impact of Anticoagulation: ARTEMIS
(Radial and Ulnar ARTEry Occlusion Meta-Analys IS)
Systematic Review and Meta-Analysis. J Am Heart Assoc
2017;6:e005430.
21. Brodman RF, Hirsh LE, Frame R. Effect of radial artery
harvest on collateral forearm blood flow and digital perfu-
sion. J Thorac Cardiovasc Surg 2002;123:512-516.
22. Yan Z, Zhou Y, Zhao Y, Zhou Z, Yang S, Wang Z. Impact of
transradial coronary procedures on radial artery. Angiology
2010;61:8-13.
23. Huzjan R, Brkljacić B, Delić-Brkljacić D, Biocina B, Sutlić
Z. B-mode and color Doppler ultrasound of the forearm ar-
teries in the preoperative screening prior to coronary artery
bypass grafting. Coll Antropol 2004;28 Suppl 2:235-241.
24. Ashraf T, Panhwar Z, Habib S, Memon MA, Shamsi F, Arif
J. Size of radial and ulnar artery in local population. J Pak
Med Assoc 2010;60:817-819.
25. Yan ZX, Zhou YJ, Zhao YX, Zhou ZM, Yang SW, Wang
ZJ. Anatomical study of forearm arteries with ultrasound
for percutaneous coronary procedures. Circ J 2010;74:686-
692.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 58-64
DOI: 10.11152/mu-3088

Comparison of the effects of adenosine, isoproterenol and their

combinations on pulmonary transit time in rats using contrast
echocardiography
Feng Su1, Yun-Yan Shi1, Bo Wang1, Xiao-Zhi Zheng1,2

1Department of Ultrasound, Yancheng Clinical College of Xuzhou Medical University, Yancheng, Jiangsu Province,
2Department of Ultrasound, Yangpu Hospital, Tongji University School of Medicine, Shanghai, China

Abstract
Aims: To compare the effects of adenosine (Ade), isoproterenol (Iso) and their combinations on pulmonary transit time
(PTT) in rats using contrast echocardiography. Material and methods: Thirty-two adult Sprague Dawley (SD) rats were
divided into four groups (n=8) according the medicines of tail-intravenous injection: Group 1, control; Group 2, Ade; Group
3, Iso; Group 4, Ade+Iso. They all underwent conventional echocardiography and contrast echocardiography with measure-
ments of PTT. Results: With Ade injection, OnsetRV-OnsetLV PTT (PTT1), PeakRV-PeakLV PTT (PTT2) and OnsetRV-PeakLV
PPT (PTT3) decreased and PTT3 had the largest decreased percentage, with the highest performance in differentiating the Ade
group from the control group [the area under receiver operating characteristic curve (AUC), sensitivity and Youden’s index
was maximal]. With Iso injection, PTT1, PTT2 and PTT1 all increased and PTT1 had the largest increased percentage, with
the highest performance in differentiating the Iso group from the control group (AUC, sensitivity and Youden’s index was
maximal). With a combination injection of Ade and Iso, the PTT values were similar to the control group and no PTT could
differentiate the Ade+Iso group from the control group. Conclusions: Ade or/and Iso exerted distinct effects on PTT. These
findings remind us that it is a necessary to consider the effects of medicine (especially cardiopulmonary vasoactive drugs) on
the PTT values. At the same time, it provides the basis for the clinical transformation of consecutive Iso/Ade treatment from
the perspective of pulmonary circulation.
Keywords: pulmonary transit time; adenosine; isoproterenol; contrast echocardiograghy; rat

Introduction evaluation of cardiopulmonary function [1-8]. It is corre-

Pulmonary transit time (PTT), defined as the time re-
quired for a volume of contrast to travel from the right
ventricle (RV) to the left atrium (LA) or the left ventri-
cle (LV), has been reported to be a metric for integrative
Received 15.02.2021  Accepted 16.06.2021
Med Ultrason
2022, Vol. 24, No 1, 58-64
Corresponding author: Xiao-Zhi Zheng, M.D, Ph.D
Department of Ultrasound, Yancheng Clinical
College of Xuzhou Medical University,
166 West Yulong Road, Yancheng 224005,
Jiangsu Province, and Department of Ultrasound,
Yangpu Hospital, Tongji University School of
Medicine, 450 Tengyue Road,
Shanghai 200090, People’s Republic of China
E-mail: zxzshyzx@126.com
Phone/fax: 021-65690520-216
021-65690520-813
lated with RV systolic and diastolic function, pulmonary
vascular status (pulmonary vascular resistance) and left
ventricular (LV) systolic and diastolic function. To date,
PTT values have been measured by a variety of imaging
methods, including contrast echocardiography, for most
species, from cats and dogs to humans, with a high reli-
ability, repeatability and accuracy [1-9]. But, the values
of PTT in rats have not been reported.
It is well known that any cardiopulmonary medica-
tion may produce an effect on cardiopulmonary function
and pulmonary vascular status, which may further result
in the changes of PTT. Adenosine (Ade) and isoproter-
enol (Iso) are two different kinds of cardiopulmonary
medication. Studies have shown that consecutive treat-
ment of isolated heart with a high dose of Iso and Ade
(Iso/Ade treatment) confers strong protection against is-
chaemia/reperfusion injury during heart surgery [10,11].

Regarding the kind of change of PTT they bring about in
vivo remains unknown. In this paper the effects of Ade,
Iso and their combinations on PTT in a rat model using
contrast echocardiography were compared.
Materials and methods
Contrast agent, adenosine and isoproterenol
preparation
A commercially available second-generation contrast
agent, SonoVue® phospholipid-shell sulfur hexafluoride
microbubbles (Bracco, Milan, Italy) was used for contrast
echocardiography. A total of 59 mg of SonoVue® was
diluted in 5 mL of saline according to the manufactur-
er’s protocol. Adenosine dry powder (Sigma, Shanghai,
China) was diluted with normal saline to 0.46 uM/mL,
and isoproterenol solution (Harvest Pharma, Shanghai,
China) was diluted with normal saline to 2.4 uM/mL.
Equipment
The conventional and contrast echocardiographic
data were acquired with a commercially available ultra-
sound system (Mindray M9, Mindray Medical Systems,
Shenzhen, China) equipped with a L12-4s transducer
(4-8 MHz), using a multi-pulse contrast-specific imag-
ing protocol (power modulation imaging) combined with
pulse inversion power Doppler.
Animal preparation and grouping
Thirty-two adult SD rats (male, 12~15 weeks old,
weighing 250~400g; SLACCAS, Shanghai, China) were
enrolled in the present study, with the approval of Ani-
mal Care and Use Committee, Xuzhou Medical Univer-
sity and Tongji University School of Medicine. These
rats were divided into four groups (n=8) according to the
medicines of tail-intravenous injection: Group 1 received
0.5 ml of sterile pyrogen-free normal saline rapidly in-
jected within 3 s at first and then 0.5 ml microbubbles
maintained (0.25 mL/min) for 2 minutes (control group);
Group 2 received 0.5 ml adenosine solution (0.23 uM)
rapidly injected within 3 s at first and then 0.5 ml micro-
bubbles maintained (0.25 mL/min) for 2 minutes (Aden
group); Group 3 received 1 ml isoproterenol solution (2.4
uM) rapidly injected within 3 s at first and then 0.5 ml
microbubbles maintained (0.25 mL/min) for 2 minutes
(Iso group); and Group 4 received 0.5 ml adenosine solu-
tion (0.23 uM)+1 ml isoproterenol solution (2.4 uM) con-
secutively within 3 s at first and then 0.5 ml microbub-
bles maintained (0.25 mL/min) for 2 minutes (Aden+Iso
group).
2D ultrasound examination
SD rats were anesthetized by intraperitoneal injection
of 10% chloral hydrate (350 mg/kg body weight). The
echocardiographic study was performed in the supine
Med Ultrason 2022; 24(1): 58-64 59
position. From parasternal short axis view, left ventricu-
lar maximal diameter at end-diastole and end-systole
(LVd and LVs), right ventricular diameter at end-diastole
and end-systole (RVd and RVs), interventricular sep-
tum thickness at end-diastole and end-systole (IVSTd
and IVSTs), right ventricular free wall thickness at end-
diastole and end-systole (RVFWTd and RVFWTs) were
measured. The interventricular septum systolic thicken-
ing rate (IVSSTR), right ventricular free wall systolic
thickening rate (RVFWSTR) was calculated and left
ventricular ejection fraction (LVEF) was obtained by real
time 2D echo method.
Real time contrast echocardiography
All examinations were done by a single physician
experienced with contrast echocardiography. Examina-
tion was performed from the parasternal short axis view.
Gain, depth, transmit focus and post-processing were op-
timized at the beginning of the study and held constant
throughout. The optimal balance between myocardial
contrast enhancement and attenuation in our setting was
achieved at a low mechanical index of 0.086 (acoustic
power 7.8%, thermal index of soft tissue 0.0). The whole
process of ventricular enhancement, i.e., from the gradu-
ally visible opacification to the plateau, then to the at-
tenuation, about 120s, was stored digitally.
PTT measurement
Image analysis was performed off-line by two ex-
perienced operators (fig 1). The clip containing contrast
agent entry and passage through the heart was carefully
reviewed. Three kinds of PTT were calculated. OnsetRV-
OnsetLV PTT (PTT1) was calculated as the number of
seconds from the first frame on which the contrast agent
started to fill right ventricular to the first frame on which
the contrast agent reached the left ventricular by dividing
this frame count by the frame rate. PeakRV-PeakLV PTT
(PTT2) was calculated as the number of seconds from
the time of the peak opacification in the right ventricle
to the time of the peak opacification in the left ventricle
determined by the time-intensity curves of contrast echo-
cardiography. OnsetRV-peakLV PTT (PTT3) was calculat-
ed as the number of seconds from the time of the initial
appearance of contrast agent in the right ventricle to the
peak opacification in the left ventricle determined by the
time-intensity curves of contrast echocardiography.
Intraobserver and interobserver variability for iden-
tification of PTT1, PTT2 and PTT3 were obtained by
repeated, blinded analysis of 10 randomly selected rats
after a minimum time interval of 2 weeks.
Statistical analysis
Results are expressed as mean ± standard deviation.
The differences among groups were tested using one-
way ANOVA. A receiver operating characteristic curve
60 Feng Su et al Effects of Ade, Iso and their combinations on pulmonary transit time in rats using contrast echocardiography

Fig 1. The time-intensity curves of contrast echocardiography of the control (a), adenosine (b), isoproterenol (c) and adenosine+
isoproterenol (d) group.

(ROC) analysis of was used to compare the performance
of PTT based on different calculation methods in differ-
entiating Aden or/and Iso group from control subjects,
determine the optimal cut-off points and validity param-
eters. Bland & Altman was used to measure the inter/
intra observer variability. A value of p<0.05 was consid-
ered statistically significant. All statistical analysis was
performed with SPSS version 16 software for Windows
(SPSS Inc, Chicago, IL).
Results
Tolerability and contrast echocardiography quality
All rats could tolerate the dosage of Ade or/and Iso
injected through the tail vein. No adverse reactions were
observed. Using the high frequency transducer and the
dosage of Sonovue mentioned above, good image quality
was achieved in all rats.
Echocardiographic and hemodynamic
characteristics
As shown in Table I, with Ade injection, heart rate
increased significantly. Compared to the control group,
the systolic cardiac chambers became smaller, the myo-
cardium became thinner, systolic thickening rate became
greater and LVEF increased significantly. With Iso injec-
tion, the heart rate increased more significantly (nearly
500 bpm). Compared to the control group, the systolic
cardiac chambers became even smaller, the myocardium
thickened at the end of systole and the LVEF increased
more significantly. But, the RVd and the difference be-
tween RVd and RVs became smaller. In addition, systolic
thickening rate in the Iso group was slightly lower than in
the Ade group. With a combination injection of Ade and
Iso, there was no change in heart rate. Compared to the
control group, the cardiac size, myocardium thickness,
systolic thickening rate, LVEF and RV function did not
change significantly, too.
PTT values
As shown in figure 2, whether it was PTT1,PTT2 or
PTT3, with the Ade injection, they all significantly de-
clined (PTT1: from 0.99±0.39 s to 0.72±0.26 s, p<0.05;
PTT2: from 4.77±2.39 s to 2.44±0.95 s, p<0.01; PTT3:
from 32.71±9.31 s to 16.59±5.23 s, p<0.001). With the
Iso injection, however, they all significantly length-
ened (PTT1: from 0.99±0.39 s to 2.87±0.59 s, p<0.001;
PTT2: from 4.77±2.39 s to 9.23±5.71 s, p<0.01; PTT3:

Table I. Comparison of echocardiographic parameters in control, adenosine (Ade), isoproterenol (Iso) and adenosine and isoproter-
enol combined treatment (Ade+Iso) groups
Variables Control Ade Iso Ade+Iso
HR, bpm 402.20±9.26 447.00±7.75** 503.67±3.21**## 409.40±7.56
RVd, mm 3.17±0.65 3.01±0.22 1.77±0.15**## 3.22±0.58
RVs, mm 2.37±0.15 1.70±0.36** 1.53±0.06 2.34±0.19
RVFWTd, mm 0.67±0.06 0.30±0.10** 0.53±0.06*## 0.64±0.07
RVFWTs, mm 0.90 ±0.11 0.53±0.06** 0.97±0.06## 0.87±0.13
RVFWSTR, % 35.72±18.89 91.67±62.91 82.22±16.17**# 35.97±19.14
LVd, mm 6.30±0.10 6.10±0.09* 6.23±0.25 6.33±0.14
LVs, mm 5.65± 0.05 4.07±0.40** 3.30±0.02**## 5.59± 0.07
LVEF (%) 63.00±1.00 80.01±4.24** 92.50±2.12**## 64.35±1.98
IVSTd, mm 1.20±0.10 0.73±0.15** 1.13±0.12*## 1.18±0.14
IVSTs, mm 1.73±0.06 1.33±0.06** 2.03±0.09*## 1.76±0.09
IVSSTR, % 45.36±16.74 85.98±30.56** 78.83±27.74**# 45.78±15.92
*p<0.05, **p<0.01 vs. control group; #p<0.05, ##p<0.01 vs. Aden group. Data are expressed as mean±standard deviation. Bpm, beat per
minute; d, at end-diastole; HR, heart rate; IVSSTR, interventricular septum systolic thickening rate; IVST, interventricular septum thickness;
LV, left ventricular; RV, right ventricular; RVFWSTR, right ventricular free wall systolic thickening rate; RVFWT, right ventricular free wall
thickness; s, at end-systole
 Med Ultrason 2022; 24(1): 58-64 61

Fig 2. Comparison of pulmonary transit times among different groups. * p<0.05, ** p<0.01, # p<0.001, vs. the control group. PTT1,
OnsetRV-OnsetLV pulmonary transit time (the time required for a volume of contrast to travel from the initial appearance in the right
ventricle to the initial appearance in the left atrium or left ventricle) determined by reviewing the contrast echocardiography in a
“frame-by-frame” manner; PTT2, PeakRV-PeakLV pulmonary transit time (the time required for a volume of contrast to travel from
the peak opacification in the right ventricle to the peak opacification in the left ventricle determined by the time-intensity curves of
contrast echocardiography); PTT3, OnsetRV-PeakLV pulmonary transit time (the time required for a volume of contrast to travel from
the initial appearance in the right ventricle to the peak opacification in the left ventricle) determined by the time-intensity curves of
contrast echocardiography.

from 32.71±9.31 s to 37.20±10.57 s, p<0.05). Interest-
ingly, with a combination injection of Ade and Iso, all
the PTT values were the same as the control group’s
(PTT1:1.02±0.43 s vs.0.99±0.39 s, p>0.05; PTT2:
4.73±2.51s vs. 4.77±2.39 s, p>0.05; PTT3: 32.94±9.35 s
vs. 32.71±9.31 s, p>0.05).
ROC curve  analysis
Comparison of the performance of PTT based on dif-
ferent calculation methods in differentiating the Ade or/
and Iso group from the control subjects, in other words,
which one was more susceptible to Ade or/and Iso among
PTT1, PTT2 and PTT3, was shown in Table II. In the
Ade group, the AUC, sensitivity and Youden’s index
were maximal in PTT3, i.e, Ade is most likely to cause
changes in PTT3. In the Iso group, the AUC, specific-
ity and Youden’s index were all were maximal in PTT1:
therefore, Iso is the most likely to cause changes in PTT1.
In the Ade+ Iso group, each AUC was less than 0.5, i.e,
the combined administration of Ade and Iso could not
cause changes in any PTT.
Reproducibility
Bland-Altman analysis showed the interobserver and
intraobserver agreement for the measurement of PTT
values were very good: intraclass correlation coefficients
were all >0.97 for interobserver and intraobserver vari-
ability.

Table II. Comparison of the performance of pulmonary transit time based on different calculation methods in differentiating adeno-
sine (Ade) or/and isoproterenol (Iso) from control subjects
Medicines AUC Sensitivity (%) Specificity (%) Youden’s index Cutoff value
Ade
PTT1 0.750(0.490-1.010) 71 75 0.46 0.835
PTT2 0.768(0.510-1.026) 71 75 0.46 3.19
PTT3 0.921(0.820-1.022)* 100* 81.4* 0.81* 31.78
Iso
PTT1 0.958(0.858-1.059) 83.3 100* 0.833* 1.44
PTT2 0.792(0.527-1.056) 100* 75 0.75 8.41
PTT3 0.500(0.176-0.824) 50 75 0.25 38.78
Ade+Iso
PTT1 0.422(0.103-0.741) 66.7 35.3 0.02 1.23
PTT2 0.438(0.127-0.748) 66.7 37.5 0.04 4.97
PTT3 0.375(0.06-0.690) 66.7 25 -0.07 12.28
*p<0.05, vs. the same validity parameters in the same medicines group. AUC, areas under receiver operating characteristic curve;
PTT1,OnsetRV-OnsetLV pulmonary transit time (the time required for a volume of contrast to travel from the initial appearance in the right
ventricle to the initial appearance in the left atrium or left ventricle) determined by reviewing the contrast echocardiography in a“frame-by-
frame” manner; PTT2, PeakRV-PeakLV pulmonary transit time (the time required for a volume of contrast to travel from the peak opacifica-
tion in the right ventricle to the peak opacification in the left ventricle determined by the time-intensity curves of contrast echocardiography);
PTT3, OnsetRV-peakLV pulmonary transit time (the time required for a volume of contrast to travel from the initial appearance in the right
ventricle to the peak opacification in the left ventricle) determined by the time-intensity curves of contrast echocardiography.
62 Feng Su et al Effects of Ade, Iso and their combinations on pulmonary transit time in rats using contrast echocardiography
Discussion
This is the first report on the effects of Ade, Iso and
their combinations on PTT in rats using contrast echocar-
diography. The results presented here indicate that Ade
or/and Iso exerted distinct effects on PTT: Ade shortened
PTT, Iso lengthened PTT and the combinations of Ade
and Iso had no significant effect on PTT. Moreover, Ade
had a greater effect on PTT3 while Iso had a greater ef-
fect on PTT1. These findings suggest that we need to
consider the effects of medicine (especially cardiopul-
monary vasoactive drugs) on the PTT values during their
usage. It also provides the basis for the translation into
clinical practice of consecutive Iso/Ade treatment from
the perspective of pulmonary circulation.
Although PTT is defined as the time required for a
volume of contrast to travel from RV to LA or LV. The
concept is still not very clear now, because it does not
specify when (initial or peak opacification?) starts in
order to time when the contrast agent reaches the heart
cavity (RV, LA or LV). Most studies take the time re-
quired for a volume of contrast to travel from RV peak
opacification to LA (or LV) peak opacification as PTT,
i.e., PeakRV-PeakLV PTT (PTT2) in our study. There are
also some studies that use the time required for a volume
of contrast to travel from RV initial appearance to LA (or
LV) the initial appearance as PTT, i.e., OnsetRV-OnsetLV
PTT (PTT1) in our study. There is another calculation of
PTT: the time required for a volume of contrast to travel
from RV initial appearance to LA (or LV) peak opacifica-
tion, i.e. OnsetRV-PeakLV PTT (PTT3) in our study. PTT1
is believed to mainly reflect the pulmonary vascular sta-
tus, such as intrapulmonary vasodilatation, pulmonary
arteriovenous fistula and abnormal angiogenesis, directly
connected arteries-veins, and so on. Our previous study
confirmed PTT1 is a reliable marker for the differential
diagnosis of pulmonary nodules [9]. PTT3 is believed to
a very important potential parameter for the calculation
of pulmonary vascular volume although some studies
have used PTT2 rather than PTT3 as a reference to calcu-
late it [7,12]. Therefore, in this study, three kinds of PTT
were calculated for comparison.
Ade and Iso, in addition to routine clinical use, also
used to mimic the potent cardioprotection of temperature
preconditioning by consecutive activation of protein ki-
nases A and protein kinases C [10,11]. Moreover, Iso and
Ade, alone or simultaneously, protected isolated rat hearts
but the consecutive treatment gave the highest protection
[13]. Such a strategy seems to be useful in cardiac sur-
gery involving ischaemic cardioplegic arrest and cardio-
pulmonary bypass. However, in vivo, it may also bring
about impacts on cardiopulmonary function and pulmo-
nary vascular status, and then result in the PTT changes.
In this study, the rats were divided into 4 groups of con-
trol, Ade, Iso, and Aden+Iso just to explore their effect on
PTT in detail. The results showed that a shortened PTT
was found in the Ade group, a lengthened PTT in the Iso
group and an unchanged PTT in the Ade+Iso group. Its
underlying mechanisms are not fully known and the possi-
ble mechanisms involved require to be further discussed.
Firstly, Ade and Iso, alone or simultaneously, have
different effects on cardiac function. Through 2D echo
examination, it was found that the ability of Ade to en-
hance RV contraction (represented by RVFWSTR and
the difference of RVd and RVs) was greater than that of
Iso, with a normal LV diastolic volume (represented by
LVd) and RV diastolic volume (i.e., the volume of sys-
temic circulation back to the heart; represented by RVd).
Enhanced right ventricular systolic function, normal RV
diastolic volume and normal LV diastolic volume (equiv-
alent to normal left atrial pressure) would result in a
shortened PTT. The ability of Iso to enhance LV contrac-
tion (represented by LVEF and the difference of LVd and
LVs) was greater than that of Ade, with a normal LV di-
astolic volume (represented by LVd), but with a reduced
RV contractility (represented by the difference of RVd
and RVs), a decrease in RV diastolic volume (represented
by RVd) and with a thicker myocardium and impaired
systolic thickening rate. Reduced right ventricular systol-
ic function, decreased RV diastolic volume (equivalent to
RV preload) and normal LV diastolic volume would re-
sult in a lengthened PTT. Compared to the control group,
the combination of Ade and Iso had no significant effect
on cardiac size, cardiac function and so on. Therefore, no
changes were seen in PPT.
Secondly, Ade and Iso, alone or simultaneously, have
different effects on pulmonary vascular status. In our
study, Ade and Iso were all administered intravenously
through the tail vein. Ade acted as a potent selective pul-
monary vasodilator by increasing intracellular cAMP via
A2 receptor agonism [14] because of its rapid endothelial
metabolism [15]. Pulmonary angiectasia would result in
a decreased pulmonary vascular volume and a shortened
PTT. Iso, a non-selective β-adrenergic agonist, acting on
the β2 receptor of the bronchial smooth muscle, caused
a strong relaxation effect on the bronchial smooth mus-
cle, and relived bronchial spasm. However, the bronchial
smooth muscle relaxation made airway resistance re-
duced, the latter further made ventilation perfusion ratio
abnormal and hypoxemia aggravated, and finally caused
a hypoxic pulmonary vasoconstriction. In this setting,
PTT was significantly prolonged. As for the effects of
Ade+Iso on pulmonary vascular status, they might cancel
each other out, so PTT did not change.

In our previous experimental studies (unpublished
data), the dose of Ade and Iso was explored, and the dose
gradient (Ade: 0.10, 0.23, 0.46 and 0.92 uM/mL; Iso: 1.2,
2.4, 4.8, 9.6 uM/mL) had been done, and each dose of
Ade and Iso had similar effects on PTT. In this study, the
optimal concentration (Ade: 0.46 uM/mL; Iso: 2.4 uM/
mL) was adopted. Our previous studies had explored the
modalities of administration of Ade and Iso, too. Ade ad-
ministered via the jugular vein was complicated and had
a low success rate. Although intraperitoneal injection of
Iso had a high success rate; it couldn’t be administered
simultaneously with Ade. Therefore, simultaneous ad-
ministration of Ade and Iso through tail vein was the best
option. Due to the small size and fast circulation of rats,
Ade may be metabolized less than we think.
Even so, our study had several limitations that should
be considered. First of all, doses of Ade and Iso used in
this study were selected according to our previous stud-
ies, without reference to those in vitro studies (sequen-
tially with 5 nM Iso and 30 μM Ade) [10,11]. Further
systematic exploration is required. Next, Ade and Iso in
this study were all administered through tail-intravenous
injection rather than subcutaneous injection, intraperito-
neal injection, heart cavity injection, pulmonary vascular
injection and bronchial injection, etc. The conclusions
about the effects of Ade and Iso, alone or simultaneously
on PTT may be one-sided. Finally, we still need more
systematic researchs, including pathology at least, to sup-
port our idea.
Conclusion
In this study, the effects of Ade, Iso and their combina-
tions on PTT in a rat model using contrast echocardiogra-
phy was compared. The findings demonstrated that Ade
or/and Iso exerted distinct effects on PTT: Ade shortened
PTT, Iso lengthened PTT, and the combinations of Ade
and Iso had no significant effect on PTT. Although it had
some limitations, this study still reminds us of the need
to consider the effects of medicine (especially cardiopul-
monary vasoactive drugs) on the PTT values. At the same
time, it provides the basis for the clinical transformation
of consecutive Iso/Ade treatment from the perspective of
pulmonary circulation.
Acknowledgements: The authors gratefully ac-
knowledge the assistance of Xiancheng Xia from the
Department of Ultrasound, The Yancheng Clinical Col-
lege of Xuzhou Medical University, Yancheng, Jiangsu
Province, P.R. China.
Conflict of interest: none
Med Ultrason 2022; 24(1): 58-64 63
References
1. Brittain EL, Doss LN, Saliba L, Irani W, Byrd BF 3rd, Mo-
nahan K. Feasibility and diagnostic potential of pulmonary
transit time measurement by  contrast echocardiography: a
pilot study. Echocardiography 2015;32:1564-1571.
2. Herold IHF, Saporito S, Bouwman RA, et al. Reliability, re-
peatability, and  reproducibility  of  pulmonary transit  time
assessment by  contrast  enhanced echocardiography. Car-
diovasc Ultrasound 2016;14:1. 
3. Monahan K, Coffin S, Lawson M, Saliba L, Rutherford R,
Brittain E. Pulmonary transit time from contrast echocardi-
ography and cardiac magnetic resonance imaging: Compar-
ison between modalities and the impact of region of interest
characteristics. Echocardiography 2019;36:119-124.
4. Zhao H, Tsauo J, Zhang X, et al. Pulmonary transit time
derived from pulmonary angiography for the diagnosis of
hepatopulmonary syndrome. Liver Int 2018;38:1974-1981.
5. Crosara S, Ljungvall I, Margiocco ML, Häggström J, Tar-
ducci A, Borgarelli M. Use of contrast echocardiography
for quantitative and qualitative evaluation of myocardial
perfusion and pulmonary transit time in healthy dogs. Am J
Vet Res 2012;73:194-201.
6. Streitberger A, Hocke V, Modler P. Measurement of pul-
monary transit time in healthy cats by use of ultrasound
contrast media “Sonovue”: Feasibility, reproducibility, and
values in 42 cats. J Vet Cardiol 2013;15:181-187.
7. Ali LA, Aquaro GD, Peritore G, et al. Cardiac magnetic
resonance evaluation of pulmonary transit time and blood
volume in adult congenital heart disease. J Magn Reson Im-
aging 2019;50:779-786.
8. Colin GC, Pouleur AC, Gerber BL, et al. Pulmonary hy-
pertension detection by computed tomography pulmonary
transit time in heart failure with reduced ejection fraction.
Eur Heart J Cardiovasc Imaging 2020;21:1291-1298.
9. Wang B, Sun F, Zheng XZ, Sun CY. A novel application of pul-
monary transit time to differentiate between benign and ma-
lignant pulmonary nodules using myocardial contrast echo-
cardiography. Int J Cardiovasc Imaging 2021;37:1215-1223.
10. Lewis M, Szobi A, Balaska D, et al. Consecutive Isoproter-
enol and Adenosine Treatment Confers Marked Protection
against Reperfusion Injury in Adult but Not in Immature
Heart: A Role for Glycogen. Int J Mol Sci 2018;19:494.
11. Khaliulin I, Halestrap AP, Bryant SM, Dudley DJ, James
AF, Suleiman MS. Clinically-relevant consecutive treat-
ment with isoproterenol and adenosine protects the fail-
ing heart against ischaemia and reperfusion. J Transl Med
2014;12:139.
12. Galanti G, Jayaweera AR, Villanueva FS, Glasheen WP,
Ismail S, Kaul S. Transpulmonary transit of microbubbles
during contrast echocardiography: implications for estimat-
ing cardiac output and pulmonary blood volume. J Am Soc
Echocardiogr 1993;6:272-278.
13. Khaliulin I, Parker JE, Halestrap AP. Consecutive pharma-
cological activation of PKA and PKC mimics the potent
cardioprotection of temperature preconditioning. Cardio-
vasc Res 2010;88:324-333.
64 Feng Su et al Effects of Ade, Iso and their combinations on pulmonary transit time in rats using contrast echocardiography
14. Aranda M, Bradford KK, Pearl RG. Combined therapy 15. Morgan JM, McCormack DG, Griffiths MJ, Morgan CJ,
with inhaled nitric oxide and intravenous vasodilators dur- Barnes PJ, Evans TW. Adenosine as a vasodilator in pri-
ing acute and chronic experimental pulmonary hyperten- mary pulmonary hypertension. Circulation 1991;84:1145-
sion. Anesth Analg 1999;89:152-158. 1149.
Review Med Ultrason 2022, Vol. 24, no. 1, 65-76
DOI: 10.11152/mu-3323

Ultrasound of the chest and mediastinum in children, interventions

and artefacts. WFUMB review paper (part 3)
Cheng Fang1, Joanna Jaworska2, Natalia Buda3, Ioana Mihaiela Ciuca4, Yi Dong5, Axel
Feldkamp6, Jörg Jüngert7, Wojciech Kosiak8, Hans Joachim Mentzel9, Corina Pienar4,
Jorge S. Rabat10, Vasileios Rafailidis1, Simone Schrading11, Dagmar Schreiber-Dietrich12,
Christoph F Dietrich13,14

1Department of Radiology, King’s College Hospital, London, United Kingdom, 2Institute of Mother and Child, Cystic
Fibrosis Department, Warszawa, Poland, 3Internal Medicine, Connective Tissue Diseases and Geriatrics Department,
Medical University of Gdansk, Poland, 4Department of Pediatrics, University of Medicine and Pharmacy “Victor
Babes” Timisoara, Romania, 5Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai, China,
6Pediatric Department, Sana Kliniken Duisburg GmbH, Germany, 7Department of Pediatrics and Adolescent Medi-

cine, University Hospital Erlangen, Germany, 8Pediatric, Hematology and Oncology Department, Medical University
of Gdansk, Poland, 9Section of Pediatric Radiology, Institute of Diagnostic and Interventional Radiology, University
hospital Jena, Germany, 10Head of Surgery Department. Universidad de Oriente. Ciudad Bolívar City. Bolivar State.
Venezuela, 11Klinik für Radiologie und Nuklearmedizin, Luzerner Kantonsspital, Switzerland, 12Localinomed, Bern,
Switzerland, 13Department Allgemeine Innere Medizin (DAIM), Kliniken Hirslanden Beau Site, Salem und
Permanence, Bern, Switzerland, 14Sino-German Research Center of Ultrasound in Medicine, The First Affiliated
Hospital of Zhengzhou University, Zhengzhou, China

Abstract
Ultrasound (US) is an ideal diagnostic tool for paediatric patients owning to its high spatial and temporal resolution, real-
time imaging, and lack of ionizing radiation and bedside availability. The lack of superficial adipose tissue and favourable
acoustic windows in children makes US the first line of investigation for evaluation of pleural and chest wall abnormalities.
In the first part of the topic the technical requirements were explained and the use of ultrasound in the lung and pleura
in paediatric patients were discussed. In the second part lung parenchymal diseases with their subpleural consolidations are
reflected. In the third part, the use of ultrasound for chest wall, mediastinum, diaphragmatic diseases, trachea, interventions
and artifacts in paediatric patients are summarized.
Keywords: lung; chest; mediastinum; guideline; CEUS

In the first part of the “Lung ultrasound in children”
topic the technical requirements were explained and the
use of ultrasound (US) in the lung and pleura in paedi-
atric patients were discussed [1]. In the second part, the
Received 05.03.2021  Accepted 21.05.2021
Med Ultrason
2022, Vol. 24, No 1, 65-76
Corresponding author: Prof. Dr. med. Christoph F. Dietrich, MBA
Department of Internal Medicine (DAIM)
Kliniken Hirslanden Bern, Beau Site,
Salem and Permanence
Schänzlihalde 11, 3031 Bern, Switzerland
E-mail: c.f.dietrich@googlemail.com
lung parenchymal diseases with their subpleural consoli-
dations were reflected [2].
Herewith, in the third part the use of US for chest wall,
mediastinum, diaphragmatic diseases, trachea, interven-
tions and artifacts in paediatric patients are summarized.
CHEST WALL
Examination technique
The superficially located structures of the thoracic
wall close to the US transducer can be optimally exam-
ined with high-frequency linear probes (7-20 MHz). Ex-
66 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)

amining both sides of the thoracic wall is necessary to 3. Benign chest wall tumours
establish symmetry. The normal bone cortex appears as Benign solid masses of the chest wall are enumerated
a uniformly hyperechoic structure with smooth margins, in Table I and figures 1-3 [3,6,8,9]. Lymph nodes are usu-
while costal cartilage is hypoechoic [3,4]. ally identified on US as hypoechoic solid structures with
Developmental abnormalities of the sternum include an echogenic fatty hilum with preserved architecture of
sternal tilt, pectus excavatum and carinatum [3,5]. Nor- blood vessels in the hilum [3,6]. In contrast, lipomas are
mal rib variants such as prominent anterior convex ribs, well defined, echogenic masses, with minimal Doppler
hypertrophic and/or wavy ribs, bifid ribs and asymmetric flow [6]. Desmoids are infiltrative hypoechoic, fibrillar
costochondral junctions may also be identified [3,5-7]. tumours with muscular fascial involvement [9].
When evaluating a costochondral mass, the advantage of
US over computed tomography (CT) or magnetic reso-
nance imaging (MRI) is that it provides a more focused
view [7]. Pathological findings should be imaged in two
planes. Ribs are examined in their oblique presentation.
In the current review breast examination is not included.
Indication and pathological findings
The main indication for chest wall US is a palpable
mass in the thoracic region. Non-tender masses are usual-
ly benign, and US may provide a definitive diagnosis [6].
1. Rib and clavicle fractures Fig 1. Clinical presentation: boy, 8 months, tumour above the
US is not the main imaging modality when evaluat- sternum (a). Transverse scan: nearly anechoic tumour (arrow)
ing rib or clavicle fractures, with radiography being the above the sternum (star): epidermal cyst (b).
first line modality, but this is controversial. Radiographic
signs may not be evident, when minor displacement oc-
curs. In these cases, US examination may show disrup-
tion of the rib cortex (fracture lines, steps), haematoma
or callous formation depending on the age of the fracture
[5-7]. In addition, US is superior to plain radiographs
when diagnosing traumatic dislodgement of costochon-
dral cartilage or fractured sternum [5,6]. Documentation
in two perpendicular axes is mandatory.
2. Chest wall infections
As sonographic findings precede radiographic ab-
normalities, US has an important role in assessing chest
Fig 2. Three-month-old boy, asymptomatic superficial con-
wall infections. US findings in cellulitis are diffusely genital tumour of the thorax and right axilla, lymphangioma.
increased echogenicity and thickness of the subcutane- Clinical presentation (a). Ultrasound (b): macrocystic tumour
ous fat [3,5,7]. When cellulitis is complicated by a sub- with nearly anechoic cysts (stars) in the subcutaneous tissue re-
cutaneous abscess, focal organised fluid collection with specting the integrity of the ventilated lung (white arrows) and
inflammatory changes in the surrounding fat can be seen the rib (white triangle).
on B-mode US and increased peripheral vascularity on
colour Doppler [5,7,8]. Furthermore, US may facilitate
drainage of the abscess by targeting the largest pocket
of collection and used to follow-up chest wall infection,
ensuring its resolution.
A normal-appearing costochondral cartilage on US in
a patient presenting with costochondral pain may suggest
costochondritis [3]. US signs of osteomyelitis include
sub periosteal fluid collection, fluid in the proximity of
the bone and bone cortex disruption [3,5]. While US
diagnosis of osteomyelitis is feasible, rib osteomyelitis
may be particularly difficult to diagnose with US alone Fig 3. Tumour of the rib cartilage at the transition to the ster-
and MRI is usually warranted. num: chondromyxoidfibroma.
 Med Ultrason 2022; 24(1): 65-76 67
Table I. Chest wall tumours
Type Origin
Benign Bone Osteochondroma
Osteoblastoma
Osteoid osteoma
Aneurysmal bone cyst
Subcutaneous tissue Lymph nodes
Lipoma/lipoblastoma
Neurofibroma
Fibroma/desmoid
Epidermal cyst
Vascular Hemangioma/haemangioendothelioma/tufted angiomas
Venous malformation
Lymphangioma
Infantile myofibromatosis
Malignant Primary Ewing sarcoma/PNET*
Rhabdomyosarcoma/Leiomyosarcoma/Liposarcoma
Osteosarcoma/Chondrosarcoma
Malignant peripheral nerve cell tumour
Secondary Neuroblastoma
Hepatoblastoma
Leukaemia/Lymphoma
*PNET, primitive neuroectodermal tumour

Osteochondroma is the most common rib tumor pre- Although US offers a rapid, first-line assessment of
senting as a non-tender, hard mass [7,8]. US shows an ex- benign solid masses of the chest wall, MRI is often the
ophytic bone lesion with a hypoechoic cartilage cap [7]. best method for their evaluation if malignancy is sus-
Vascular masses are usually associated with change pected [8].
in colour of the overlying skin. Haemangiomas are the 4. Malignant chest wall tumours
most common vascular lesions in infancy and childhood. Malignant chest wall tumours are uncommon in
Typically, they are well circumscribed, but on grey-scale children (fig 4, fig 5) and metastatic tumours are more
may show heterogeneous echogenicity with non-specific common than primary neoplastic lesions (Table I) [8].
features. However, haemangiomas demonstrate high ves-
sel density and high Doppler frequency shifts can be di-
agnosed on US with high specificity and sensitivity when
both criteria are met [3,5-7].
Other vascular malformations, such as haemangioen-
dothelioma, tufted angiomas and infantile myofibroma-
tosis may have similar US appearances, without meet-
ing both criteria [3]. Thus, additional imaging studies are
necessary, including CT or MRI.
Venous malformations are characterized by sponge-
like serpiginous channels and anechoic cystic spaces [3,
5-7]. In contrast to haemangiomas, the venous flow is low
and often not detectable on Doppler US [6,7]. In venous
malformations, phleboliths may be identified on US [6,7].
Lymphatic malformations are seen on US as multiple,
cystic lesions with internal septae, without internal flow
[3,6,7]. Although typically anechoic, their echogenicity Fig 4. Hodgkin lymphoma in a 13 y/o girl. Retrosternal inho-
will increase after infection or haemorrhage [6,7]. Lym- mogeneous hypoechoic tumour with fracture and destruction of
phangioma is more commonly found in the axilla [6]. the sternum (arrow), transversal (a) and longitudinal (b) plane.
68 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
Fig 5. No thymus after heart surgery, 10-year-old boy with
Down syndrome.
Clinically, malignant chest wall tumours are associated
with pain and tenderness [3]. On US, they demonstrate
heterogeneous echogenicity with increased Doppler flow
[6]. The cortex of the adjacent bony structures may show
abnormal thickening or disruption [3,6]. US may facili-
tate histological diagnosis through US-guided biopsy.
MEDIASTINUM
Examination technique
The mediastinum is imaged in the parasternal longitu-
dinal section and scanned by tilting the ultrasound trans-
ducer. The same applies to the parasternal cross-sections
in the intercostal space. Since the sternum of premature
infants and young newborns is mostly cartilaginous, it
can be penetrated by US when scanning in a trans-sternal
approach. For older children, jugular or infra-clavicular
approach is used. More superficial structures, such as the
thymus, can be examined best by using a high-frequency
linear transducer. Structures of the mediastinum require
deeper penetration in older children with a low-frequen-
cy sector transducer. The mediastinum is the intrathorac-
ic and extrapleural space between the lungs, containing
organs from different systems, such as the thymus, tra-
chea, oesophagus and lymph nodes. The heart and major
thoracic vessels are also a part of it, but will not be con-
sidered in this publication.
Thymus
The thymus is horseshoe-shaped on cross-sectional
images, with the aorta and the pulmonary artery found
dorsally on the right and left side respectively. In lon-
gitudinal section images, it appears triangular to oval
and is found cephalic to the aortic arch and the heart.
The thymus is homogeneous in echogenicity and sur-
rounded by a fibrous capsule, which can be visualised
as an echogenic boundary layer. When examined with a
high-resolution transducer, it shows echogenic lines and
dots within the tissue, corresponding to normal septa and
vessels. This typical echo texture allows thymic tissue to
be differentiated from pathological changes in the medi-
astinum. The echogenicity resembles to that of the liver,
sometimes being slightly more echogenic. When com-
pared to the thyroid though, thymus is less echogenic.
In older children, the thymus becomes gradually more
echogenic. The thymus tissue is very soft and as result
the organ demonstrates intermittent deformation, caused
by the contraction of the heart. This can be assessed in
real-time with US.
1. Thymic aplasia / hypoplasia
Thymus aplasia is characterised by absence of thy-
mus tissue (fig 6). This occurs with immune deficiency
syndrome, especially Di-George syndrome. In this case,
US is the method of first choice to demonstrate the char-
acteristic absence of thymus.
Thymic hypoplasia is very difficult to definitively
diagnose due to insufficient standard values. It is only
justified to raise the possibility of thymic hypoplasia if
the thymus is significantly reduced and can hardly be dis-
tinguished from adjacent structures (fig 7).
2. Thymic hyperplasia
This is not an actual clinical entity, hence the name
“physiological thymic hyperplasia”. The question often
arises when a chest X-ray demonstrates a shading in the
area of ​​the mediastinum, being either the thymus or an-
other abnormal process. This question can be answered
clearly by US due to the typical echo texture and echo-
genicity. The synchronous deformability, induced by car-
diac motion is never present with atelectasis, infiltrates
or tumours.
3. Aberrant thymus
In rare cases, thymus tissue can also be found ectopi-
cally [10-12]. Such an ectopic thymus is found pulling
Fig 6. A 7 days-old newborn after heart catheter: thymus with
air inclusions; spontaneous recovery after a few days.

Fig 7. Fibrosarcoma, 10 months old boy, left dorsolateral chest
wall tumor adjacent to the lower rib edge. B-Mode: inhomoge-
neous almost smoothly limited (a). Colour Doppler: only low
blood flow peripheral (b).
along the thymopharyngeal duct from the lower jaw to
the mediastinum. Such a thymus usually remains asymp-
tomatic, but it can lead to vascular compression and dif-
ficulty swallowing when displaced.
4. Structural changes of the thymus
Structural changes in childhood are very rare. These
include the thymoma (paraneoplastic syndrome), thy-
mus carcinoma and lymphosarcoma. Leukaemia can also
lead to structural changes in the thymus. These diseases
cannot be differentiated with US. The thymus is usu-
ally shown relatively larger in size. The echogenicity is
mostly heterogeneous, possibly due to cystic parts. The
echogenicity can be increased but also reduced.
4.1. Lymphomas
Lymphomas are usually located in the anterior and
middle mediastinum, hence often causing shift of the
large vessels.
The position and width of the trachea is important.
Especially in T-cell lymphoma rapid growth is possible
and decision for early therapy is necessary before the tra-
chea is collapsed. Imaging findings include both enlarged
lymph nodes and diffuse thymus infiltration. Contrary to
inflammatory changes, lymphomas are less vascularised
on colour Doppler technique, but a confident differen-
tiation is not possible with Doppler technique only. On
US, lymph nodes in lymphomas appear mostly multiple,
rounded oval and hypoechoic. They are often polycyclic
limited and are in the vicinity of the large vessels. Differ-
entiation from other solid processes is not always possi-
ble. US can be used to observe thoracic wall or pericardial
infiltration and to look for infradiaphragmal lymph nodes.
4.2. Solid masses
Solid masses are rare in childhood, and differentia-
tion based on US alone is not possible. They are usually
hyperechoic and heterogeneous, often containing very
echogenic parts with acoustic shadowing, representing
calcifications. There may also be some cystic compo-
Med Ultrason 2022; 24(1): 65-76 69
nents. A better classification can be achieved through
the precise localisation of the lesion. Teratomas (15%
of the mediastinal masses), teratocarcinomas, chorionic
carcinomas, yolk sac tumours and thymolipomas (mostly
relatively homogeneous) are located in the anterior and
middle mediastinum.
Neuroblastoma, ganglioneuroma and neurofibroma
are usually found in the posterior mediastinum next to
the vertebral column and can be visualised from a poste-
rior view (sitting position).
4.3. Cystic masses
Such lesions are shown sonographically if they are
liquid-filled or empty, with fluid appearing hypoechoic
on US. Differentiation is only feasible based on the topo-
graphical location.
The most common cystic masses are thymic cysts and
mediastinal lymphangiomas. They usually have smooth
walls. Bleeding can occasionally make the contents of
the cyst appear more echogenic. Ultrasound may be
helpfully to differentiate bronchogenic cyst from layered
foregut duplication cysts (oesophageal duplication) [13].
Trachea
The echopoor tracheal cartilage and the ventral wall
can be visualised sonographically. It is possible to meas-
ure the thickness of the cartilage, the width of the trachea
[14,15] or to evaluate movements of the trachea. US may
be helpfully to differentiate bronchogenic cyst from lay-
ered foregut duplication cysts (oesophageal duplication).
Haemangioma can be visualised in the subglottic position.
Beside the vertebral column, neurogenic cysts are possi-
ble. The adjacent air filling of the trachea leads to dorsal
sound cancellation and dirty shadowing. Thus, patho-
logical changes in the trachea are difficult to evaluate.
The position of endotracheal tubes can be checked us-
ing US. With a subglottic visualisation of the trachea, the
width of the sound cancellation correlates very well with
the size of the tube to be selected [16]. The tube is shown
in cross-section as two short, parallel echogenic lines. In
longitudinal section, the tube below the ventral wall of
the trachea is shown as longitudinal, parallel echogenic
lines. The movement of the tube facilitates identification.
This means that both the depth of the intubation and the
faulty intubation can be displayed.
Diaphragm
US of the diaphragm is a reliable method for evalu-
ation of anatomy and function of the diaphragm. It was
firstly described in 1969 by Cohen followed by Haber
in 1975 [17], who showed that US is superior to fluor-
70 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
oscopy in assessing diaphragmatic movement disorders
[18]. The diaphragm has a crucial role in breathing. Mul-
tiple methods for evaluation were developed including
fluoroscopic sniff test, electromyography and nerve con-
duction, which are used in detection of diaphragmatic
dysfunction. However, these are time-consuming and
procedure-related complications might occur [19]. US
evaluation of the diaphragm has proved to be useful in
patients with suspicion of diaphragm paralysis for dec-
ades. From diaphragm paralysis [20], to the evaluation
of respiratory function [21], diaphragm US has been con-
firmed to be a valuable technique, useful in the diagnosis
of different signs and symptoms like paradoxical respi-
ration, dyspnoea or asymmetric radiographic findings.
Other conditions that can be diagnosed by US include
diaphragmatic hernias, eventrations, peridiaphragmatic
abscess, thoracic tumours, pleural collections and lesions
causing diaphragm paralysis [17].
Examination technique
Diaphragmatic US is performed ideally with the pa-
tient in the supine position or in a sitting position if the
former is not possible, e.g., in children with a neuromus-
cular disorder.
Lateral parts of the diaphragm can be imaged in an
axillary longitudinal section and appear similar to mus-
cles in being generally hypoechoic. Above the spleen and
the liver, the diaphragm appears as a thin hypoechoic
three-layered muscular structure when using high fre-
quency transducers. The interface between air-filled tis-
sue (lung) and solid organs (liver, spleen) is generated
by changes in the acoustic impedance and appears as a
bright echo line using low frequency transducers. This
so-called bright diaphragm line does not represent the
diaphragm itself but an interface. The position of the
diaphragm and respiratory diaphragmatic movement are
the focus of the evaluation. The best examination posi-
tion is the sub-diaphragmatic longitudinal view along the
anterior axillary line. Comparison between the left and
right side can be performed by tilting the transducer to
the cranial sub sternal cross-section. In this case, the right
side of the diaphragm can be very well seen while show-
ing the left side can be challenging due to the air-filled
stomach. Because of the necessary depth of penetration,
low-frequency convex transducers (3- 9 MHz) should be
used. By using M-mode, the mobility of the diaphragm
can be shown and measured.
Deep breathing might be difficult in non-compliant
children [17]. The linear, high frequency transducers (7-
17 MHz) are primarily used for the diaphragmatic muscle
assessment and the convex probe (micro convex for chil-
dren and thin patients) or a sectorial for smaller children
[22]:
• The anterior central view from the subxiphoid
window is recommended for the left and right
hemidiaphragm. This view allows the comparison
between the two hemidiaphragms and their move-
ments, using the sectorial or a micro convex probe,
in B mode, transversely positioned transabdomi-
nal, cranially upright oriented [19].
• The lateral views, in the so-called apposition area,
between the axillar anterior line and mid-clavicu-
lar line, with the linear high frequency transducer,
positioned longitudinally, perpendicular to the
lower ribs;
• The right hemidiaphragm is easily seen through
the liver window, while the left diaphragm is not
easy through the spleen window, because of stom-
ach [23] or other abdominal content [20].
A good image may be obtained between the midclav-
icular and anterior axillary lines in the subcostal area and
by directing the probe medially and cranially to achieve
the best view of the right hemidiaphragm at the apposi-
tion zone [24]. For the posterior view, in the sitting posi-
tion, using a convex probe, the renal parenchyma may
conduct the US to enable the diaphragm to be assessed in
children, but the window might be smaller and not pos-
sible in every child as it is thoracic wall size dependent
[17].
The intercostal approach can be used but the inter-
costal space is different in children, compared to adults.
However, the preferred intercostal spaces in adults be-
tween 7th-8th rib or 8-9th rib even 9-10th are potentially
applicable in supine children [17]. The estimation of the
lateral diaphragm parts is important, because of their
main role in breathing compared to the anterior central
zone which has less effect on inspiration [25]. Diaphrag-
matic US has a few limitations: firstly, standard reference
parameters such as thickness are only available in some
populations; secondly, it is difficulty to visualize the left
diaphragm in overweight children. Lastly, assessment of
the diaphragm motion can be affected by inspiratory ef-
fort and interposition of abdominal organs [17].
Diaphragm parameters
The measurable diaphragm parameters are: diaphrag-
matic thickness excursion and thickening fraction. The
first is a static parameter while the others are dynamic
parameters.
The diaphragm thickness (DT) is increased in inspi-
ration and decreases in expiration. DT is well correlated
with body weight in children and adults. Other param-
eters like trans diaphragmatic pressure is higher in chil-
dren compared to adults [26]. Normal DT value scales
are available in healthy adults [27] and children [26] but
only in limited populations. Atrophy of the diaphragm

occurs in ventilated children with acute respiratory fail-
ure, especially in those receiving neuromuscular blockers
[28]. DT is a very important parameter for detection of
diaphragmatic atrophy of mechanically ventilated chil-
dren [29] and is a reliable prognostic indicator for venti-
lator-induced diaphragmatic atrophy [29] and for the risk
for extubation failure in children [30]. US can be used
for diagnosing atrophy secondary to other diseases such
as myodystrophy, malnutrition, trauma, compression
and associated breathing disorders with restrictive dys-
function, malnutrition and electrolyte disturbances (hy-
pophosphatemia, hypokalaemia, hypocalcaemia) [31].
Diaphragmatic excursion (DE) is assessed by M-
mode US, measuring the distance of diaphragm move-
ment between end inspiration and end expiration [32] and
normal reference values exist in small adult populations
[33] and also a few in children [20]. Decreased DE can be
suggestive of a number of diaphragm disorders, such as
neuromuscular diseases (including hypocalcaemia, hy-
pokalaemia, hypophosphatemia [31]), neuropraxia sec-
ondary to induced hypothermia, ventilated patients and
pathology in the vicinity, from lung and pleura (pneumo-
nia, pleural effusions) or abdomen (peritonitis).
Diaphragm thickening fraction (DTF) is defined
as the fraction of the difference between the thickness-
at-end inspiration and thickness-at-end expiration /
thickness-at-end-expiration [19]. The increased DTF is
strongly correlated with spontaneous breathing fraction
and can predict weaning success, correlated with an in-
creased percentage of successful extubation [34], even in
post transplanted patients [35].
Diaphragm motility
The evaluation of diaphragm motility is essential
when there is a suspicion of diaphragmatic paralysis. This
could be performed by chest X-ray, which can demon-
strate elevation of a hemidiaphragm, but does not provide
information on diaphragm motion. CT would be more
specific in detecting the structural changes but provide
less information on motility and expose the patients to
increased ionizing radiations. Diaphragm motility might
be evaluated by MRI which does not involve ionizing
radiation, but it is more time consuming and expensive
and sedation might be required [19,36,37].
It has been shown that US is a “highly sensitive meth-
od of demonstrating generalised or localised abnormali-
ties of diaphragmatic motion” [18] and new data suggest
that diaphragm motility is a predictor for effective ex-
tubation in children [38], confirming the importance of
US for diaphragm evaluation. One of the most impor-
tant indications for diaphragm US is after cardiac sur-
gery where there is a suspicion of diaphragm paralysis.
The diaphragm paralysis can be promptly evaluated by
Med Ultrason 2022; 24(1): 65-76 71
US and it has been shown to be superior to fluoroscopy
[39].
Specific pathological findings
Diaphragmatic conditions range from congenital mal-
formations such as diaphragmatic hernia to diaphragm
paralysis secondary to phrenic nerve injury in cardiac
surgery or muscular dystrophies and neuromuscular dis-
orders [40]. Detecting traumatic diaphragm rupture using
US has been suggested as an additional examination in
“Focused Abdominal Sonography for Trauma” (FAST)
examination for adults and children [41,42].
In diaphragmatic impairment, lung function assessed
by spirometry reveals a restrictive pattern, but spirom-
etry is challenging to perform in children, particularly in
those with neuromuscular diseases. It has been found that
a significant correlation exists between lung volumes and
diaphragm function in adults [43] and a linear correla-
tion between diaphragmatic thickening fraction and lung
volumes was demonstrated in children [27]. However,
this correlation was not confirmed in a study evaluating
the relationship between body plethysmography and dia-
phragm respiratory course in adults [44].
Congenital diaphragmatic disorders include anatomi-
cal defects such as Bochdalek and Morgagni congenital
hernia, eventration, diaphragmatic agenesis and func-
tional disorders including diaphragm paralysis, atrophy,
posttraumatic and iatrogenic injury during surgery and
neuromuscular diseases, all have important effects in
childhood [27].
Congenital diaphragmatic hernia (CDH) is one of
the most frequent major  congenital anomalies, with a
significant postnatal mortality rate due to acute compli-
cations or development of pulmonary hypertension [45].
Diaphragmatic hernia can be easily diagnosed with US.
Chest X-ray is the first imaging modality postnatally. In
fact, most cases are diagnosed prenatally by foetal US
combined with foetal MRI for risk evaluation (e.g., lung
volumetry). CDH can be diagnosed on chest X-ray when
gas containing bowel or stomach are projected over the
lung fields; if the solid organs such as liver are involved.
US is the method of choice for diagnosis and can differ-
entiate CDH from pleural effusion in post-operative pa-
tients [22]. US before corrective surgery can accurately
evaluate the size of the defect in CDH and can provide
anatomical information, which can consequently influ-
ence the surgical approach [46].
US GUIDED/ASSISTED INTERVENTIONS
As discussed previously, US has an important role in
assessing all thoracic compartments: chest wall, pleura,
lung, mediastinum and diaphragm. Thus, interventional
72 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
procedures involving any of these compartments are of-
ten best and most safely performed using US guidance
or assistance. US guided drainage of chest wall or lung
abscesses aids the microbiological diagnosis, while of-
fering a minimally invasive therapeutic opportunity [3].
Accessible mediastinum lymph nodes and solid tumours
of the chest wall or lung may be biopsied via US guid-
ance [3,4,6,7]. US guidance is preferred for peripheral
lung biopsy for both children and adults [38,47-49]. The
utility and low radiation burden of US guided inser-
tion of arterial and venous catheters is well established
[5,50,51]. In addition, venous malformations of the chest
wall may benefit from US guided sclerotherapy [6,52].
Diagnostic and therapeutic US guided drainage of pleural
effusion and empyema is well described in the literature
[53,54]. As per recent guidelines, fibrinolytic treatment
is initiated via transthoracic catheter [55]. The use of US
guided insertion of small-bore pig-tail catheters appears
safe and efficient both for symptomatic pleural effusion
[56] and empyema [57]. Although, US is not usually em-
ployed for identifying pneumothorax in some cases it
may be a more accessible and rapid method of diagnosis
and treatment [58].
US guided minimally invasive therapeutic approach-
es are employed for congenital pulmonary airway mal-
formation (CPAM) with sequestration, both in the foetal
and neonatal period [6,7,59]. In a case series, the authors
describe the occlusion of the feeding vessel in utero via
interstitial laser, thrombogenic coil embolization and ra-
dio-frequency ablation [59].
Interventions in the pleural space
US is the most specific and sensitive imaging study
used to confirm the presence and type of pleural effusion,
as well as the optimal site for tube placement [55]. It can
also signal complications such as the presence of septa-
tion or debris within the pleural space [55]. The size of
the pleural effusion, cultures and respiratory compromise
guide the decision for drainage of the pleural effusion
[60] (Table II).
In children, 10F and 12F drainage tubes are consid-
ered equally effective. When choosing the site of the
thoracocentesis, whether diagnostic or therapeutic, one
should always have in mind the “safe triangle”. Sam-
ples of the pleural effusion must be sent for biochemical
Table II. Pleural drainage decision.
Size of effusion (lateral decubitus) Culture
Small: <10 mm or opacifies <25% of hemithorax Unknown/Negative
Moderate: >10 mm or opacifies <50% of hemithorax Negative/Positive
Large: Opacifies >50% of hemithorax Positive (empyema)
and microbiologic testing. If there is suspicion that the
pleural effusion is not secondary to infection, cytological
analysis is mandatory.
When therapeutic thoracocentesis is performed, the
drainage tube is placed to one-way suction, ranging from
-5 to -20 cm of water [55].
Currently, there is no consensus regarding pleural
drainage volumes in children. Volumes equal to 10 ml/
kg with a maximum of 1.5 l are generally considered
safe. Re-expansion pulmonary oedema (RPE) is the most
feared complication of large and rapid pleural effusion
drainage. Still, RPE is rare in children.
Intrapleural fibrinolytic agents are used successfully
for pleural drainage. There are several protocols used
for pleural sclerotherapy using either urokinase or tissue
plasminogen activator [61,62].
Pleuritic pain must be managed with analgesics in or-
der to ensure appropriate expansion of the lung after the
procedure. When the output of the chest tube decreases to
< 1 ml/kg/24 hours, it may be removed [61].
ARTEFACTS
Recognition of artefacts is crucial in order to avoid
misinterpretation of the findings and false diagnosis.
Physicians should thus familiarise themselves with the
terminology and nature of these technical phenomena.
The term “beam width artefact” describes the visu-
alisation of echoes generated from a point lying outside
the US beam, inside the imaging plane and represents an
expression of US lateral resolution. To understand this
artefact, it is important to remember that the US beam is
bow-tie shaped, being narrowed at the level of the focus
point and then widens reaching a width larger than the
probe width. As a result, a reflection originating from a
point lying outside the width of the transducer but within
the widened width of the beam will be erroneously dis-
played inside the field-of-view. In practice, this explains
why bright echoes can be falsely displayed within an-
echoic structures such as the gallbladder or an anechoic
pleural effusion. The solution to this problem is always to
keep the focal point at the level of structure under exami-
nation, using multiple focal zones or placing the point of
interest at the centre of transducer. In that way, the beam
Pleural drainage +/- fibrinolysis
No
No (no respiratory compromise and no empyema)
Yes (respiratory compromise or empyema)
Yes

will have its narrowest width at the level of examination,
visualising only the area of interest and leaving adjacent
echoes outside the imaging field [63,64].
The false visualization of echoes external to the imag-
ing field can also be caused by the “side lobe artefact”.
The radial expansion of piezoelectric crystals gives rise
to multiple low-amplitude waves of energy situated
laterally to the main array. When these waves encoun-
ter strong reflectors, the returning signals are shown as
originating from within the main imaging field. This ar-
tefact can also result in the false visualization of echoes
within anechoic structures such as the gallbladder or a
pleural effusion [63]. Potential solutions include decreas-
ing gain, using harmonics or different acoustic windows
[64]. The reflection of the US beam produces a number
of artefacts. When the US beam is completely reflected
by a strongly reflective surface such as bone or calculi,
the absence of any echo signal behind these surfaces is
known as ‘acoustic shadowing”. In the lungs, ‘acoustic
shadowing” occurs with the ribs hindering evaluation of
underlying pathology.
When the US beam encounters two parallel highly re-
flective structures, part of the beam is repeatedly reflect-
ed between the two structures. Some of those will reach
the transducer and will appears abnormally brightly as
a result of the time the beam has been reflected. This is
called the “reverberation artefact” and can be seen in
lung US when the beam hits catheters visualised as two
parallel echogenic lines.
The “comet tail” artefact is similar to the above, be-
ing only different in that it appears triangular with echoes
gradually narrowing due to the decreasing attenuation of
each reflected echo. The difference results from the fact
that the strong reflective areas are closely situated and
thus hardly discernible by the beam.
The “ring down” artefact is another form of artefact
produced by the sound energy generated by fluid trapped
between air-bubbles. This appears as a white vertical
straight line below the gas bubble and can be seen in lung
abscesses or colonic loops under the diaphragm. Similar
to the mechanism of ring-down artefact, aggregates of
gas may give rise to “dirty shadowing”, consisting of an
area without useful signal. This can be seen with nor-
mal lung parenchyma and gas bubbles in lung abscesses
[63,64] Reverberation and comet tail artefact can be miti-
gated by reducing gain or using a different acoustic win-
dow. On the contrary, if these artefacts need to be accen-
tuated in order to better detect pathologic gas, the spatial
compounding function should be turned off [64].
The “mirror image” artefact is seen when the echo
generated from a strong reflective surface such as the dia-
phragm encounters another object during its path back to
Med Ultrason 2022; 24(1): 65-76 73
the transducer. Part of the echo will be reflected back a
second time to create a false secondary echo which cre-
ates a ‘mirror image’ of the true lesion on the opposite
side of the diaphragm. The commonest example is seen
as reflection of a liver lesion into the thorax. This arte-
fact may also result in the reflection of abdominal ascites
into the hemi thorax, thus mimicking a pleural effusion
[63,64].
Another important artefact is the “speed displace-
ment” artefact, in which US travels with different speed
through different materials. As a result, the time required
for a reflected signal to reach the transducer is different.
As the ultrasound machine assumes an equal speed for all
echoes reflected, this causes the artefactual visualization
of echoes in deeper locations. In lung US, this explains
why the diaphragm may appear discontinuous with some
of its segments being visualised deeper because of differ-
ent speed of sound through fatty parts of the liver [63].
In contrast to acoustic shadowing, acoustic enhance-
ment occurs when the US beam travels through an an-
echoic structure such a pleural effusion, beam attenua-
tion is lower than that of the adjacent beam, meaning that
the next reflected echo will be visualised brighter [63].
An important artefact arising from the lung-pleura in-
terface is call B-line artefact (BLA). In healthy children
with a normally aerated lung, nothing is visible on US
other than horizontal hyperechoic lines below the pleural
lines known as A lines, these are reverberation artefacts.
In patients with alveolar – interstitial syndrome, a dis-
ease involving the alveolar space and pulmonary intersti-
tial, this instead appears as multiple narrow hyperechoic
laser-beam which reaches up to the edges of the screen.
This is known as a “comet tail” and is also known as BLA
[65], however, the latter is the preferred term when refer-
ring to the lung. Both, mild or severe alveolar – intersti-
tial syndrome can lead to this artefact. In patients with
localised disease, the involved lobe can also be localised
based on correlation of the intercostal rib space and pres-
ence of the “comet tail” artefact with sensitivity ranging
from 86% to 93% and specificity of 87% to 96% [66]. A
number of factors have been described to influence the
detection of BLA [67,68]. Although BLA was initially
described in alveolar – interstitial syndrome, any condi-
tions which cause loss of lung aeration peripherally lead-
ing to increased density will give rise to BLA.
Artifacts in contrast-enhanced ultrasound (CEUS)
Contrast-enhanced ultrasound (CEUS) is a well-es-
tablished technique for many abdominal organs in adults
but has also been shown to have lung applications [69].
In Europe, intravenous CEUS in children is still per-
formed off-license. Most CEUS artefacts occur in abdom-
inal applications but they can also affect lung imaging.
74 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
Non-linear artefacts or incomplete suppression may be
encountered with highly echogenic interfaces or struc-
tures, which are projected on the contrast-specific image.
These signals should not be perceived as true enhance-
ment and can be differentiated from microbubbles signal
by comparing the contrast-specific and low-mechanical
index B-mode image prior to contrast administration, as
they will appear identical in both images. In the case of
lung US, gas bubbles from abscesses can be visualised
as echogenic areas projected on contrast-specific images.
When assessing a lung consolidation, it should be kept
in mind that near-field signal loss can be caused by the
disruption of microbubbles incurred by the higher acous-
tic pressure in this part of the imaging plane. Imaging
plane signal loss can be seen when scanning on the same
plane for a long period of time. The continuous acoustic
pressure disrupts microbubbles. This mainly interferes
with liver examinations for the characterization of focal
lesions and is less common in the lung where the trans-
ducer typically sweeps through areas of consolidation or
pleural pathology to detect necrosis. High doses of mi-
crobubbles cause shadowing similar to ribs, due to the at-
tenuation of beam. This can be observed in intracavitary
applications of CEUS, where the solution to be adminis-
tered needs to be carefully reconstituted, including only
a drop (0.1 ml) of microbubbles [70,71].
Conclusion
The series of reviews is intended to summarize and il-
lustrate current knowledge on thoracic ultrasound in pae-
diatric patients. The reader is warmly welcome to stimu-
late a discussion and to illustrate the initiated WFUMB
Atlas.
Conflict of interest: none
References
1. Jaworska J, Buda N, Ciuca IM, et al. Ultrasound of the
pleura in children, WFUMB review paper. Med Ultrason
2021. doi: 10.11152/mu-3058.
2. Dietrich CF, Buda N, Ciuca IM, et al. Lung ultrasound in
children, WFUMB review paper (part 2). Med Ultrason
2021. doi: 10.11152/mu-3059.
3. Mong A, Epelman M, Darge K. Ultrasound of the pediatric
chest. Pediatr Radiology 2012;42:1287-1297.
4. Joshi P, Vasishta A, Gupta M. Ultrasound of the pediatric
chest. Br J Radiol 2019;92:20190058.
5. Cox M, Soudack M, Podberesky DJ, Epelman M. Pediat-
ric chest ultrasound: a practical approach. Pediatr Radiol
2017;47:1058-1068.
6. Coley BD. Chest sonography in children: current indica-
tions, techniques, and imaging findings. Radiol Clin North
Am 2011;49:825-846.
7. Supakul N, Karmazyn B. Ultrasound of the pediatric chest–
the ins and outs.Semin Ultrasound CT MR 2013;34:274-
285.8.
8. Newman B. Thoracic neoplasms in children. Radiol Clin
North Am 2011;49:633-664.
9. Milos RI, Moritz T, Bernathova M, et al. Superficial
desmoid tumors: MRI and ultrasound imaging characteris-
tics. Eur J Radiol 2015;84:2194-2201.
10. Yildiz AE, Elhan AH, Fitoz S. Prevalence and sonographic
features of ectopic thyroidal thymus in children: A retro-
spective analysis. J Clin Ultrasound 2018;46:375-379.
11. Yildiz AE, Ceyhan K, Siklar Z, et al. Intrathyroidal Ectopic
Thymus in Children: Retrospective Analysis of Grayscale
and Doppler Sonographic Features. J Ultrasound Med
2015;34:1651-1656.
12. Bang MH, Shin J, Lee KS, Kang MJ. Intrathyroidal ectopic
thymus in children: A benign lesion. Medicine (Baltimore)
2018;97:e0282.
13. Balakrishnan K, Fonacier F, Sood S, Bamji N, Bostwick H,
Stringel G. Foregut Duplication Cysts in Children. JSLS
2017;21:e2017.00017.
14. Wanner MR, Marine MB, Dahl JP. Ultrasound diagnosis of
tracheal cartilaginous sleeve in a patient with Pfeiffer syn-
drome. Pediatr Radiol 2018;48:1814-1816.
15. Diwakar A, Adam RJ, Michalski AS, et al. Sonographic
evidence of abnormal tracheal cartilage ring structure in
cystic fibrosis. Laryngoscope 2015;125:2398-2404.
16. Gollu G, Onat Bermede A, Khanmammadov F, et al. Use of
ultrasonography as a noninvasive decisive tool to determine
the accurate endotracheal tube size in anesthetized children.
Arch Argent Pediatr 2018;116:172-178.
17. Sarwal A, Walker FO, Cartwright MS. Neuromuscular ul-
trasound for evaluation of the diaphragm. Muscle Nerve
2013;47:319-329.
18. Haber K, Asher M, Freimanis AK. Echographic evaluation
of diaphragmatic motion in intra-abdominal diseases. Radi-
ology 1975;114:141-144.
19. Fayssoil A, Behin A, Ogna A, et al. Diaphragm: Pathophys-
iology and Ultrasound Imaging in Neuromuscular Disor-
ders. J Neuromuscul Dis 2018;5:1-10.
20. Riccabona M, Sorantin E, Ring E. Application of M-mode
sonography to functional evaluation in pediatric patients.
Eur Radiol 1998;8:1457-1461.
21. Riccabona M. Ultrasound of the chest in children (mediasti-
num excluded). Eur Radiol 2008;18:390-399.
22. Trinavarat P, Riccabona M. Potential of ultrasound in the
pediatric chest. Eur J Radiol 2014;83:1507-1518.
23. Cohen E, Mier A, Heywood P, Murphy K, Boultbee J, Guz
A. Excursion-volume relation of the right hemidiaphragm
measured by ultrasonography and respiratory airflow meas-
urements. Thorax 1994;49:885-889.
24. El-Halaby H, Abdel-Hady H, Alsawah G, Abdelrahman A,
El-Tahan H. Sonographic Evaluation of Diaphragmatic Ex-
cursion and Thickness in Healthy Infants and Children. J
Ultrasound Med 2016;35:167-175.
25. Houston JG, Morris AD, Howie CA, Reid JL, McMillan N.
Technical report: quantitative assessment of diaphragmatic

movement--a reproducible method using ultrasound. Clin
Radiol 1992;46:405-407.
26. Rehan VK, McCool FD. Diaphragm dimensions of the
healthy term infant. Acta Paediatr 2003;92:1062-1067.
27. McCool FD, Tzelepis GE. Dysfunction of the diaphragm. N
Engl J Med 2012;366:932-942.
28. Goligher EC, Laghi F, Detsky ME, et al. Measuring dia-
phragm thickness with ultrasound in mechanically venti-
lated patients: feasibility, reproducibility and validity. In-
tensive Care Med 2015;41:642-649.
29. Johnson RW, Ng KWP, Dietz AR, et al. Muscle atrophy in
mechanically-ventilated critically ill children. PLoS One
2018;13:e0207720.
30. Terhart MN, Hanekom S, Lupton-Smith A, Morrow B. Re-
liability of ultrasonic diaphragm thickness measurement in
mechanically ventilated infants and children: A pilot study.
South Afr J Crit Care (Online) 2018;34:22-27.
31. McParland C, Resch EF, Krishnan B, Wang Y, Cujec B, Gal-
lagher CG. Inspiratory muscle weakness in chronic heart
failure: role of nutrition and electrolyte status and systemic
myopathy. Am J Respir Crit Care Med 1995;151:1101-
1107.
32. Turton P, ALAidarous S, Welters I. A narrative review of
diaphragm ultrasound to predict weaning from mechanical
ventilation: where are we and where are we heading? Ultra-
sound J 2019;11:2.
33. Boussuges A, Gole Y, Blanc P. Diaphragmatic motion stud-
ied by m-mode ultrasonography: methods, reproducibility,
and normal values. Chest 2009;135:391-400.
34. Glau CL, Conlon TW, Himebauch AS, et al. Progressive
Diaphragm Atrophy in Pediatric Acute Respiratory Failure.
Pediatr Crit Care Med 2018;19:406-411.
35. Sharma A, Karna ST, Tandon M, et al. Use of ultrasound-
guided preoperative diaphragmatic thickness as a predictor
of postoperative weaning failure in recipients and donors
scheduled for living donor liver transplant surgery. Saudi J
Anaesth 2018;12:406-411.
36. Caraiani C, Dong Y, Rudd AG, Dietrich CF. Reasons for in-
adequate or incomplete imaging techniques. Med Ultrason
2018;20:498-507.
37. Caraiani C, Petresc B, Dong Y, Dietrich CF. Contraindica-
tions and adverse effects in abdominal imaging. Med Ultra-
son 2019;21:456-463.
38. Lee EP, Hsia SH, Hsiao HF, et al. Evaluation of diaphrag-
matic function in mechanically ventilated children: An ul-
trasound study. PLoS One 2017;12:e0183560.
39. Sanchez de Toledo J, Munoz R, Landsittel D, et al. Diag-
nosis of abnormal diaphragm motion after cardiothoracic
surgery: ultrasound performed by a cardiac intensivist vs.
fluoroscopy. Congenit Heart Dis 2010;5:565-572.
40. Dube BP, Dres M. Diaphragm Dysfunction: Diagnos-
tic Approaches and Management Strategies. J Clin Med
2016;5:113.
41. Blaivas M, Brannam L, Hawkins M, Lyon M, Sriram K.
Bedside emergency ultrasonographic diagnosis of dia-
phragmatic rupture in blunt abdominal trauma. Am J Emerg
Med 2004;22:601-604.
Med Ultrason 2022; 24(1): 65-76 75
42. Gangahar R, Doshi D. FAST scan in the diagnosis of acute
diaphragmatic rupture. Am J Emerg Med 2010;28:387.e1-3.
43. Ayoub J, Metge L, Dauzat M, et al. Diaphragm kinet-
ics coupled with spirometry. M-mode ultrasonographic
and fluoroscopic study; preliminary results. J Radiol
1997;78:563-568.
44. Scott S, Fuld JP, Carter R, McEntegart M, MacFarlane NG.
Diaphragm ultrasonography as an alternative to whole-
body plethysmography in pulmonary function testing. J
Ultrasound Med 2006;25:225-232.
45. Brownlee EM, Howatson AG, Davis CF, Sabharwal AJ.
The hidden mortality of congenital diaphragmatic hernia: a
20-year review. J Pediatr Surg 2009;44:317-320.
46. Hattori K, Takamizawa S, Miyake Y, et al. Preoperative so-
nographic evaluation of the defect size and the diaphragm
rim in congenital diaphragmatic hernia - preliminary expe-
rience. Pediatr Radiol 2018;48:1550-1555.
47. Jarmakani M, Duguay S, Rust K, Conner K, Wagner JM.
Ultrasound Versus Computed Tomographic Guidance for
Percutaneous Biopsy of Chest Lesions. J Ultrasound Med
2016;35:1865-1872.
48. Safai Zadeh E, Keber CU, Dietrich CF, et al. Perfusion pat-
terns of peripheral pulmonary granulomatous lesions using
contrast-enhanced ultrasound (CEUS) and their correlation
with immunohistochemically detected vascularization pat-
terns. JUltrasound Med 2021. doi: 10.1002/jum.1573049.
49. Safai Zadeh E, Beutel B, Dietrich CF, et al. Perfusion
Patterns of Peripheral Pulmonary Lesions in COVID-19
Patients Using Contrast-Enhanced Ultrasound (CEUS):
A Case Series. J Ultrasound Med 2021. doi:10.1002/
jum.15624.
50. Bajaj M, Wells J, Liyanage A, Evans S, Hamill J. Radiation
burden of pediatric ultrasound-guided percutaneous central
venous access devices: A prospective cohort study. J Pedi-
atr Surg 2018;53:802-807.
51. Jenssen C, Brkljacic B, Hocke M, et al. EFSUMB Guide-
lines on Interventional Ultrasound (INVUS), Part VI - Ul-
trasound-Guided Vascular Interventions. Ultraschall Med
2016;37:473-476.
52. Kumar S, Bhavana K, Kumar S, Kumar P. Ultrasound-
guided polidocanol foam sclerotherapy for treating venous
malformations. J Clin Ultrasound 2018;46:23-31.
53. Corcoran JP, Tazi-Mezalek R, Maldonado F, et al. State of
the art thoracic ultrasound: intervention and therapeutics.
Thorax 2017;72:840-849.
54. Safai Zadeh E, Gorg C, Dietrich CF, Gorlach J, Alhyari A,
Trenker C. Contrast-Enhanced Ultrasound for Evaluation
of Pleural Effusion: A Pictorial Essay. J Ultrasound Med
2021. doi:10.1002/jum.15705.
55. Feola GP, Hogan MJ, Baskin KM, et al. Quality Improve-
ment Standards for the Treatment of Pediatric Empyema. J
Vasc Interv Radiol 2018;29:1415-1422.
56. Miraglia R, Maruzzelli L, Piazza M, et al. Real-time ultra-
sound-guided placement of a pigtail catheter in supine posi-
tion for draining pleural effusion in pediatric patients who
have undergone liver transplantation. J Clin Ultrasound
2016;44:284-289.
76 Cheng Fang et al Ultrasound of the chest and mediastinum in children, interventions and artefacts. WFUMB review paper (part 3)
57. Lewis MR, Micic TA, Doull IJM, Evans A. Real-time
ultrasound-guided pigtail catheter chest drain for compli-
cated parapneumonic effusion and empyema in children -
16-year, single-centre experience of radiologically placed
drains. Pediatr Radiol 2018;48:1410-1416.
58. Migliaro F, Sodano A, Capasso L, Raimondi F. Lung
ultrasound-guided emergency pneumothorax nee-
dle aspiration in a very preterm infant. BMJ Case Rep
2014;2014:bcr2014206803.
59. Baud D, Windrim R, Kachura JR, et al. Minimally inva-
sive fetal therapy for hydropic lung masses: three different
approaches and review of the literature. Ultrasound Obstet
Gynecol 2013;42:440-448.
60. Bradley JS, Byington CL, Shah SS, et al. Executive sum-
mary: the management of community-acquired pneumonia
in infants and children older than 3 months of age: clinical
practice guidelines by the Pediatric Infectious Diseases So-
ciety and the Infectious Diseases Society of America. Clin
Infect Dis 2011;53:617-630.
61. Sonnappa S, Cohen G, Owens CM, et al. Comparison of
urokinase and video-assisted thoracoscopic surgery for
treatment of childhood empyema. Am J Respir Crit Care
Med 2006;174:221-227.
62. Hawkins JA, Scaife ES, Hillman ND, Feola GP. Current
treatment of pediatric empyema. Semin Thorac Cardiovasc
Surg 2004;16:196-200.
63. Feldman MK, Katyal S, Blackwood MS. US artifacts. Ra-
diographics 2009;29:1179-1189.
64. Baad M, Lu ZF, Reiser I, Paushter D. Clinical Significance
of US Artifacts. Radiographics 2017;37:1408-1423.
65. Lichtenstein D, Meziere G, Biderman P, Gepner A, Barre
O. The comet-tail artifact. An ultrasound sign of alve-
olar-interstitial syndrome. Am J Respir Crit Care Med
1997;156:1640-1646.
66. Stefanidis K, Dimopoulos S, Kolofousi C, et al. Sonograph-
ic lobe localization of alveolar-interstitial syndrome in the
critically ill. Crit Care Res Pract 2012;2012:179719.
67. Dietrich CF, Mathis G, Blaivas M, et al. Lung artefacts and
their use. Med Ultrason 2016;18:488-499.
68. Dietrich CF, Mathis G, Blaivas M, et al. Lung B-line arte-
facts and their use. J Thorac Dis 2016;8:1356-1365.
69. Sidhu PS, Cantisani V, Dietrich CF, et al. The EFSUMB
Guidelines and Recommendations for the Clinical Practice
of Contrast-Enhanced Ultrasound (CEUS) in Non-Hepatic
Applications: Update 2017 (Short Version). Ultraschall
Med 2018;39:154-180.
70. Fetzer DT, Rafailidis V, Peterson C, Grant EG, Sidhu P,
Barr RG. Artifacts in contrast-enhanced ultrasound: a pic-
torial essay. Abdom Radiol (NY) 2018;43:977-997.
71. Dietrich CF, Ignee A, Hocke M, Schreiber-Dietrich D,
Greis C. Pitfalls and artefacts using contrast enhanced ul-
trasound. Z Gastroenterol 2011;49:350-356.
Review Med Ultrason 2022, Vol. 24, no. 1, 77-84
DOI: 10.11152/mu-2961

Transvaginal three-dimensional ultrasound for preoperative

assessment of myometrial invasion in patients with endometrial
cancer: a systematic review and meta-analysis
Tatiana Costas1, Rocío Belda2, Juan Luis Alcázar3

1Department of Obstetrics and Gynecology, University Hospital Salamanca, Salamanca, 2Department of Obstetrics

and Gynecology, General University Hospital, Valencia, 3Department of Obstetrics and Gynecology, University
Hospital of Navarre, School of Medicine, University of Navarra, Pamplona, Spain

Abstract
Aim: The aim of this meta-analysis is to evaluate the diagnostic accuracy of three-dimensional transvaginal ultrasound
subjective assessment (3D-TVS) in the preoperative detection of deep myometrial invasion (MI) in patients with endometrial
cancer, using definitive frozen section diagnosis after surgery as the reference standard. Material and methods: A search for
studies evaluating the role of 3D-TVS for assessing myometrial invasion in endometrial cancer from January 1990 to Novem-
ber 2020 was performed in PubMed/MEDLINE and Web of Science. The Quality Assessment of Diagnostic Accuracy Studies
2 evaluated the quality of the studies (QUADAS-2). All analyses were performed using MIDAS and METANDI commands.
Results: Nine studies comprising 581 women were included. The mean prevalence of deep MI was 39.8%. QUADAS as-
sessment showed that most studies had a high risk for the patient selection domain. Overall, the pooled estimated sensitivity,
specificity, positive likelihood and negative likelihood ratio of 3D-TVS for detecting deep MI were 84% (95% CI, 73-90%),
82% (95% CI, 75-88%), 5 (95% CI, 3.1-7.1) and 0.20 95% CI, 0.11-0.35). respectively. Conclusions: 3D-TVS has an accept-
able diagnostic performance for detecting MI in women with endometrial cancer.
Keywords: endometrial cancer; meta-analysis; myometrial invasion; systematic review; three-dimensional transvaginal
ultrasound

Introduction features for EC are FIGO stage (International Federation

Cancer of the corpus uteri, mainly endometrial can-
cer (EC), is the sixth most frequent form of cancer in
women worldwide, being the most common cause of gy-
naecologic malignancy in developed countries, with an
incidence of 12.9/100.000 women and a mortality rate
of 2.4/100.000 women. The most important prognosis
Received 25.11.2020  Accepted 31.01.2021
Med Ultrason
2022, Vol. 24, No 1, 77-84
Corresponding author: Juan Luis Alcazar, MD, PhD
ORCID ID: 0000-0002-9700-0853
Department of Obstetrics and Gynecology.
Clínica Universidad de Navarra.
36 Avenida Pío XII, 3110 Pamplona Spain
Phone: +34-948-296234
Fax: +34-948296500
E-mail: jlalcazar@unav.es
of Gynaecology and Obstetrics), myometrial infiltration
(MI), histological type and differentiation grade. Among
these, MI ≥50% is associated with both pelvic lymph
node involvement and extension into the parametrium
[1]. Even in low-grade endometroid tumours, lymph
node metastasis risk increases from 4% to 15% if the
tumour invades more than 50% of the myometrium [2].
These women are at a higher risk of recurrence and might
benefit from an extended surgical staging with systematic
pelvic and para-aortic lymphadenectomy in order to tailor
adjuvant therapy. It is important to avoid overtreatment,
as most of these women are elderly with comorbidities.
Lymphadenectomy shows no survival benefit in low-
risk endometrial cancer patients and, in fact, results in
increased complications and higher morbidity [3-7]. For
this reason, many authors advocate intraoperative gross
or frozen section evaluation of the myometrial invasion
78 Tatiana Costas et al Transvaginal 3D US for preoperative assessment of myometrial invasion in patients with endometrial cancer
in clinical stage I endometrial carcinoma to aid the deci-
sion of whether lymphadenectomy should be performed
or not. Frozen section has shown to be highly accurate
(94%) for determining myometrial invasion [8]. Howev-
er, it may be time consuming and it is not performed in all
hospitals. Therefore, a method that could reliably assess
myometrial invasion preoperatively would be helpful to
provide individual tailoring of the surgical approach [9].
There is no consensus about the optimal imaging tech-
nique for evaluation of myometrial and cervical invasion.
Two-dimensional transvaginal ultrasound (2D-TVS) and
magnetic resonance imaging (MRI) have shown accura-
cies of 69-74% and 66-90% respectively [10-13]. Three-
dimensional transvaginal ultrasonography (3D-TVS) has
several advantages when evaluating myometrial infiltra-
tion [14]. However, although several studies have shown
the utility of this technique in myometrial evaluation of
EC, the routine implementation in clinical practice has
not been stablished so far. Theoretically, from an efficient
point of view, the assessment of pre-surgical staging with
3D-TVS reveals a notorious benefit in terms of economic
approach as compared to MRI. Since the advent of 3D-
TVS, several studies have been published aiming at eval-
uating the role of this technique for detecting the depth of
myometrial infiltration in endometrial cancer.
On the other hand, more hospitals, especially in de-
veloped countries, tend to implement 3D-TVS in their
daily practices, revealing an increasing need of training
and investigation of the actual performance of this tech-
nique. To the best of our knowledge, not a single meta-
analysis analysed the diagnostic performance of 3D-TVS
for detecting myometrial invasion in women with EC.
Thus, the purpose of this systematic review and
meta-analysis is to evaluate the diagnostic accuracy of
3D-TVS in the preoperative detection of deep myome-
trial invasion in patients with endometrial cancer, using
definitive frozen section diagnosis after surgery as the
reference standard.
Material and methods
Protocol and registration
This systematic review and meta-analysis has been
made according to the PRISMA Statement [15]. All
methods for inclusion and exclusion criteria, data extrac-
tion and quality assessment were specified in advance.
The protocol was not registered.
Data sources and searches
Two electronic databases, PubMed/Medline and Web
of Science were screened in November 2020 by three of
the authors (TC, RB and JLA) to identify eligible stud-
ies. “Endometrial cancer”, “ultrasound” and “three-di-
mensional” were used as a search terms. No other meth-
odological filters in the database were included to avoid
possible omission of relevant studies, according to the
recommendation of Leeflang et al [16].
Study selection and data collection
Two authors (TC and RB) screened the titles iden-
tified by the searches to exclude irrelevant articles, not
strictly related to the topic. Abstracts of these articles
were revised and some of these articles were excluded
due to no relevant content (not relevant topic) or non-
English languages. Full text publications were indepen-
dently analysed by two authors (TC and RB) who read
and collected data of all the publications.
Inclusion criteria were: 1) Use of 3D-TVS as index
test to assessment myometrial invasion preoperatively
with subjective impression estimation of myometrial in-
vasion (less than 50% of myometrium or more) by the
investigator/s; 2) Reference standard was surgical patho-
logical data; 3) Presence of results sufficient to construct
the 2x2 table of diagnostic performance as minimum
data requirement. To avoid inclusion of duplicate cohorts
from at least two pair of studies reported by the same
authors, the study period of each study was examined. If
dates overlapped, the latest study published or the study
that could provide us a 2x2 table was chosen. Disagree-
ments arising during the process of study selection and
data collection were resolved by consensus among three
authors (JLA, TC and RB)
The PICOS (Patients, Intervention, Comparator,
Outcome, Study design) criteria used for inclusion and
exclusion of studies are shown in Table I. Diagnostic
accuracy and additional useful information on patients
and procedures were retrieved from a selected primary
study written independently by the same authors (JLA,
TC and RB).
Risk of bias in individual studies
Quality assessment was conducted using the tool pro-
vided by QUADAS (Quality Assessment of Diagnostic
Accuracy Studies –2). The QUADAS-2 format includes
four domains: 1) patient selection, 2) index test, 3) refer-
ence standard, 4) flow and timing. For each domain, the
risk of bias and concerns regarding applicability was ana-
lysed and rated as low, high and unclear risk. The qual-
ity assessment was used to provide an evaluation of the
overall quality of the studies and to investigate potential
sources of heterogeneity.
Three authors (JLA, TC and RB) evaluated the meth-
odological quality independently. Disagreements were
solved by discussion among these authors. The assess-
ment of quality was based on whether the study described
the study´s design as well as inclusion and exclusion cri-
teria for the patient selection domain. The evaluation of

quality for the index test domain was based on whether
the study reported on how the index test (3D-TVUS)
was performed and interpreted. The evaluation of qual-
ity for the reference standard domain examined the ref-
erence standard used in each study and if sonographers
and pathologists were blind or not to the index test. For
the flow-and-timing domain’s evaluation, a description
of the time elapsed from the index test assessment to the
reference standard result was evaluated.
Statistical analysis
We extracted or derived information on diagnostic
performance of 3D-TVS. Although other techniques
were used in the studies (such as RMI or 2D-TVS), only
the data from 3D-TVS evaluations were recollected. Pri-
mary outcome was pooled sensitivity, specificity, posi-
tive predictive value, negative predictive value and for
instance, true positive, true negative, false positive and
false negative values were obtained. Post-test probabili-
ties were calculated and plotted on Fagan nomograms.
Heterogeneity of all studies was graphically ex-
plored, drawing forest plots of sensitivity and specificity.
We then formally assessed the presence of heterogeneity
for sensitivity and specificity using Cochran´s Q test and
the I2 Index. A test for heterogeneity examines the null
hypothesis that all studies are evaluating the same effect
(Higgins et al [17]).
Cochran´s Q statistic is computed by summing the
squared deviations of each study´s estimate from the
overall meta-analytic estimate, weighting each study´s
contribution in the same manner as in the meta-analysis.
A p-value <0.1 indicates heterogeneity. The I2 index de-
scribes the percentage of total variation across studies
that is due to heterogeneity rather than chance. Accord-
ing to Higgins et al values of 25%, 50%, and 75% would
be considered to indicate low, moderate and high hetero-
geneity, respectively [17].
Summary receiver-operating characteristics (sROC)
curves for each approach were plotted to illustrate the
relationship between sensitivity and specificity.
All analyses were performed using Meta- analytical
integration of Diagnostic Accuracy Studies (MIDAS)
and METANDI commands in STATA version 12 for Win-
dows (Stata Corporation, College Station, TX, USA). p-
value <0.05 was considered statistically significant.
Results
Search results
The electronic search provided 108 citations. A flow-
chart summarizing literature identification and selection
is given in Figure 1. Eighty-three of them were excluded
due to no relevant topic (i.e. 3D MRI) or being review
Med Ultrason 2022; 24(1): 77-84 79
Fig 1. Flow chart showing studies selection process.
papers, 12 were excluded after reading abstract (not rel-
evant to the review) and one due to other language than
English (Chinese n=1). The full text of the remaining
twelve articles were examined. One of them [18] was
excluded due to the impossibility of obtain 2x2 tables.
Alcázar et al [19] and Mascilini et al [20] analysed myo-
metrial invasion with TDS and tumour/uterine volume
3D ratio without subjective impression as a method. For
that reason, they were also discarded.
Nine studies were finally taken into account
[9,14,18,21-26] (Table I). All studies analysed myome-
trial invasion with “subjective impression method”. No
additional relevant studies were found from references
cited in the papers included in the review. A flowchart
summarizing literature identification and selection is
given in Figure 1.
Characteristics of included studies
The nine studies [9,14,18,21-26] published between
2009 and 2019, reported data from 581 patients, 231
women having myometrial invasion >50% in the defini-
tive pathological samples. This was considered as the
prevalence and, therefore, pre-test probability.
The mean patient age was reported in 8 out of the 9
studies, in Trujillo et al no reference to this item being
made [22]. Patients’ mean age was 61.4 years, ponderated
by the number of patients in each study. The prevalence
of ≥ 50% myometrial invasion was 39.8%. Hormonal
status differentiating between pre- and postmenopausal
women was given in 3 out of 9 studies [14,21,26]. The
differentiation between histological grades was referred
in all papers except for Green et al [23] and Jantarasaen-
80 Tatiana Costas et al Transvaginal 3D US for preoperative assessment of myometrial invasion in patients with endometrial cancer
Table I. Characteristics of the studies included in the current meta-analysis according to PICO criteria
Author, Study’s Setting Consecutive N Cases with Index Method of Observers Reference
year Design recruitment MI≥ 50% test assessment test
Alcázar, Prospective Multi- Yes 113 27 3D-TVS Subjective Two Definitive
2009 [9] center histology
Saarelainen, Prospective Single Yes 20 12 3D-TVS Subjective Single Definitive
2012 [18] center histology
Jantarasaengaram, Prospective Single Yes 40 11 3D-TVS Subjective Single Definitive
2014 [14] center histology
Christensen, Retrospective Single Unclear 110 47 3D-TVS Subjective Two Definitive
2015 [24] center histology
Rodríguez-Trujillo, Retrospective Single Yes 98 39 3D-TVS Subjective Single Definitive
2016 [22] center histology
Ergeneglu, Prospective Single Unclear 45 9 3D-TVS Subjective Single Definitive
2016 [21] center histology
Green, Retrospective Single Yes 58 31 3D-TVS Subjective Multiple Definitive
2018 [23] center* histology
Yildrim, Prospective Single Unclear 40 45 3D-TVS Subjective Multiple Definitive
2018 [26] center histology
Yang, Retrospective Single Unclear 78 22 3D-TVS Subjective Two Definitive
2019 [25] center histology
* In the case of Green et al, women included as subjects of the study came from the same center. However, the examiners were spread all
over Europe, N - number of patients, MI - myometrial invasion

garam et al [14], This last author excludes all G3 cases as
being considered high-risk profile. All studies described
data regarding tumour histological type (endometroid
and non-endometroid) except for Christensen et al [24].
As mentioned before, in all studies selected for quan-
titative analysis, method to estimate myometrial invasion
was the subjective impression on 3D-TVS by examiner/s.
Most of the publications studied in this meta-analysis
make a comparison between subjective impression with
3D-TVS and MRI [18,22,24-26]. Two papers also com-
pared subjective impression with 3D-TVS and objective
methods with 3D-TVS [9,21]. Another paper compares
2D-TVS and 3D-TVS [23]. Only one publication uses the
subjective impression of 3D-TVS as the only method as-
sessed [14]. Pathological evaluation of the removed uter-
us was considered as a reference standard in all papers.
Methodological quality of included studies
A graphical display of the risk evaluation of bias and
concerns regarding applicability of the selected studies,
according to predefined criteria, is shown in figure 2.
Regarding risk of bias and the domain patient selec-
tion, all of them explain specifically patterns of inclusion.
Three studies do not include advanced stage patients
[14,22,24]. Christensen et al [24] included endometrial
hyperplasia with atypia in preoperative histology into the
group of patients studied. This last case can increase the
risk of overestimating the accuracy of 3D-TVS for the
estimation of <50% of MI. In contrast, the exclusion of
advanced stage cases was not considered a risk of bias,
in spite that this technique is relevant especially for early
stages since true tumour stage can only be determinated
after surgery. One study excluded patients with leiomio-
mas greater than 3 cm and submucous myomas [21]. This
study was considered as a high risk for bias in patient se-
lection domain since selecting exclusively ideal patients
for the validation of the 3D-TVS, can suppose a bias
in the validity of the test. Two studies [9,21] excluded
patients whose videos are incomplete and one study ex-
cludes patients with suboptimal image resolution (classi-
fied by score 3 of 4 of quality) [14]. Incomplete videos is
understandably criteria for exclusion, but if only optimal
image resolution is selected, the scenario left is made
up of exclusively ideal situations and this can lead to an
overestimation of the accuracy of the test.
Regarding the study design, prospective design with
US prior to surgery is carried out in 5 of the 9 studies
[9,14,18,21,26]. The rest of papers describe retrospec-
tive methodology. Christensen et al [24] evaluated the
volume of the uterus 6 months after surgery, the US ex-
aminers being blinded to the pathological result. In 2 of
the retrospective studies the examiner was blinded to the
pathology [23,24] and in 2 studies no explanation was
provided [22,25].
Concerning the domain index test, all studies describe
clearly the index test as well as how it was performed
and interpreted. However, Christensen et al [24], describ-
ing a 4-step subjective methodology (1- initial subjec-
tive evaluation, 2- TUI function, 3- render function and
 Med Ultrason 2022; 24(1): 77-84 81

Fig 2. Histogram plot quality assessment (risk of bias and concerns about applicability) for all studies included in the meta-analysis.
QUADAS indicates Quality Assessment of Diagnosis Accuracy Studies.

4- final evaluation with all gathered information), did not Reference standard concern of applicability was not
clearly define the importance given to each of these sub found to be high in any case, as histological definitive sam-
methods, leaving a riddle to understand the real criteria ple is usually the reference test in all gynaecology clinics .
used in this study. All studies adopted the same pre-spec- Diagnostic performance of 3D-TVS for detection of
ified threshold to define deep MI (≥50% of myometrial deep myometrial invasion
thickness in any of the three spatial orthogonal planes). We analysed the overall sensitivity and specificity
Four out of nine studies disposed ≥2 TVS examiners, of 3D-TVS in all studies to determine pulled sensitivity,
and the remainder had one expert examiner for all pa- specificity, LR+ and LR– of subjective impression in 3D-
tients [9,23,25,26]. From the group which included ≥2 TVS in detecting deep MI. Results were 84% (95% CI,
examiners, Christensen et al [24] disposed of examiners 73-90%), 82% (95% CI, 75-88%), 5.0 (95% CI, 3.1-7.1)
with limited experience in 3D TVS and Alcazar et al [9] and 0.2 95% CI, 0.1-0.3) respectively. Diagnostic odds
mixed both experts and beginners. The rest of them were ratio (DOR) was 23.0 (95% CI, 9.0-59.0). sROC curve is
experts. In two of the retrospective cases, as mentioned shown in figure 3. Area under the curve was 0.89 (95%
before, the blind condition of examiners to histological CI: 0.86-0.92)
final results is not clearly defined, considering, in conse- Heterogeneity was moderate for sensitivity (I2, 65.5%
quence, an unclear risk of bias 22.25]. (95% CI 41.4-90.0%)) and for specificity (I2, 65.1 %
Concerning the domain reference standard, all studies (95% CI 40.3-89.9%)). Data are shown in figure 4.
stated that histology was analysed after uterus removal,
but most of them did not describe how this was done
(frozen section method was only mentioned in 2 of the
studies [20,25]). Three papers do not mention if the pa-
thologist was blind to the 3D-TVS conclusions, leaving
the risk of bias in these cases unclear [14,21,26].
Time and flow were not specified in four of the stud-
ies included [21-23,25].
Concerning regarding applicability, all studies were
considered as low concern for patient selection domain.
The incidence of EC is high; however, advanced stages
of this neoplasia are not usually susceptible for 3D-TVS
as they already implicate a surgical attitude; EC is usu-
ally diagnosed in early stages, motivating the patients’
selection.
Index test has not been found to implicate a high con-
cern of applicability in any of the studies of this meta-
analysis, except for Christensen et al [24] as a conse- Fig 3. Summary receiver operating characteristic curve for 3D-
quence of the above mentioned 4-step method. TVS.
82 Tatiana Costas et al Transvaginal 3D US for preoperative assessment of myometrial invasion in patients with endometrial cancer

Fagan nomograms show that a positive test signifi-

cantly increased the pretest probability of MI 50% while
a negative significantly decreased the pre-test probability
of MI ≥ 50% (fig 5). We did not observe publication bias
(fig 6).

Discussion

EC is considered a neoplasia with good prognosis

in terms of survival when the diagnosis is made at an
early stage. The differences between the survival rates
observed in stage 1 compared to those seen in stage 3
are overwhelming: according to the American Cancer
Society the survival rate in early stages reaches 95% but
Fig 4. Forest plot of studies evaluated for subjective impres-
it decreases to 17% in distant spreading [27]. Thus, the sion in 3D-TVS for the assessment of myometrial invasion.
importance to find tests that lead to a better staging has Summary sensitivity and specificity as well as heterogeneity
become a challenge for the gynecological community. statistics (Cochran’s Q and I2) are shown.
As there is a really improvement of the rates in early
cancer detection, new needs emerge from the elevated
number of patients waiting for attending. In conse-
quence, the efficiency of tests used for pre-surgical stag-
ing started to be an important object of investigation. In
this context, MRI and 2D-TVS are currently considered
the imaging techniques of choice for estimating MI be-
fore surgery [28]. Alcázar et al [9] in 2009 were the first
authors that studied the performance of 3D-TVS. Since
then, several studies using 3D-TVS for detecting MI
have been reported.
We found that the overall diagnostic performance of
3D-TVS in detecting deep MI in women with EC have
A pooled sensitivity of 84% and pooled specificity of
81%, subjective method being the reference measure-
ment technique.
To the best of our knowledge, this is the first meta-
analysis that evaluates subjective impression in 3D-TVS Fig 5. Fagan normogram for detecting miometrial invasion
as the method of estimation of MI in women with EC with subjective impression in 3D-TVS.
reported to date.
The principal limitation in our meta-analysis is the
low number of papers reported. The total sum of patients
studied of 581. In this group of studies, moderate hetero-
geneity has been also found. Besides, the small sample
the data reported should be interpreted carefully. In addi-
tion, there has not been a proper analysis of the objective
methods used by some authors. However, we do consider
that this could have increased the heterogeneity and we
decided to exclude them. Due to the reduced number of
papers published that describe objective methods, sub-
analysis was non-viable.
Regarding generalizability, Alcazar et al published a
meta-analysis in which 2D-TVS showed 78% (95% CI,
72-83) sensibility and 81% (95% CI, 71-87) specificity Fig 6. Deeks funnel plot for assessing publication bias (p = 0.19)
[29]. In this same study, objective methodology with ESS indicates effective sample size.

Karlson and Gordon’s methods for the estimation of my-
ometrial infiltration showed a sensibility of 84-85% and
82-80% specificity, respectively. In 2017, Alcázar com-
pared 2D-TVS and MRI by meta-analysis. In this case,
with a sample of 560 patients they reached a sensitivity
of 75% and specificity of 86% for 2D-TVS [28].
Subjective 3D-TVS has, then, slightly superior re-
sults than those obtained by subjective methods in 2D-
TVS and very similar results from objective methods of
2D-TVS.
MRI has been considered a good imaging test for
evaluating MI. Recent studies demonstrated a pooled
sensitivity ranging from 81% to 90% and pooled spec-
ificities ranging from 82% to 89%. No differences were
found in these studies between contrast enhanced MRI
and diffusion weighted MRI [28,30-32].
When comparing data from several meta-analyses
about MRI with the data we have obtained in our me-
ta-analysis regarding 3D-TVS, we observe that, roughly,
diagnostic performance seems to be similar [28,30-32].
Objective methods in 3D-TVS have been studied.
Alcazar describes TDS method as an objective method
[9]. In this case, sensibility reaches 81.7% and specificity
41.5%. Ergenoglu validates this method demonstrating
sensibility 89% and specificity 61% [21]. Another objec-
tive method described described by Mascilini et al [20]
is tumor volume/uterine volume ratio. With this method,
sensitivity reported in some studies was 90.5% and 60%,
respectively, whereas, specificity was 28.5% and 85%,
respectively [19,20].
However, we did not find a representative number of
prospective studies dedicated exclusively to the compari-
son of 2D-TVS and 3D-TVS methods for MI assessment
in EC. Also, lack of information has been identified re-
garding the objective methods used for 3D-TVS assess-
ment. We think it is important to dedicate the investiga-
tion to the objective methods used in 3D-TVS evaluation
for improving the quality of comparison with subjective
3D-TVS methods.
In conclusion, subjective impression in 3D-TVS
seems to be an appropriate method for the estimation of
MI of EC in early stages. However, as we have observed
a high risk of patient selection bias, it is necessary to in-
crease the number of prospective well-designed studies
in this area.
Conflict of interest: none
References
1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D.
Global cancer statistics. CA Cancer J Clin 2011;61:69-90.
Med Ultrason 2022; 24(1): 77-84 83
2. AlHilli MM, Podratz KC, Dowdy SC, et al. Risk-scoring
system for the individualized prediction of lymphatic dis-
semination in patients with endometroid endometrial can-
cer. Gynecol Oncol 2013;131:103-108.
3. Morrow CP, Bundy BN, Kurman RJ, et al. Relationship
between surgical–pathological risk factors and outcome
in clinical stage I and II carcinoma of the endometrium:
a Gynecologic Oncology Group study. Gynecol Oncol
1991;40:55–65.
4. Brown AK, Madom L, Moore R, Granai CO, DiSilvestro
P. The prognostic significance of lower uterine segment in-
volvement in surgically staged endometrial cancer patients
with negative nodes. Gynecol Oncol 2007;105:55–58.
5. Nag S, Erickson B, Parikh S, Gupta N, Varia M, Glasgow
G. The American Brachytherapy Society recommendations
for high–dose–rate brachytherapy for carcinoma of the en-
dometrium. Int J Radiat Oncol Biol Phys 2000;48:779-790.
6. Kitchener H, Swart AM, Qian Q, Amos C, Parmar MK.
Efficacy of systematic pelvic lymphadenectomy in endo-
metrial cancer (MRC ASTEC trial): a randomised study.
Lancet 2009;373:125–136.
7. ASTEC/EN.5 Study Group, Blake P, Swart AM, Orton J,
et al. Adjuvant external beam radiotherapy in the treatment
of endometrial cancer (MRC ASTEC and NCIC CTG EN.5
randomised trials): pooled trial results, systematic review,
and meta–analysis. Lancet 2009;373:137–146.
8. Noumoff JS, Menzin A, Mikuta J, Lusk EJ, Morgan M,
LiVolsi VA. The ability to evaluate prognostic variables on
frozen section in hysterectomies performed for endometrial
carcinoma. Gynecol Oncol 1991;42:202–208.
9. Alcázar JL, Galván R, Albela S, et al. Assessing myometrial
infiltration by endometrial cancer: uterine virtual navigation
with three-dimensional US. Radiology 2009;250:776-783.
10. DelMaschio A, Vanzulli A, Sironi S, et al. Estimating the
depth of miometrial involvement by endometrial carcino-
ma: efficacy of transvaginal sonography vs MR imaging.
AJR Am J Roentgenol 1993;160:533–538.
11. Antonsen SL, Jensen LN, Loft A, et al. MRI, PET/CT and
ultrasound in the preoperative staging of endometrial can-
cer – a multicenter prospective comparative study. Gynecol
Oncol 2013;128:300–308.
12. Ortoft G, Dueholm M, Mathiesen O, et al. Preoperative
staging of endometrial cancer using TVS, MRI, and hys-
teroscopy. Acta Obstet Gynecol Scand 2013;92:536–545.
13. Rieck GC, Bulman J, Whitaker R, Leeson SC. A retrospec-
tive review of magnetic resonance imaging in assessing the
extent of myometrial infiltration for patients with endome-
trial carcinoma. J Obstet Gynaecol.2005;25:765–768.
14. Jantarasaengaram S, Praditphol N, Tansathit T, Vipupinyo
C, Vairojanavong K. Three-dimensional ultrasound with
volume contrast imaging for preoperative assessment of
myometrial invasion and cervical involvement in wom-
en with endometrial cancer. Ultrasound Obstet Gynecol
2014;43:569–574.
15. Preferred Reporting Items for Systematic Reviews and Me-
ta-Analyses (PRISMA) website. Available at: http://www.
prisma-statement.org/
84 Tatiana Costas et al Transvaginal 3D US for preoperative assessment of myometrial invasion in patients with endometrial cancer
16. Leeflang MMG,  Scholten RJPM,  Rutjes AWS,  Reitsma
JB, Bossuyt PMM. Use of methodological search filters to
identify diagnostic accuracy studies can lead to the omission
of relevant studies. J Clin Epidemiol 2006;59:234– 240.
17. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Meas-
uring inconsistency in meta–analyses. BMJ 2003;327:557–
560.
18. Saarelainen SK, Kööbi L, Järvenpää R, Laurila M, Mäen-
pää JU. The preoperative assessment of deep miometrial
invasion by three-dimensional ultrasound versus MRI
in endometrial carcinoma. Acta Obstet Gynecol Scand
2012;91:983–990.
19. Alcazar JL, Pineda L, Martinez-Astorquiza Corral T, et al.
Transvaginal/transrectal ultrasound for assessing myome-
trial invasion in endometrial cancer: a comparison of six
different approaches. J Gynecol Oncol 2015;26:201-207.
20. Mascilini F, Testa AC, Van Holsbeke C, Ameye L, Timmer-
man D, Epstein E. Evaluating myometrial and cervical in-
vasion in women with endometrial cancer: comparing sub-
jective assessment with objective measurement techniques.
Ultrasound Obstet Gynecol 2013;42:353-338.
21. Ergenoglu M, Akman L, Terek MC, et al. The prediction
of myometrial infiltration by three-dimensional ultrasonog-
raphy in patients with endometrial carcinoma: a valida-
tion study from Ege University Hospital. Med Ultrason
2016;18:201-206.  Gynecol 2015;46:405-413.
22. Rodríguez-Trujillo A, Martínez-Serrano MJ, Martínez-
Román S, et al. Preoperative Assessment of Myometrial
Invasion in Endometrial Cancer by 3D Ultrasound and
Diffusion-Weighted Magnetic Resonance Imaging: A Com-
parative Study. Int J Gynecol Cancer 2016;26:1105-1110.
23. Green RW, Valentin L, Alcazar JL, et al. Endometrial can-
cer off-line staging using two-dimensional transvaginal ul-
trasound and three-dimensional volume contrast imaging:
Intermethod agreement, interrater reliability and diagnostic
accuracy. Gynecol Oncol 2018;150:438-445.  32. Das SK, Niu XK, Wang JL, et al. Usefulness of DWI in pre-
24. Christensen JW, Dueholm M, Hansen ES, Marinovskij E,
Lundorf E, Ørtoft G. Assessment of myometrial invasion in
endometrial cancer using three-dimensional ultrasound and
magnetic resonance imaging. Acta Obstet Gynecol Scand
2016;95:55–64.
25. Yang T, Tian S, Li Y, et al. Magnetic Resonance Imaging
(MRI) and Three-Dimensional Transvaginal Ultrasonog-
raphy Scanning for Preoperative Assessment of High
Risk in Women with Endometrial Cancer. Med Sci Monit
2019;25:2024-2031.
26. Yildirim N, Saatli B, Kose S, et al. Predictability of myo-
metrial, lower uterine segment and cervical invasion with
3D transvaginal ultrasonography and magnetic resonance
imaging in endometrial cancer patients: a prospective co-
hort study. Med Ultrason 2018;20:348-354.
27. American cancer society. Survival rates for endometrial
cancer. Available from: https://www.cancer.org/cancer/
endometrial-cancer/detection-diagnosis-staging/survival-
rates.html. Accessed November 2020.
28. Alcázar JL, Gastón B, Navarro B, Salas R, Aranda J, Guer-
riero S. Transvaginal ultrasound versus magnetic resonance
imaging for preoperative assessment of myometrial infil-
tration in patients with endometrial cancer: a systematic
review and meta-analysis. J Gynecol Oncol 2017;28:e86.
29. Alcázar JL, Orozco R, Martinez-Astorquiza Corral T, et
al. Transvaginal ultrasound for preoperative assessment of
myometrial invasion in patients with endometrial cancer:
a systematic review and meta-analysis. Ultrasound Obstet
30. Andreano A,  Rechichi G,  Rebora P,  Sironi S,  Valsecchi
MG, Galimberti S. MR diffusion imaging for preoperative
staging of myometrial invasion in patients with endometrial
cancer: a systematic review and meta–analysis. Eur Radi-
ol 2014;24:1327– 1338.
31. Luomaranta A, Leminen A, Loukovaara M. Magnetic reso-
nance imaging in the assessment of high–risk features of
endometrial carcinoma. A meta–analysis.  Int J Gynecol
Cancer 2015;25:837– 842.
operative assessment myometrial invasion in patients with
endometrial carcinoma: a systematic review and meta–
analysis. Cancer Imaging 2014;14:32.
Review Med Ultrason 2022, Vol. 24, no. 1, 85-94
DOI: 10.11152/mu-2871

Diagnostic value of endobronchial ultrasound elastography

for differentiating benign and malignant hilar and mediastinal lymph
nodes: a systematic review and meta-analysis
Jiangfeng Wu*, Yue Sun*, Yunlai Wang, Lijing Ge, Yun Jin, Zhengping Wang
* the authors share the first authorship

Department of Ultrasound, The Affiliated Dongyang Hospital of Wenzhou Medical University, Dongyang, Zhejiang,
China

Abstract
Aims: In the present study, a meta-analysis was performed to evaluate the diagnostic value of endobronchial ultrasound
(EBUS) elastography for differentiating benign and malignant hilar and mediastinal lymph nodes (LNs). Material and
methods: A comprehensive literature search was carried out through PubMed, Embase, and Cochrane Library. Two authors
screened the papers and extracted the data independently and any discrepancies were resolved by discussion. The methodolog-
ical quality of each included study was assessed by the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Sensitivity,
specificity, positive likelihood ratio, negative likelihood ratio, and area under the curve were calculated to evaluate the value
of EBUS elastography for hilar and mediastinal LNs. Results: Seventeen studies with the number of 2307 LNs were included.
There was significant heterogeneity across the included studies. The pooled sensitivity, specificity, positive likelihood ratio,
negative likelihood ratio and diagnostic odds ratio for the diagnosis of hilar and mediastinal LNs by EBUS elastography were
0.90 (95% confidence interval [CI], 0.84-0.94), 0.78 (95% CI, 0.74-0.81), 4.1 (95% CI, 3.4-4.9), 0.12 (95% CI, 0.07-0.21)
and 33 (95% CI, 17-64), respectively. Furthermore, area under the curve was calculated to be 0.86 (95% CI, 0.82-0.88).
Conclusion: EBUS elastography is a valuable technology in the differentiation of benign and malignant hilar and mediastinal
LNs and could provide supplementary diagnostic information during endobronchial ultrasound-guided transbronchial needle
aspiration. The combination of EBUS elastography and B-mode EBUS could improve the diagnostic accuracy for hilar and
mediastinal LNs.
Keywords: endobronchial ultrasound; elastography; lymph nodes; diagnosis; meta-analysis

Introduction proaches [1,2]. The adequate treatment and prognosis of

Lung cancer is among the most common diagnosed
malignant tumours with a high mortality rate worldwide
and poor 5-year survival rate of only 16%, despite the
significant advances in diagnostic and therapeutic ap-
Received 30.11.2020  Accepted 20.02.2021
Med Ultrason
2022, Vol. 24, No 1, 85-94
Corresponding author: Jiangfeng Wu
Department of Ultrasound, The Affiliated
Dongyang Hospital of Wenzhou Medical
University, 60 Wuning West Road,
Dongyang 322100, Zhejiang, China
E-mail: wjfhospital@163.com
Phone: 18257937213
lung cancer is determined by a proper diagnosis and stag-
ing of the disease [2,3].
Endobronchial ultrasound-guided transbronchial nee-
dle aspiration (EBUS-TBNA) is considered as a highly
sensitive, specific, safe and minimally invasive tech-
nology, which has developed into the most widely used
standard sampling technique in the diagnosis of hilar and
mediastinal lymph nodes (LNs) [4,5]. Current interna-
tional guidelines recommend that EBUS-TBNA should
be carried out before mediastinoscopy, especially for the
pathological diagnosis and staging of LNs in patients
with lung cancer [4-6]. Compared with conventional me-
diastinoscopy, which is relevant to a surgical operation
under general anaesthesia, EBUS-TBNA has a compa-
rable diagnostic accuracy, has a relatively lower risk of
86 Jiangfeng Wu, Yue Sun et al Benign and malignant hilar and mediastinal lymph nodes & endobronchial US elastography
adverse events and is much more cost-efficient [5-8].
The evaluation of the LNs during endobronchial ultra-
sound (EBUS) according to conventional B-mode fea-
tures such as size, shape, echogenicity, distinct border,
central hilar structure and coagulation necrosis sign, has
been found helpful in the detection of LNs metastasis [9].
Furthermore, studies demonstrated that vascular pattern
on power Doppler mode during EBUS can also predict
nodal metastasis in patients with lung cancer. However, it
is still difficult for certain cases to predict LNs metastasis
only by conventional B-mode features [10,11].
Ultrasound elastography exhibits potential applica-
tion to the differential diagnosis of benign and malignant
LNs [12,13]. In recent years, EBUS elastography has
been introduced as a novel modality in the evaluation of
hilar and mediastinal LNs [14-17]. However, the diag-
nostic performance of EBUS elastography for the differ-
entiation of benign and malignant hilar and mediastinal
LNs is variable across previous studies, with the sensitiv-
ity ranging from 64% to 100% and the specificity ranging
from 65% to 92% [14-16].
Whether EBUS elastography can be considered as
a valuable scanning modality is still a controversial is-
sue. Therefore, we performed a meta-analysis to assess
the diagnostic performance of EBUS elastography in the
noninvasive discrimination between benign and malig-
nant hilar and mediastinal LNs.
Material and methods
Meta-analysis principles
The present meta-analysis was carried out on the basis
of the Preferred Reporting Items for Systematic Reviews
and Meta-Analyses (PRISMA) guidelines, which include
27 items and provide specific guidance for the reporting
of systematic reviews [18]. We registered our protocol
with the International Platform of Registered Systematic
Review and Meta-analysis Protocols (INPLAY) (regis-
tration number: INPLASY2020110117).
Search strategy
The Pubmed, EMBASE and Cochrane Library were
systematically searched from inception to November
2020, to identify English-language studies on EBUS elas-
tography for differentiating benign and malignant LNs.
The search terms were as follows: “elasticity”, “elasto-
grams”, “elastography”, “elastosonography”, “elasticity
imaging”, “endobronchial ultrasound”, “EBUS”, “EBUS
elastography”, “lymph node”, “lymph nodes”, “lymphad-
enopathy”, and “lymphaden”. Detailed search terms are
provided in supplementary file 1. Reference lists of the
included studies and relevant reviews were also screened
to find additional relevant studies.
Inclusion and exclusion criteria
Two reviewers (JW and YS) screened the titles and
the abstracts of retrieved studies independently. Before
identifying the literature, they established the inclusion
and exclusion criteria together to increase validity and
reproducibility. Inconsistencies between the reviewers
were resolved through discussion.
The inclusion criteria were as follows: (1) diagnostic
studies were included; (2) studies evaluating the diag-
nostic accuracy of EBUS elastography in distinguishing
malignant and benign hilar and mediastinal LNs were in-
cluded; (3) a reference standard was adopted to confirm
malignant intrathoracic LNs, such as cytology obtained
by EBUS-TBNA or other method, histology of surgical
resection, or more than 6 months of follow-up; (4) if sev-
eral diagnostic methods were used in a study, only the
best result was chosen.
The exclusion criteria were as follows: (1) case re-
ports, letters, consensus statements, and unpublished
articles; (2) studies without sufficient data to construct
diagnostic 2x2 tables; (3) studies that contained an over-
lapped population.
Data extraction
Two reviewers (JW and YS) independently extracted
the relevant data from the eligible studies using a pre-
designed data collection form. Any discrepancies were
resolved by discussion with the senior author. For eligi-
ble studies, the following items were extracted: last name
of the first author, year of publication, country, study
type, sample method, malignancy prevalence, diagnostic
method, blinding method, cut off, ultrasound equipment,
probe frequency, sample size, lymph node, short-axis di-
ameter, mean age, gender, standard reference, time be-
tween EBUS elastography and the standard reference.
Study quality assessment
The Quality Assessment of Diagnostic Accuracy
Studies-2 (QUADAS-2) tool [19] was used to assess the
risk of bias and methodological quality. The quality of
each eligible study was assessed by an appraisal of the
risk of bias of four domains and clinical applicability of
three domains of the study characteristics. Four domains
included patient selection, index test, reference standard
and flow and timing. Every domain was assessed for risk
of bias, and the first three domains were assessed for ap-
plicability. The quality assessment was carried out using
RevMan 5.3 software (Nordic Cochrane Centre, Copen-
hagen, Denmark).
Statistical analysis
This meta-analysis was performed by StataSE 15
(Stata Corporation, College Station, Texas). All statisti-
cal analyses were performed by one author (JW), who
has experience in performing meta-analysis. The pooled
 Med Ultrason 2022; 24(1): 85-94 87
estimates of sensitivity, specificity, positive likelihood
ratio (PLR), negative likelihood ratio (NLR) and diag-
nostic odds ratio (DOR) with corresponding 95% con-
fidence intervals (CIs) were calculated by a bivariate
random effect model in this study, which indicated the
accuracy of EBUS elastography in the differentiation of
benign and malignant hilar and mediastinal LNs. Fur-
thermore, the summary receiver operator curve (SROC)
was constructed and the area under the curve (AUC)
was calculated. An AUC close to 0.5 reveals a poor test,
while an AUC close to 1.0 shows a perfect diagnostic
test [20]. The inconsistency index (I2) and the Cochrane
Q test were utilized to evaluate the heterogeneity among
included studies with a p-value <0.1 or I2 >50% indicat-
ing significant heterogeneity [21]. The Deeks’ funnel
plot asymmetry test was applied to evaluate publication
bias [22], through a p-value >0.05 showing no significant
publication bias.
Meta-regression analyses utilizing several covari-
ates were performed to investigate the potential causes
of heterogeneity: study design (prospective versus other),
year published (2014-2017 versus 2019-2020), reference
standard (pathology versus pathology or follow-up), di-
agnostic method (quantitative versus qualitative) and
sampling (consecutive versus others). Fig 1. Flowchart of study selection

Results enroll patients [36]. The sample size of the enrolled

Study selection
In the three databases, our literature search yielded
110 publications for consideration. Figure 1 shows the
flow chart of the study selection. Finally, 17 publications
were included in this meta-analysis [14-16,23-36].
Characteristics of the included studies
The 17 enrolled studies were published between
2014-2020 and written in English [14-16,23-36]. A to-
tal of 1360 patients with 2307 hilar and mediastinal LNs
were included in these studies. Five studies were per-
formed in China [15,23,27-29], 3 in Japan [14,24,35],
2 in Netherlands [16,33], 1 in Slovenia [25], 1 in Thailand
[26], 1 in Taiwan [30], 1 in Spain [31], 1 in France [32],
1 in Turkey [34] and 1 in Germany [36]. There were 10
prospective [15,16,25,26,29,31-34,36] and six retrospec-
tive [14,24,27,28,30,35] studies and one study did not de-
pict the study design [23]. Double blinding between the
index test and standard reference was found in ten stud-
ies [15,24,25,27,28,31,32,34-36] and single blinding in
four studies [14,23,29,33]. Three studies [16,26,30] did
not report a blinding method. The patients were consecu-
tively enrolled in eight studies [14,15,24,25,29,32,34,35]
and eight studies did not report the sample method
[16,23,26-28,30,31,33]. One study did not consecutively
studies ranged from 21 to 327 and the mean age ranged
from 56 to 68 years. The malignancy prevalence of LNs
ranged from 33% to 80% across different studies. The
mean size of short-axis diameter of LNs ranged from 8
to 20 mm. Nine studies [16,23-26,29,33-35] used quan-
titative diagnostic methods (strain ratio, stiff area ratio,
blue colour proportion, or strain histogram) while eight
studies used qualitative diagnostic methods (3, 4 or 5 col-
our classifications) [14,15,27,28,30-32,36]. Table I and II
epitomizes the data extracted from the enrolled studies.
More details are showed in supplementary file 2.
Quality assessment
The quality assessment results of the risk of bias
and applicability concerns of the enrolled studies were
shown graphically in figure 2. With respect to the patient
selection domain, eight studies were considered as “un-
known” [16,23,26-28,30,31,33], because they did not re-
port the sample method. One was considered as having
high bias because the patients were not enrolled consecu-
tively [36]. Six studies were considered as having high
bias because part of the patients were excluded inappro-
priately [14,15,28,30,32,36]. Concerning the index test
domain, 6 studies [14,16,23,26,29,30] were considered
as “unknown” because the blinded status of the golden
standard was not explicitly reported. Concerning the ref-
88 Jiangfeng Wu, Yue Sun et al Benign and malignant hilar and mediastinal lymph nodes & endobronchial US elastography
Table I. Primary data extracted from included studies for meta-analysis
Author Country Study Blind Sample Pa- Male/ Age (year) Lymph Malig- Malignancy
type method method tient female node nant prevalence
(n) (n) node (n) (%)
Izumo Japan Retro Single Consecutive 30 17/13 67.1±15.5 75 42 56
[14], 2014 blinded
He China NR Single NR 40 26/14 65 (median) 68 42 62
[23], 2015 blinded
Nakajima Japan Retro Double Consecutive 21 15/6 63 (30–80) 49 16 33
[24], 2015 blinded
Rozman Slovenia Pro Double Consecutive 33 25/8 67.5±8.2 80 34 43
[25], 2015 blinded
Korrungruang Thailand Pro NR NR 72 41/31 58.3±12.5 120 96 80
[26], 2017
Sun China Retro Double NR 56 32/24 56 68 35 51
[27], 2017 blinded
Gu China Pro Double Consecutive 60 49/11 62 (26-82) 133 89 67
[15], 2017 blinded
Huang China Retro Double NR 47 29/18 60.19±11.03 78 33 42
[28], 2017 blinded
Ma China Pro Single Consecutive 60 40/20 61.6 79 39 49
[29], 2018 blinded
Lin Taiwan Retro NR NR 94 65/29 62.8 (20-97) 206 74 36
[30], 2019
Roca Spain ProDouble NR 27 16/11 68±10 42 18 43
[31], 2019 blinded
Fournier France Pro Double Consecutive 114 80/34 60.2±12.2 217 120 55
[32], 2019 blinded
Verhoeven Netherlands Pro Single NR 63 39/24 64.3 (41–83) 120 45 38
[33], 2019 blinded
Çağlayan Turkey Pro Double Consecutive 119 69/50 63.2±12.4 221 93 42
[34], 2020 blinded
Uchimura Japan Retro Double Consecutive 132 91/41 71 (median) 149 68 46
[35], 2020 blinded
Gompelmann Germany Pro Double Not 65 34/31 63 (52-82) 77 55 71
[36], 2020 blinded consecutive
Verhoeven Netherlands Pro NR NR 327 200/127 66 (26-90) 525 253 48
[16], 2020
Pro, prospective; Retro, retrospective; NR, not reported

erence standard domain, 4 studies [16,26,30,33] were
considered as “unknown” because the blinded status of
index test was not definitely reported. With regard to
the flow and timing domain, 6 studies were considered
as having high bias because they did not enroll all the
patients for analysis [14,15,28,30,32,36]. Regarding ap-
plicability, for patient selection, index test and reference
standard domains, all studies were considered to have
low concerns.
Data synthesis and publication bias
Summaries of diagnostic sensitivity and specificity of
EBUS elastography for differentiating malignant and be-
nign hilar and mediastinal LNs were analysed by the ran-
dom effects method on the basis of significant statistical
heterogeneity (I2=89.56% for sensitivity; I2=69.31% for
specificity). Overall, the pooled sensitivity and specific-
ity of EBUS elastography were 0.90 (95% CI, 0.84-0.94)
and 0.78 (95% CI, 0.74-0.81; fig 3). The pooled PLR,
NLR, and DOR of EBUS elastography were 4.1 (95%
CI, 3.4-4.9), 0.12 (95% CI, 0.07-0.21), and 33 (95% CI,
17-64), respectively and the post-test probability was
50% and 3% (fig 4). The AUC under the SROC curve for
the value of EBUS elastography in the diagnosis of LNs
was 0.86 (95% CI, 0.82-0.88; fig 5).
The Deeks’ funnel plot was carried out to evaluate
the publication bias of the eligible studies. However, as
 Med Ultrason 2022; 24(1): 85-94 89
Table II. Characteristics of the included studies
Author Diagnostic Cut-off Short-axis Reference Time between US Sen Spe
method value diameter standard reference and equipment (%) (%)
(mm) index test
Izumo 3 classifications Predominantly 15.0 median Pathology After EBUS BF-UC260FW 100 92
[14] blue (5.0–50.0)
He Strain ratio 32.07 17 median Pathology After EBUS EB‑1970UK 88 81
[23]
Nakajima Stiff area ratio 0.311 8.44 (5.0–21.7) Pathology After EBUS BF-UC260FW 81 85
[24] or follow-up
Rozman Strain ratio 8 11.1±4.5 Pathology After EBUS BF-UC180F 88 85
[25] or follow-up
Korrungruang Strain ratio 2.5 18.8±7.9 Pathology After EBUS BF-UC180F 100 71
[26] or follow-up
Sun 5 classifications Score: 4-5 19.92±9.09 Pathology After EBUS EB1970, Pentax, 86 82
[27] or follow-up Japan
Gu 3 classifications Predominantly NR Pathology After EBUS BF-UC260FOL8 100 65
[15] blue and EU-C2000
Huang 3 classifications Predominantly NR Pathology After EBUS BF-UC260F and 96 87
[28] blue NA-201SX-4022
Ma Blue color 36.70% NR Pathology After EBUS BF-UC260FW 92 68
[29] proportion
Lin 3 classifications Predominantly 13.6 (2.0-56.3) Pathology After EBUS EU-ME2 91 83
[30] blue or follow-up PREMIER PLUS
Roca 3 classifications Predominantly 15 ± 6 Pathology After EBUS PENTAX, 72 92
[31] blue or follow-up EB 1970 UK
Fournier 3 classifications Predominantly 16.2 ± 8.5 Pathology After EBUS OlympusBF 87 68
[32] blue UC 180F and
OlympusEU-ME2
Verhoeven Strain histogram 78 9.95 (4–26) Pathology After EBUS Pentax 93 75
[33] or follow-up EB-1970UK
echo-endoscopes
Çağlayan Strain ratio 2.47 16.2±11.1 Pathology After EBUS BF-UC180F 75 65
[34]
Uchimura Stiffness area 0.41 8 Pathology After EBUS BF-UC260FW 88 80
[35] ratio or follow-up
Gompelmann 3 classifications Predominantly NR Pathology After EBUS NR 71 67
[36] blue
Verhoeven Strain histogram 78 12.3 (3-50) Pathology After EBUS Pentax 64 76
[16] or follow-up EB1970UK and
EB19-J10U
Sen, sensitivity; Spe, specificity; NR,not reported; US, ultrasound; EBUS, endobronchial ultrasound

shown in figure 6, which indicated the publication bias
existed (p=0.00). This indicated publication bias might
be part of source of heterogeneity.
Meta-regression and subgroup analyses
Due to the significant heterogeneity among studies,
a meta-regression analysis was performed to explore
other potential sources of heterogeneity. The covariables
included study design (prospective versus others), year
of publication (2014-2017 versus 2018-2020), reference
standard (pathology versus pathology or follow-up), di-
agnostic method (quantitative versus qualitative), and
sampling (consecutive versus others). Among the various
potential covariates, study design and year of publication
were associated with the heterogeneity of the sensitivity,
while study design, year of publication, reference stand-
ard, diagnostic method, and sampling were associated
with the heterogeneity of the specificity (fig 7).
The subgroups with respect to study design, prospec-
tive studies had a lower diagnostic performance to others
(sensitivity: 0.89 and 0.92, p=0.04; specificity: 0.73 and
0.83, p=0.00). The subgroups with regard to year of pub-
lication, studies published between 2018 and 2020 had a
90 Jiangfeng Wu, Yue Sun et al Benign and malignant hilar and mediastinal lymph nodes & endobronchial US elastography

Fig 4. Fagan nomogram for detecting malignant hilar and me-

diastinal LNs

Fig 2. Quality assessment of the included studies using QUA-

DAS-2 tool

Fig 5. Summary receiver operating characteristics curve of

EBUS elastography for malignant hilar and mediastinal LNs

Fig 3. Forest diagrams of EBUS elastography for differentiat-

ing hilar and mediastinal LNs

lower diagnostic performance to studies between 2014

and 2017 (sensitivity: 0.84 and 0.95, p=0.00; specificity:
0.75 and 0.82, p=0.00). The subgroups concerning ref- Fig 6. Funnel plot of EBUS elastography for malignant hilar
erence standard, the sensitivity between pathology and and mediastinal LNs
 Med Ultrason 2022; 24(1): 85-94 91
Discussion

In this meta-analysis, including 17 studies and 2307

Fig 7. Meta-regression and subgroup analyses
pathology or follow-up was similar (sensitivity: 0.92 and
0.89, p=0.33), but the specificity of subgroup of pathol-
ogy or follow-up was higher (specificity: 0.80 and 0.73,
p=0.00). The sensitivity between quantitative method
and qualitative method was similar (sensitivity: 0.89 and
0.92, p=0.05), while the specificity of qualitative method
was higher (specificity: 0.80 and 0.76, p=0.00). Finally,
a similar sensitivity was detected between consecutive
sampling and others (sensitivity: 0.92 and 0.89, p=0.24),
while the specificity of other sampling methods was
higher (specificity: 0.79 and 0.76, p=0.00) (Table III).
hilar and mediastinal LNs, we have evaluated the diag-
nostic accuracy of EBUS elastography for the identifi-
cation of hilar and mediastinal LNs. We found that the
pooled sensitivity, specificity and DOR of EBUS elas-
tography for identifying hilar and mediastinal LNs were
0.90 (95% CI, 0.84-0.94), 0.78 (95% CI, 0.74-0.81), and
33 (95% CI, 17-64), respectively. The PLR, NLR, and
SROC AUC were 4.1 (95% CI, 3.4-4.9), 0.12 (95% CI,
0.07-0.21), and 0.86 (95% CI, 0.82-0.88), respectively.
The findings of this meta-analysis show that EBUS elas-
tography has a well diagnostic performance for hilar and
mediastinal LNs.
A previous meta-analysis in 2018 [37], which com-
prised a total of 504 patients, depicted 7 studies, pub-
lished between 2014 and 2017, evaluating the diagnostic
accuracy of EBUS elastography in the diagnosis of hilar
and mediastinal LNs. The 7 studies were also included
in our meta-analysis. The pooled sensitivity, specificity,
PLR, NLR, and DOR were 0.93 (95% CI, 0.85-0.97),
0.85 (95% CI, 0.78-0.90), 6.3 (95% CI, 4.2-9.2), 0.08
(95% CI, 0.04-0.18), and 74 (95% CI, 33-168), respec-
tively. Compared to it, our meta-analysis comprising
a total of 1206 patients reported 17 studies and the in-
cluded studies were published between 2014 and 2020.
Although our results are somewhat lower to the previous
study, our meta-analysis includes more additional objec-
tive studies (17 studies versus 7 studies) to support clini-
cal practice of EBUS elastography for the diagnosis of
hilar and mediastinal LNs.

Table III. Meta-regression and subgroup analyses

Covariate N Sensitivity (95% CI) p-value Specificity (95% CI) p-value
Study design 0.04* 0.00*
Prospective 10 0.89 (0.82 - 0.96) 0.73 (0.69 - 0.77)
Others 7 0.92 (0.85 - 0.99) 0.83 (0.79 - 0.87)
Year of publication 0.00* 0.00*
2018-2020 9 0.84 (0.75 - 0.92) 0.75 (0.70 - 0.80)
2014-2017 8 0.95 (0.91 - 0.99) 0.82(0.76 - 0.87)
Reference standard 0.33 0.00*
Pathology 8 0.92 (0.86 - 0.99) 0.73 (0.67 - 0.78)
Pathology or follow-up 9 0.89 (0.81 - 0.96) 0.80 (0.77 - 0.84)
Diagnostic method 0.05 0.00*
Quantitative 9 0.89 (0.82 - 0.97) 0.76 (0.71 - 0.81)
Qualitative 8 0.92 (0.85 - 0.98) 0.80 (0.75 - 0.86)
Sampling 0.24 0.00*
Consecutive 8 0.92 (0.85 - 0.98) 0.76 (0.70 - 0.81)
Others 9 0.89 (0.81 - 0.96) 0.79 (0.75 - 0.84)
N, number of studies; *, statistical significance (p<0.05); CI, confidence interval
92 Jiangfeng Wu, Yue Sun et al Benign and malignant hilar and mediastinal lymph nodes & endobronchial US elastography
EBUS elastography used qualitative diagnostic meth-
ods or quantitative diagnostic methods for diagnosing
hilar and mediastinal LNs in eligible studies. Qualita-
tive diagnostic methods in enrolled studies included
the 3-, 4-, or 5-image pattern classifications and quan-
titative diagnostic methods included the stiffness area
ratio, the strain histogram method, and the strain ratio.
Generally, unlike qualitative diagnostic methods, quan-
titative diagnostic methods could avoid the disadvan-
tage relative to subjective evaluations by using specific
numerical values [16,33]. In the subgroup analysis, 9
studies [16,23-26,29,33-35] using quantitative diagnos-
tic methods showed pooled sensitivity of 0.89 (95% CI,
0.81-0.96) and specificity of 0.76 (95% CI, 0.71-0.81);
in comparison, the pooled sensitivity of the 8 stud-
ies [14,15,27,28,30-32,36] using qualitative diagnostic
methods was 0.92 (95% CI, 0.85-0.98) and the specific-
ity was 0.80 (95% CI, 0.75-0.86). A higher specificity of
qualitative diagnostic methods was observed (p=0.00);
in contrast, the sensitivities were comparative (p=0.05).
However, in three studies using qualitative diagnostic
methods [14,15,28,30,32,36], the comparative analysis
was performed between LNs with colour type 1 (not pre-
dominantly blue) and 3 (predominantly blue) and LNs
with a mixed colour (type 2) were excluded. This might
have resulted in patient selection bias. On the other hand,
qualitative methods are operator dependent and may
not provide an objective diagnosis. Therefore, the result
should be interpreted with caution and large prospective
studies are still required to verify the present result.
Ultrasound elastography is mainly used to detect the
tissue stiffness [38,39]. Several previous studies found
out that some malignant LNs were very soft, while some
benign LNs appeared quite hard [26,32,40,41]. Possible
explanations for the findings may lie in the high vascular
invasion or necrotic structure of malignant LNs that ap-
pear soft under elastographic evaluation. By comparison,
nonspecific inflammation, having areas of hard fibrotic or
anthracotic tissues frequently occur in benign LNs and,
therefore, the strain within the node may increase as the
“stiffer” components of the node are assessed, ultimately
influencing the application of elastography techniques.
This confirms again that EBUS elastography reveals tis-
sue stiffness and not malignancy or benignity.
Studies that compared conventional EBUS imaging
to EBUS elastography have shown that EBUS elastogra-
phy is superior to imaging by EBUS alone [14,23,25,30].
However, a prospective study by Gu et al [15] showed
that EBUS elastography in combination with conven-
tional EBUS B-mode features could improve the speci-
ficity from 65% to 72.7%. The study by Fujiwara et al
[17] also demonstrated that the sensitivity and specific-
ity of the stiffer area comprised more than 31% of the
entire lymph node area for diagnosing malignancy were
72.1% and 84%, and by adding sonographic features to
the elastographic findings, the sensitivity and specificity
improved to 93.7% and 89.4%. A retrospective study of
149 mediastinal and hilar LNs by Uchimura et al [35]
reported that by adding B-mode sonography to EBUS
elastography, the sensitivity increased to 98.3%. There-
fore, the combination of EBUS elastography and B-mode
EBUS may lead to a higher diagnostic accuracy than
either technology alone for differentiating benign and
malignant LNs. In clinical practice, the procedure of B-
mode EBUS is always before EBUS elastography, so the
combination of EBUS elastography and B-mode EBUS
for assessing the LNs is more feasible.
As the significant heterogeneity in the present study
(sensitivity: I2=89.56%, p=0.00; specificity: I2=69.31%,
p=0.00), meta-regression analyses were carried out to ex-
plore the sources of heterogeneity. We found that study
design and year of publication accounted for part of the
significant sources of heterogeneity in terms of sensitiv-
ity and study design, year of publication, reference stand-
ard, diagnostic method and sampling in terms of specific-
ity. Furthermore, the indicated publication bias might be
another source of heterogeneity. On the other hand, there
were other factors such as specialties of EBUS perform-
ers, different levels of experience and different equip-
ment, which might also have played a role in significant
heterogeneity among studies. Further meta-regression
analyses was not carried out to explore the sources of
heterogeneity on the basis of the factors referred above
because of the insufficient information supplied in the
eligible studies.
EBUS elastography cannot currently replace the more
accurate EBUS-TBNA for differentiating hilar and medi-
astinal LNs, which has been considered as the first-line
technology for hilar and mediastinal lymph node staging
of lung cancer in many clinic guidelines [42,43]. How-
ever, EBUS elastography as a supplemental modality
could be helpful for operators to perform EBUS-TBNA
efficiently. It could help in selecting the most suspicious
lymph node for biopsy in stations where multiple LNs
are found and help operators to choose harder areas of
LNs for fine needle aspiration (FNA) to avoid necrosis
or blood vessels, then improving the quality of the speci-
mens and decreasing the number of punctures [14,23,24].
It is important to consider several limitations regard-
ing this study. First, only studies written in English were
enrolled, which may result in potential selection bias.
Second, only several included studies [16,26,27,32,36]
evaluated the intra- or inter-observer variability. There-
fore, further studies are needed to evaluate the variabil-

ity. Third, the diagnostic accuracy of EBUS elastogra-
phy was acquired from enrolled studies which provided
various elastic parameters with diverse threshold values,
which may result in heterogeneity. Therefore, the clas-
sification of LNs by EBUS elastography still needs to be
further explored, developed and improved because of the
lack of uniform standards. Finally, most of the enrolled
studies had methodological disadvantages, especially in
domains such as patient selection, the index test, refer-
ence standard and flow and timing, so improvements in
the future study design are needed to address the issue.
Conclusion
In summary, EBUS elastography is a valuable tech-
nology in the differentiation of benign and malignant hilar
and mediastinal LNs with high sensitivity and moderate
specificity, which may provide supplementary diagnostic
information, increase the diagnostic yield, and reduce the
number of unnecessary biopsies during EBUS-TBNA.
Furthermore, the combination of EBUS elastography and
B-mode EBUS could improve the diagnostic accuracy
for hilar and mediastinal LNs. However, the conclusion
of this study should be interpreted with caution because
of the significant heterogeneity and the publication bias.
Large prospective international multicentre studies are
still required to support the present conclusion.
Conflict of interest: none
References
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA
Cancer J Clin 2017;67:7-30.  endobronchial elastography and sonographic findings dur-
2. Bremnes RM, Busund LT, Kilvær TL, et al. The role of
tumor-infiltrating lymphocytes in development, progres-
sion, and prognosis of non-small cell lung cancer. J Thorac
Oncol 2016;11:789-800.
3. Woodard GA, Jones KD, Jablons DM. Lung cancer staging
and prognosis. Cancer Treat Res 2016;170:47-75.  diagnostic test accuracy studies: The PRISMA-DTA State-
4. Divisi D, Zaccagna G, Barone M, Gabriele F, Crisci R.
Endobronchial ultrasound-transbronchial needle aspiration
(EBUS/TBNA): a diagnostic challenge for mediastinal le-
sions. Ann Transl Med 2018;6:92.
5. Naur TMH, Nilsson PM, Pietersen PI, Clementsen PF,
Konge L. Simulation-based training in flexible bronchos-
copy and endobronchial ultrasound-guided transbron-
chial needle aspiration (EBUS-TBNA): a systematic re-
view. Respiration 2017;93:355-362.  views of evaluations of diagnostic and screening tests. BMJ
6. Medford AR, Bennett JA, Free CM, Agrawal S. Mediasti-
nal staging procedures in lung cancer: EBUS, TBNA and
mediastinoscopy. Curr Opin Pulm Med 2009;15:334-342.
7. Yasufuku K, Pierre A, Darling G, et al. A prospective con-
trolled trial of endobronchial ultrasound-guided transbron-
Med Ultrason 2022; 24(1): 85-94 93
chial needle aspiration compared with mediastinoscopy for
mediastinal lymph node staging of lung cancer.  J Thorac
Cardiovasc Surg 2011;142:1393-400.e1.
8. Verdial FC, Berfield KS, Wood DE, et al. Safety and costs
of endobronchial ultrasound-guided nodal aspiration and
mediastinoscopy. Chest 2020;157:686-693. 
9. Hylton DA, Turner J, Shargall Y, et al. Ultrasonographic
characteristics of lymph nodes as predictors of malignancy
during endobronchial ultrasound (EBUS): a systematic re-
view. Lung Cancer 2018;126:97-105. 
10. Darwiche K, Özkan F, Wolters C, Eisenmann S. Endobron-
chial ultrasound (EBUS) - update 2017. Ultraschall Med
2018;39:14-38.
11. Agrawal S, Goel AD, Gupta N, Lohiya A, Gonuguntla HK.
Diagnostic utility of endobronchial ultrasound (EBUS)
features in differentiating malignant and benign lymph
nodes - a systematic review and meta-analysis. Respir Med
2020;171:106097.
12. Ophir J, Alam SK, Garra B, et al. Elastography: ultrason-
ic estimation and imaging of the elastic properties of tis-
sues. Proc Inst Mech Eng H 1999;213:203-233.
13. Carlsen J, Ewertsen C, Sletting S, et al. Ultrasound elas-
tography in breast cancer diagnosis.  Ultraschall Med
2015;36:550-562.
14. Izumo T, Sasada S, Chavez C, Matsumoto Y, Tsuchida T.
Endobronchial ultrasound elastography in the diagnosis
of mediastinal and hilar lymph nodes.  Jpn J Clin Oncol
2014;44:956-962. 
15. Gu Y, Shi H, Su C, et al. The role of endobronchial ultra-
sound elastography in the diagnosis of mediastinal and hilar
lymph nodes. Oncotarget 2017;8:89194-89202. 
16. Verhoeven RLJ, Trisolini R, Leoncini F, et al. Predictive
value of endobronchial ultrasound strain elastography in
mediastinal lymph node staging: the e-predict multicenter
study results. Respiration 2020;99:484-492.
17. Fujiwara T, Nakajima T, Inage T, et al. The combination of
ing endobronchial ultrasound-guided transbronchial needle
aspiration for predicting nodal metastasis.  Thorac Cancer
2019;10:2000-2005. 
18. McInnes MDF, Moher D, Thombs BD, et al. Preferred re-
porting items for a systematic review and meta-analysis of
ment. JAMA 2018;319:388-396.
19. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2:
a revised tool for the quality assessment of diagnostic ac-
curacy studies. Ann Intern Med 2011;155:529-536.
20. Hanley JA, McNeil BJ. The meaning and use of the area
under a receiver operating characteristic (ROC) curve. Ra-
diology 1982;143:29-36.
21. Deeks JJ. Systematic reviews in health care: systematic re-
2001;323:157-162. 
22. Deeks JJ, Macaskill P, Irwig L. The performance of tests
of publication bias and other sample size effects in system-
atic reviews of diagnostic test accuracy was assessed. J Clin
Epidemiol 2005;58:882-893.
94 Jiangfeng Wu, Yue Sun et al Benign and malignant hilar and mediastinal lymph nodes & endobronchial US elastography
23. He HY, Huang M, Zhu J, Ma H, Lyu XD. Endobronchial
ultrasound elastography for diagnosing mediastinal and hi-
lar lymph nodes. Chin Med J (Engl) 2015;128:2720-2725.  34. Çağlayan B, İliaz S, Bulutay P, et al. The role of endo-
24. Nakajima T, Inage T, Sata Y, et al. Elastography for predict-
ing and localizing nodal metastases during endobronchial
ultrasound. Respiration 2015;90:499-506.
25. Rozman A, Malovrh MM, Adamic K, Subic T, Kovac V,
Flezar M. Endobronchial ultrasound elastography strain
ratio for mediastinal lymph node diagnosis. Radiol Oncol
2015;49:334-340.
26. Korrungruang P, Boonsarngsuk V. Diagnostic value of
endobronchial ultrasound elastography for the differ-
entiation of benign and malignant intrathoracic lymph
nodes. Respirology 2017;22:972-977.
27. Sun J, Zheng X, Mao X, et al. Endobronchial ultrasound
elastography for evaluation of intrathoracic lymph nodes: a
pilot study. Respiration 2017;93:327-338.
28. Huang H, Huang Z, Wang Q, et al. Effectiveness of the
benign and malignant diagnosis of mediastinal and hilar
lymph nodes by endobronchial ultrasound elastography.  J
Cancer 2017;8:1843-1848.
29. Ma H, An Z, Xia P, et al. Semi-quantitative analysis of
EBUS elastography as a feasible approach in diagnos-
ing mediastinal and hilar lymph nodes of lung cancer pa-
tients. Sci Rep 2018;8:3571.
30. Lin CK, Yu KL, Chang LY, et al. Differentiating malignant
and benign lymph nodes using endobronchial ultrasound
elastography. J Formos Med Assoc 2019;118:436-443.
31. Hernández Roca M, Pérez Pallarés J, Prieto Merino D, et al.
Diagnostic value of elastography and endobronchial ultra-
sound in the study of hilar and mediastinal lymph nodes. J
Bronchology Interv Pulmonol 2019;26:184-192.
32. Fournier C, Dhalluin X, Wallyn F, et al. Performance of
endobronchial ultrasound elastography in the differentia-
tion of malignant and benign mediastinal lymph nodes: re-
sults in real-life practice.  J Bronchology Interv Pulmonol
2019;26:193-198.
33. Verhoeven RLJ, de Korte CL, van der Heijden EHFM.
Optimal endobronchial ultrasound strain elastography
assessment strategy: an explorative study.  Respiration
2019;97:337-347.
bronchial ultrasonography elastography for predicting
malignancy.  Turk Gogus Kalp Damar Cerrahisi Derg
2020;28:158-165.
35. Uchimura K, Yamasaki K, Sasada S, et al. Quantitative
analysis of endobronchial ultrasound elastography in com-
puted tomography-negative mediastinal and hilar lymph
nodes. Thorac Cancer 2020;11:2590-2599.
36. Gompelmann D, Kontogianni K, Sarmand N, et al. Endo-
bronchial ultrasound elastography for differentiating benign
and malignant lymph nodes. Respiration 2020;99:779-783.
37. Chen YF, Mao XW, Zhang YJ, et al. Endobronchial ul-
trasound elastography differentiates intrathoracic lymph
nodes: a meta-analysis. Ann Thorac Surg 2018;106:1251-
1257.
38. Krouskop TA, Wheeler TM, Kallel F, Garra BS, Hall T.
Elastic moduli of breast and prostate tissues under com-
pression. Ultrason Imaging 1998;20:260-274.
39. Lyshchik A, Higashi T, Asato R, et al. Elastic moduli of
thyroid tissues under compression.  Ultrason Imaging
2005;27:101-110.
40. Seo M, Sohn YM. Differentiation of benign and metastatic
axillary lymph nodes in breast cancer: additive value of
shear wave elastography to B-mode ultrasound. Clin Imag-
ing 2018;50:258-263.
41. Bae SJ, Park JT, Park AY, et al. Ex vivo shear-wave elas-
tography of axillary lymph nodes to predict nodal metasta-
sis in patients with primary breast cancer. J Breast Cancer
2018;21:190-196.
42. VanderLaan PA, Wang HH, Majid A, Folch E. Endo-
bronchial ultrasound-guided transbronchial needle as-
piration (EBUS-TBNA): an overview and update for
the cytopathologist.  Cancer Cytopathol 2014;122:561-
576. 
43. Herth FJ, Krasnik M, Vilmann P. EBUS-TBNA for the
diagnosis and staging of lung cancer. Endoscopy 2006;38
Suppl 1:S101-S105.
Review Med Ultrason 2022, Vol. 24, no. 1, 95-106
DOI: 10.11152/mu-3217

How to perform shear wave elastography. Part I

Giovanna Ferraioli1, Richard G Barr2, André Farrokh3, Maija Radzina4, Xin Wu Cui5, Yi Dong6,
Laurence Rocher7, Vito Cantisani8, Eleonora Polito8, Mirko D’Onofrio9, Davide Roccarina10,11,
Yasunobu Yamashita12, Manjiri K. Dighe13, Daniela Fodor14, Christoph F Dietrich15

1Department of Clinical, Surgical, Diagnostic and Pediatric Sciences, Medical School University of Pavia, Pavia, Italy,
2Northeastern Ohio Medical University, Rootstown, Ohio, USA, 3University Hospital Schleswig-Holstein, Campus
Kiel, Department of Gynecology and Obstetrics, Germany, 4Diagnostic Radiology Institute, Paula Stradins Clinical
University Hospital, Riga, Latvia, 5Sino-German Tongji-Caritas Research Center of Ultrasound in Medicine, Department
of Medical Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology,
Wuhan, China, 6Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai, China, 7Service de
Radiologie, APHP Hôpitaux Paris Saclay, Hôpital Antoine Béclère, Clamart, France. Université Paris Saclay,
Le Kremlin-Bicêtre, France, 8Department of Radiology, Oncology, Anatomo-Pathology, Sapienza-University of Rome,
Rome, Italy, 9Department of Radiology, University Hospital G.B. Rossi, University of Verona, Italy, 10Sheila Sherlock
Liver Unit and UCL Institute for Liver and Digestive Health, Royal Free Hospital, London, UK, 11SOD Medicina
Interna ed Epatologia, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy, 12Second Department of Internal
Medicine, Wakayama Medical University, Wakayama, Japan, 13Department of Radiology, University of Washington,
Seattle, USA, 142nd Internal Medicine Department, “Iuliu Hatieganu” University of Medicine and Pharmacy,
Cluj-Napoca, Romania, 15Department Allgemeine Innere Medizin (DAIM), Kliniken Hirslanden Beau Site, Salem
und Permancence, Bern, Switzerland

Abstract
We recently introduced a series of papers describing how to do certain techniques. This article is the first part of a review
of shear wave elastography (SWE). It reports the principles and interpretation of the technique and describes how to optimize
it. Normal values, pitfalls and artefacts for the examination of liver, breast. thyroid and salivary gland with shear wave elastog-
raphy are presented. The manuscript provides specific tips for applying SWE as part of a diagnostic US examination.
Keywords: ultrasound; elastography; elastometry; technique

Introduction plied force either by palpation or an Acoustic Radiation

Ultrasound (US) elastography is a method to deter-
mine tissue stiffness. It is similar to palpation used in
the physical examination. According to the current EF-
SUMB [1-5] and WFUMB guidelines [6-9] two types of
US elastography can be defined: strain elastography (SE)
and shear wave elastography (SWE). Strain uses an ap-
Received 30.01.2021  Accepted 15.03.2021
Med Ultrason
2022, Vol. 24, No 1, 95-106
Corresponding author: Prof. Christoph F. Dietrich
Department Allgemeine Innere Medizin (DAIM),
Kliniken Hirslanden Beau Site,
Salem und Permancence, Bern, Switzerland
Phone: +41798347180
E-mail: c.f.dietrich@googlemail.com
Force Impulse (ARFI) method to create and receive in-
formation about tissue displacement associated with the
elastic restoring forces in the tissue that act against shear
deformation as a function of time and space to display
biomechanical properties. This method has been previ-
ously described in detail [10,11].
Aim
Two articles on “how to perform strain imaging tech-
niques” have been recently published using conventional
[10] and endoscopic US elastography [11]. This article
is the first part of a review of SWE applied to several or-
gans. It describes how to optimize the examination tech-
nique, discussing normal values, pitfalls and artefacts for
the examination of the liver, breast. thyroid and salivary
glands. The manuscript provides more specific tips for
applying SWE as part of a diagnostic US examination.
96 Giovanna Ferraioli et al
Shear wave based elastography – how does it work?
SWE techniques include vibration controlled tran-
sient elastography (VCTE) and ARFI based techniques.
The shear waves are generated by a body-surface vibra-
tion, as in VCTE, or by the push-pulse of a focused US
beam, as in ARFI techniques. In VCTE, a body-surface
vibration creates a shear wave, which then travels to the
organ of interest. The frequency of the vibration is con-
trolled (50 Hz), as are its shape and amplitude. VCTE
is implemented on the Fibroscan®, which is a dedicated
device that does not display an anatomical image. In
ARFI-based techniques, the shear waves are generated
directly in the tissue. A convex or linear transducer trans-
mits focused US pulses (also known as a push pulses or
ARFI) that generate shear waves. The pulses are repeated
several times over a short period of time, and the shear
waves generated travel at a much slower rate than US.
B-mode tracking pulses are used to detect the propaga-
tion velocity of the shear wave [12] by measuring the
difference in arrival time (time lag) between two points
at known distances apart from each other [1,6,13]. Such
push pulses generate much slower shear waves off-axis
[14]. ARFI-based techniques include point shear-wave
elastography (pSWE) and multidimensional SWE (2D-
SWE, 3D-SWE). pSWE measures the stiffness at the
focal (~1cm3) point in the tissue whereas with 2D-SWE
the stiffness is measured over a much larger area and a
color-coded image of the qualitative elastic properties is
displayed on the monitor of the US system [3,4].
Shear wave speed
The shear wave speed (SWS) is almost one thousand
times lower than the velocity of US in soft tissues, the
shear waves attenuate very rapidly and some do not prop-
agate in the simple fluids [14]. The shear wave propagates
faster in stiffer tissue than in softer tissue. The expected
SWS in the liver in normal and pathologic states is typi-
cally in the range 0.7 to 5.0 m/s (1.5 kPa to 75 kPa). For
breast cancers it can be up to 10 m/s (300 kPa) and even
higher for normal tendons. Pathology in any tissue often
creates changes in tissue stiffness making elastography a
method to characterize pathological changes. SWS val-
ues may vary depending on the vendor; therefore, vendor
specific cut-off values may be necessary.
Differences of equipment
Different equipment may give different values of
stiffness within the same tissue in the same patient. This
is because the measured values of SWS will vary with
a number of system factors, in particular shear-wave
vibration mean frequency and bandwidth. In addition,
measurement bias may occur due to the algorithm em-
ployed to calculate relative shear wave arrival time and
speed.
How to perform shear wave elastography. Part I
kPa or can they change to m/s?
The propagating speed [12] of the generated shear
wave is reported in meters per second (m/s) but can also
be converted to Young’s modulus values in kilopascals
(kPa) by applying the formula E=3ρVs2, where E is tis-
sue elasticity, Vs is the shear wave speed, and ρ is the
density of tissue in kg/m3, and making some assumptions
[1-3,6]. One main reason why it is preferable to report
results in units of ms-1 rather than kPa is the fact that the
SWS is measured by the scanner in ms-1. However, main-
ly for liver application, the units of the Young’s modu-
lus are largely used, as many clinicians are familiar with
them.
Angle of insonation
The angle of insonation has a significant influence on
the measurement, which is of importance when a curved
transducer is used [15]. The shear waves are generated
perpendicular to the ARFI push pulse therefore the B-
mode tracking must be in the same angle to accurately
estimate the SWS.
Region of interest
The ROI should be positioned so that the push pulse
is generated perpendicular to the center of the transducer
surface. For more information about the technology we
also refer to the recently published guidelines on elastog-
raphy [1,7,15,16].
Does the size and/or the shape of the ROI influences
measurements?
The size of the ROI depends on the tissue being
evaluated. Even though a larger ROI would give SWE
information over a larger amount of tissue, it risks the in-
clusion of artifacts particularly in heterogeneous masses.
By using two different 2D-SWE US systems, it has been
shown that, for the assessment of breast lesions, a small
round ROI (approximately 2 mm in diameter) placed
over the stiffest area of the lesion was more accurate than
a larger ROI manually drawn along the margin of the le-
sion [17]. In another study that assessed the influence on
the accuracy of 2D-SWE in evaluating breast lesions by
using three different ROI size (1, 2 and 3 mm), the diag-
nostic accuracy was not affected by changing the ROI
size [18]. In general, malignant lesions are heterogene-
ous in stiffness and using the area of highest stiffness is
more accurate in characterizing the lesion. However, for
homogenous tissue like liver, a larger ROI can average
the stiffness over a larger area of tissue.
In ex vivo study involving porcine muscle, a sig-
nificant increase of SWS (p<0.001) was observed for
larger ROI widths. In this animal model, the SWS was
also influenced by several other factors, including probe
frequency, applied pressure, muscle orientation, different
machine settings, and placement depth [19].

Artefacts
SWE images are reconstructed using time-of-flight
based images. In heterogeneous tissues these algorithms
might introduce a variety of artifacts. One of them is SWE
under- and overestimation from reflections at stiffness in-
terfaces. Reflected waves violate the assumption of a sin-
gle direction of propagation, leading to artifacts in SWE
images [20]. To avoid this, directional filters had been
applied [20,21]. By separating the forward and backward
components, it is possible to almost entirely remove the
reflected wave [21]. It is highly recommended in transient
shear wave applications to avoid reflection artifacts. For
liver assessment, common artifacts include reverberation
from the liver capsule, respiratory/cardiac motion and
vessel pulsation/loss of the SWE signal (fig 1). The pen-
etration of the US beam can also generate artifacts since
consistent elasticity estimates cannot be obtained in the
far field due to attenuation of the ARFI pulse. The most
consistent estimates are generally obtained near the focus
zone of the ARFI pulse, where the largest displacements
is generated by the push pulses [22]. A detailed analysis
of all the artifacts is out of the scope of this review article
and can be found elsewhere [22-24].
Fig 1. Example of the reverberation artifact from the liver
capsule in SWE. The red and teal areas are the artifacts; the
blue areas are the accurate stiffness measurements. Note that
in p-SWE a color map is not provided, so it is critical to place
the ROI box 1.5-2 cm below the liver capsule. Whereas in 2D-
SWE the artifact can be identified on the color map and be
avoided.
Med Ultrason 2022; 24(1): 95-106 97
Liver
Main objective, clinical value
The liver is an important target organ for the use of
elastography; stiffness correlates with the degree of fi-
brosis and indirectly with portal hypertension and the
risk of developing hepatocellular carcinoma. Due to the
large overlap between stiffness values, guidelines do not
recommend the use of SWE to differentiate benign and
malignant focal liver lesions [15,25-28].
The most important clinical management may be
summarized as follows:
1. SWE values within the normal range can rule out
compensated advanced chronic liver disease (cA-
CLD) when in agreement with the clinical and labo-
ratory data.
2. SWE technologies perform best to rule out cirrhosis.
3. SWE technologies can be used as first line assessment
for the severity of liver fibrosis but are much less reli-
able in differentiating intermediate stages of fibrosis.
4. An interquartile range/median (IQR/M) ≤30% with
measurements taken in kPa or <15% when taken in
m/s is the most important reliability criterion [29].
“Knobology”
Prerequisites
The user should always refer to the manufacturer’s
recommendations for a good quality measurement. The
parameters that should be taken into account vary from
one manufacturer to another and include judgment of the
signal-to-noise ratio or the stability of the signal over
time (2D-SWE acquisitions). Several manufacturers
have developed quality criteria for either pSWE or 2D-
SWE techniques. The users must always refer to them
when they are available.
Transient elastography: probe selection
In transient elastography (TE) three different probes
are available (S, M and XL probes). The S probe is used
in children with a thoracic belt <75 cm whereas the XL
probe is dedicated to overweight/obese subjects with
more reliable results as compared with the M probe. The
XL probe must be used when the skin-to-liver capsule
distance is higher than 25 mm. Limiting factors for the
XL probe are a skin-to-liver capsule distance >3.4 cm
and extreme obesity (BMI >40 kg/m²) [3,4,28]. Values
obtained with XL probe are usually lower than with the
M probe, therefore no recommendation on the cut-offs to
be used can be given.
ARFI-based techniques: transducer (frequency) selection
In adults, the convex transducer is used for perform-
ing the elastography studies, whereas in children the
choice of the probe, either the linear or the curvilinear
one, depends on the body habitus and age. Generally, the
98 Giovanna Ferraioli et al
same rule used for the choice between the two probes
for the B-mode image of the liver applies also to the as-
sessment of liver stiffness (LS) in children. However, it
should be kept in mind that the difference in frequency
between the two probes gives different readings in the
same subject. In phantom studies, it has been shown that
the readings with the higher frequency of the linear trans-
ducer are higher than those obtained with the convex
transducer. Moreover, in children the acquisition could
be more challenging due to the lack of cooperation and
this could affect the feasibility of the technique [30,31].
Description of (other) parameters
The strength of the push-pulse is higher in the center
of the transducer, thus the sampling should be done in
the central area of the image, whereas the sampling at the
edge should be avoided.
The influence of depth on the estimation of the elas-
tic properties is not negligible [32]. The acoustic push
pulse is progressively attenuated as it traverses the tissue.
The results with the lowest variability are obtained at a
depth of 4-5 cm from the skin surface [33]. The attenua-
tion is higher in stiffer liver, thus in cirrhotic or steatotic
patients, measurements are more variable [15]. The ROI
box should be perpendicular to the transducer.
Region of interest (ROI) size, shape, others
The region of interest should be in between 2-6 cm
below the liver capsule.
In pSWE the size of the region of interest (ROI) is
small and fixed by the manufacturer because the tech-
nique assesses the stiffness at a single location by using a
sequence of push-pulses, generally up to five.
In 2D-SWE the size of the ROI is user-adjustable and
can theoretically be as large as the ARFI FOV image.
However, the larger the ROI the higher the risk of includ-
ing artifacts. Thus, generally the ROI’s size in SWE tech-
nique may influence the quality of the elastogram. Fol-
lowing EFSUMB guidelines and recommendations, we
suggest using an ROI of 2.5x2.5 cm in size [3,4]. Many
vendors have quality or confidence maps, which help to
identify and avoid artifacts [15].
Position of the transducer
The measurements should be performed through
the intercostal space rather than the subcostal approach
yielding the highest intra- and interobserver agreement
[15,34-36].
Description of quality parameters
The most important criterion for a measurement of
good quality seems an IQR/M ≤30% when the results are
reported in Young’s modulus [29]. This ratio, in fact, is a
measure of the variability between consecutive acquisi-
tions, and studies have reported a decrease in accuracy
when this criterion is not fulfilled [37-41]. For measure-
How to perform shear wave elastography. Part I
ment reported in m/s the IQR/M should be ≤15% because
the conversion between the two is not linear [29].
Pre-compression
Pre-compression should be avoided.
How many measurements?
Based on literature data, for the pSWE technique the
EFSUMB and WFUMB guidelines have recommended
to use the median value of 10 acquisitions [3,4,9]. How-
ever, some studies have shown that the accuracy does not
decrease when fewer acquisitions (up to five) are obtained
[38-40,42]. For 2D-SWE, the EFSUMB updated guide-
lines have recommended to obtain at least three acquisi-
tions [3,4]. The updated WFUMB and SRU guidelines
are more cautious and have suggested five acquisitions
when a quality factor is available [9,43]. The higher num-
ber of acquisitions suggested by the WFUMB updated
guidelines may give a better estimation of the variabil-
ity assessed through the calculation of the IQR/M ratio.
Reproducibility
The intra-observer reproducibility of VCTE [44-46],
pSWE [34,36,47-49] and 2D-SWE [50-52] for LS as-
sessment is excellent with ICC above 0.90.
How to use shear wave elastography
The transducer should be positioned in an intercostal
space; perpendicular to the liver in both superior/inferior
and right/left planes, avoiding the ribs or the lung arti-
facts. As the SWS is calculated based on B-mode, the
quality of the B-mode US image affects the quality of
the SWE acquisitions. The most common limitations en-
countered with US, i.e. poor acoustic window, limited
penetration, and rib or lung shadowing, may influence
both the feasibility and the performance of the SWE
techniques. Some of these limitations can be avoided,
thus the operator should obtain an optimal scan of the
liver before launching the acquisition. The perpendicular
position of the transducer can be assessed by looking at
the liver capsule that appears as a sharp white line, paral-
lel to the transducer’s line (fig 2). Motion of the probe
or of the patient affects the quality of the measurement
as well. The patient should breathe normally while the
operator is searching for the best acoustic window and
for the best area of liver parenchyma where the sample
box will be positioned. This area should be homogene-
ous, i.e., free of vessels or ligaments. Before launching
the acquisition, the operator asks the patient to hold the
breath in a neutral position without performing a Vals-
alva’s maneuver for the few seconds needed for the ac-
quisition [15]. Special applications in pediatric patients
are discussed elsewhere [30,31,53].
Tips and tricks
Depth as assessed by the skin-to-liver capsule dis-
tance may influence the SWS values assessed by all

Fig 2. Figure demonstrating the positioning of the liver capsule
and FOV box in liver stiffness assessment. The transducer, liver
capsule and top of the FOV box should be parallel lines. The
liver capsule should be a sharp echogenic line.
SWE-techniques. Due to the attenuation of the US beam,
the depth for reliable measurements is up to 7 cm in most
systems; measurements performed deeper have a lower
signal/noise ratio. Using a deep abdominal probe may
allow for measurements at a greater depth in high BMI
patients. In ARFI techniques, the US beam that generates
the shear waves is also attenuated by the interaction with
the tissue that it traverses thus, its strength is inversely
related to the depth; this attenuation is higher in cases
of liver steatosis or severe fibrosis. Due to these factors,
measurements in patients with significant liver steatosis
or severe fibrosis could have a higher rate of unreliable
results or failures. This is also true in obese patients with
thick subcutaneous tissue due to higher attenuation of
the US beam in the near field. In staging liver fibrosis,
Metavir-derived cutoff values are system-specific and
could not be applied interchangeably across different
US systems. A recent study has shown that the agree-
ment between LS measurements obtained with different
US systems is good to excellent; however, the difference
between values was higher than two kPa, assigning the
patient to different stages of liver fibrosis [54]. Because
the overlap of LS values between METAVIR-derived
scores is as large if not larger than the difference between
vendors, the updated SRU consensus advises that sepa-
rate cut-off values for each vendor are not required when
determining the likelihood for cACLD [29]. The SWE
values might be overestimated in certain diseases, e.g.
sinusoidal obstruction syndrome due to congestion [55].
Med Ultrason 2022; 24(1): 95-106 99
Normal reference values
For all equipment, a SWE measurement within the
normal range, in a subject without other clinical or labo-
ratory evidence of liver disease, may exclude significant
liver fibrosis with a high degree of certainty. For both
VCTE and ARFI-based techniques, there is consensus in
considering that values ≤5 kPa (1.3 m/s) have high prob-
ability of being normal [29,56].
What to avoid?
Confounding factors that may lead to an increase of
LS independently from liver fibrosis have been listed
elsewhere [57-59]. Briefly, eating may increase the stiff-
ness of the liver, thus measurements are performed in the
fasting status of at least 4 hours. LS does not necessarily
reflect liver fibrosis, but can reflect many other physi-
ological or pathological conditions including hepatic
inflammation (elevated transaminase level) [60-63], ob-
structive cholestasis [64], neoplastic and other infiltration
of the liver and hepatic congestion [65,66]. Recently, it
has been reported that portal vein thrombosis is also a
confounder [67]. On the other hand, SWE can play a role
in cases of liver congestion due to right-sided heart fail-
ure, congenital heart diseases or valvular diseases as well
as in the hepatic sinusoidal obstruction syndrome or in
the Budd-Chiari syndrome [31,68].
Specific artifacts
Measurements should be performed at least 1-2 cm
below the liver capsule to avoid reverberation artifacts.
However, when using 2D-SWE with a quality map the
measurement can be taken closer to the liver capsule as
the artifact can be visualized and avoided. This is helpful
in high BMI or steatotic patients since the reverberation
artifact in these patients can be as small as 5mm and visu-
alizing the artifact on 2D-SWE may help with placing the
ROI closer to the liver surface and still avoid the rever-
beration artifact.
Breast
Main objective, clinical value
Various studies have shown that malignant and be-
nign breast tumours differ significantly in their elastic-
ity [7,69-75]. Benign alterations tend to be softer than
malignant lesions. This fact forms the basis for the use
of elastography to differentiate between different breast
tumors. SWE is a new method introduced in 2009 and,
unlike strain elastography, allows quantitative measure-
ment of tissue stiffness. SWE is not only capable of the
differentiation between benign and malignant tumours,
but can also be used for therapy monitoring under neoad-
juvant chemotherapy [76-79]. Recently, the fifth edition
of the ACR BI-RADS Atlas 2013 incorporated elasticity
100 Giovanna Ferraioli et al
assessment of breast lesions as one of the associated fea-
tures of ultrasound [80].
“Knobology”
Transducer (frequency) selection
A standard 5 cm wide linear transducer is very well
suited to perform SWE of breast lesions. Depending on
vendor, transducers of 9 MHz to 18 MHz are SWE ena-
bled. Before the elastography mode is activated, a high-
quality B-mode image must first be set, because the elas-
togram is derived from it. It is recommended to use higher
US frequencies in the assessment of superficial breast le-
sions and lower frequencies for better depth penetration
for lesions located deep inside the breast. The operator
must be aware that lower frequencies result in a lower
spatial resolution. The US probe must be placed perpen-
dicular on the skin of breast directly above the lesion with
enough contact to the breast tissue to obtain a good B-
mode image while avoiding excessive pre-compression.
Pre-compression can be recognized when fatty tissue that
should normally appear blue (soft) on the color map has
a different color (fig 3). The examiner must not move
the US probe while the elastogram is being obtained.
Region of interest (ROI) size
There are several approaches in setting the ROI size.
One way is to use a small ROI placed at the site of the
stiffest area within the mass or within 3 cm surrounding
the mass. Another way is to use larger ROIs that cover
the entire lesion. No general standard is given within
published guidelines. A recent paper evaluating 154
breast lesions came to the conclusion that a small ROI
measuring the mean or maximum stiffness value is supe-
rior to medium sized or large sized ROI in distinguishing
between benign and malignant lesions [17]. Regardless
of the size of the ROI, it has been shown that minimum
stiffness value is the least significant and should there-
fore not be used.
Description of quality parameters
Some vendors provide a quality map, which is a
color-coded map that can be superimposed on the B-
mode US image and provides information of the qual-
ity of the shear wave propagation and the image quality.
Even green distributions indicate a high quality elasto-
gram, whereas yellow or red areas should not be used for
assessment [81].
Pre-compression
Pre-compression is an important factor of influence
changing the appearance of a lesion in the elastogram.
If too much pre-compression were applied, the lesion
would appear stiffer than it really is [82]. Therefore, the
recommendation is to apply a large amount of gel and
then place the US transducer on the breast. The subcuta-
neous fatty tissue should appear in dark blue. If it appears
How to perform shear wave elastography. Part I
Fig 3. 2D-SWE of an invasive ductal cancer. The irregular hy-
poechoic lesion is the malignancy. Note that the skin surface
is parallel to the transducer, there is a small amount of cou-
pling gel between the transducer and skin confirming minimal
precompression and that the area of highest stiffness is just ad-
jacent to the lesion.
in green or even red than too much pre-compression is
used and needs to be corrected.
Checking reproducibility
To assess the quality and reproducibility of the elas-
tography image, place the probe and hold it still while the
elastogram builds superimposed on the B-mode image.
Wait 5-10 seconds until the elastograms shows a consist-
ent and permanent color pattern. Also check that the fatty
tissue is displayed soft, which indicates that not too much
pre-compression is applied. Then freeze the picture and
proceed with the measurements within the elastogram.
How to use shear wave elastography
The basic recommendations about performing SWE
as described above also apply to the use in breast tu-
mours. SWE should be used as a lesion-based adjunct to
conventional B-mode imaging using all ACR BI-RADS
criteria giving the examiner more information in order
to make a final assessment. It should not be used as a
screening tool without a lesion. However, if a palpable
lesion is present with no B-mode findings elastography
may identify an isoechoic lesion. A quantitative assess-
ment using the mean or maximum stiffness values can
be used. Alternatively, the color pattern of the elastogram
can be analyzed by using different color pattern scores
[83,84].
Artifacts
There is a well-documented artifact with SWE in
malignant lesions. They may appear as soft lesions even
though they are very stiff. This artifact is not infrequent
and can be recognized by evaluating the quality map. In
these false negative cases the quality map usually will
confirm that there are poor shear wave and the results
should not be used. These cases of false negative lesions
on SWE are always true positives on SE. Therefore, the

combination of SE and SWE will improve diagnostic ac-
curacy [71,72,85].
Tips and tricks
• Use ultrasound frequency according to lesion lo-
calization (depth);
• Use small ROI and mean or maximum stiffness
values for assessment;
• Change the clinical procedure for BI-RADS 4a
and BI-RADS 3 lesions according to SWE meas-
urements;
• If the stiff rim sign occurs measure within the stiff
rim and not inside the uncolored lesions center.
Normal reference values
Normal fatty tissue: mean stiffness values 5-10 kPa
(1.3-1.8 m/s); breast parenchyma: mean stiffness values
30-50 kPa (3.1-4.1 m/s) [86].
Differentiating between benign and malignant lesions
the following cut-off values are reported: maximum stiff-
ness values 33.3-80 kPa, (3.3-5.0 m/s); mean stiffness
values 46.7-93.8 kPa, (4.0-5.6 m/s).
What to avoid?
During examination it should be avoided the pre-
compression by checking the stiffness of fatty tissue
which should be in the normal range, the use of minimum
stiffness values for the differentiation between benign
and malignant breast lesions and measurements in areas
of poor quality on the quality map.
Thyroid
Main objective, clinical value
Despite fine needle aspiration (FNA) being the gold
standard in the diagnosis of thyroid neoplasms, US has a
paramount role in the diagnostic process. Thyroid nod-
ules are indeed present in almost 50% of the population
and so performing cytology on one or multiple targets
on each patient is not feasible; the detection of certain
suspect features on B-mode US is then fundamental in
deciding which nodules should be assessed with FNA
[87,88]. SWE is a quick, readily available tool, and ef-
fective in increasing US sensitivity in the detection of
thyroid neoplasms [89].
“Knobology”
Prerequisites
It is recommended to verify the manufacturer’s instruc-
tions about quality parameters. No patient preparation is
required; patient has to lie down in supine position with
a pillow or a towel used to extend the patient’s neck [2].
Select an appropriate transducer and frequency se-
lection
Select a high-end linear transducer, usually 7 to 18
MHz [90].
Med Ultrason 2022; 24(1): 95-106 101
Fig 4. SWE examination of a thyroid with several nodules. The
ROI is placed in each nodule to avoid including normal thyroid
or other structures. In this case the more central nodule with a
mean stiffness value of 183 kPa was a papillary carcinoma. The
other two nodules were benign.
Region of interest (ROI) size, shape, others
The sample box features depend on the SWE method
being used. In pSWE the small fixed-size ROI should be
completely included within the nodule. In 2D-SWE the
box should be large enough to include the whole nodule,
avoiding nearby vessels or gland areas with cystic or fi-
brotic changes (fig 4) [5,91].
Description of quality parameters
For pSWE 3 to 10 measurements should be acquired
at the same location and the average of these should be
calculated. For 2D-SWE at least three measurements
must be performed [5,91].
Pre-compression
An abundant quantity of gel should be used to avoid
pre-compression, since it may alter tissue elastic modulus
thus causing artifacts. The operator places the transduc-
er perpendicular to the target nodule without pressure,
maintaining only slight contact with the skin. A manufac-
turer quality control tool should be used if available [92].
Normal reference values
Normal values may vary depending on the manufac-
turer; however, guidelines suggest that benign nodules
show a mean elasticity of 15.3-28 kPa [5]. Recent studies
state that the optimal cut-off between benign and malig-
nant nodules is 34.5-37.5 kPa [93,94].
What to avoid?
As stated before, it is important to avoid compression
artifacts, which may jeopardize measurement’s accuracy
and reliability. Certain neck morphologies may be chal-
lenging when correctly placing the probe perpendicular
to the target nodule; previous neck surgery and subse-
quent fibrosis may as well represent an obstacle [2].
102 Giovanna Ferraioli et al
Salivary glands
Main objective, clinical value
Salivary glands are readily accessible to high reso-
lution US, which is the initial imaging modality when
clinically indicated. SWE is useful for the assessment of
diffuse diseases, such as Sjögren syndrome, parotitis in
pediatric patients or damage due to irradiation [95-97]. As
for the evaluation of focal lesions, a substantial overlap
of stiffness values has been reported [94]. A multipara-
metric approach to allow a better differentiation between
benign and malignant lesions has been suggested [90].
“Knobology”
Prerequisites
Before starting shear wave measurement, it is recom-
mended to verify the quality of the shear wave generation
by referring to the manufacturer’s quality parameters.
Transducer (frequency) selection
SWE is performed with a high-frequency linear trans-
ducer, typically 7 to 12 MHz, with patient lying in the
supine position with a pillow or a towel used to extend
the patient’s neck [90,98].
Region of interest (ROI) size, shape, others
The sample box should be positioned in a region of
the gland free of vessels or cystic or fibrotic transforma-
tion, and between 1 to 2 cm from the anterior glandular
contour.
Description of quality parameters
For pSWE, the fixed ROI should be placed at the
point of interest. Three to 10 measurements should be
acquired and the median value calculated. For 2D-SWE,
because a larger FOV is available, at least three measure-
ments must be performed [5].
Pre-compression
Pre-compression should be avoided.
Tips and tricks
For obtaining a prompt and reliable SWE acquisition,
it is recommended to hold the transducer perpendicular
to the plane being explored, avoiding any movement of
the transducer or of the patient when the acquisition has
been launched. A sufficient quantity of gel has to be used
and minimal pressure should be applied to avoid pre-
compression because it can alter tissue elastic modules
and produce artifacts. False negative cases do occur and
if the B-mode findings are suggestive of a malignancy a
biopsy should not be cancelled based on elastography.
When there are several similar nodules, SWE can be used
to select the stiffest lesion for sampling.
Normal reference values
Literature data suggest that the mean stiffness of the
salivary glands is quite uniform (approximately 11±3
kPa) (fig 5) [99].
How to perform shear wave elastography. Part I
Fig 5. 2D-SWE of a normal parotid gland.
What to avoid
As explained for other organs, artifacts should be lim-
ited for a prompt and reliable elastography acquisition.
The interpretation of a 2D-SWE elastogram is operator
dependent and the choice of an adequate ROI is challeng-
ing due to the multiplicity and complexity of the struc-
tures in the neck region.
Specific artifacts
The reliability of the measurement may be affected
by some artifacts that may arise in the region of the neck,
generally due to the proximity to the skin, to the osseous
plane (ramus of mandible) or to the eventual presence
of focal convex bulge of the skin, which generates local
inhomogeneity and falsify the real elasticity of the tissue.
Some very stiff cancers show a circular stiff area in the
surroundings of the actual lesion. This artifact is called
the “stiff rim sign”. In 3D SWE there is a similar sign in
the c-plane called the “crater sign” [100]. The reason for
these artifacts is still under discussion but it may indicate
a very low shear wave amplitude within the cancer due to
attenuation of US energy resulting in no colour coded in-
formation in the elastogram [101]. Furthermore, the “stiff
ring sign” and the “crater sign” can be used as predictors
of malignancy [102].
Acknowledgement. The authors thank the Bad Mer-
gentheimer Leberzentrum e.V. for support.
References
1. Bamber J, Cosgrove D, Dietrich CF, et al. EFSUMB guide-
lines and recommendations on the clinical use of ultrasound
elastography. Part 1: Basic principles and technology. Ul-
traschall Med 2013;34:169-184.
2. Cosgrove D, Piscaglia F, Bamber J, et al. EFSUMB guide-
lines and recommendations on the clinical use of ultrasound
elastography. Part 2: Clinical applications. Ultraschall Med
2013;34:238-253.

3. Dietrich CF, Bamber J, Berzigotti A, et al. EFSUMB
Guidelines and Recommendations on the Clinical Use of
Liver Ultrasound Elastography, Update 2017 (Long Ver-
sion). Ultraschall Med 2017;38:e16-e47.
4. Dietrich CF, Bamber J, Berzigotti A, et al. EFSUMB
Guidelines and Recommendations on the Clinical Use of
Liver Ultrasound Elastography, Update 2017 (Short Ver-
sion). Ultraschall Med 2017;38:377-394.
5. Săftoiu A, Gilja OH, Sidhu PS, et al. The EFSUMB Guide-
lines and Recommendations for the Clinical Practice of
Elastography in Non-Hepatic Applications: Update 2018.
Ultraschall Med 2019;40:425-453.
6. Shiina T, Nightingale KR, Palmeri ML, et al. WFUMB
guidelines and recommendations for clinical use of ultra-
sound elastography: Part 1: basic principles and terminol-
ogy. Ultrasound Med Biol 2015;41:1126-1147.
7. Barr RG, Nakashima K, Amy D, et al. WFUMB guidelines
and recommendations for clinical use of ultrasound elastog-
raphy: Part 2: breast. Ultrasound Med Biol 2015;41:1148-
1160.
8. Ferraioli G, Filice C, Castera L, et al. WFUMB guidelines
and recommendations for clinical use of ultrasound elas-
tography: Part 3: liver. Ultrasound Med Biol 2015;41:1161-
1179.
9. Ferraioli G, Wong VW, Castera L, et al. Liver Ultrasound
Elastography: An Update to the World Federation for Ul-
trasound in Medicine and Biology Guidelines and Recom-
mendations. Ultrasound Med Biol 2018;44:2419-2440.
10. Dietrich CF, Barr RG, Farrokh A, et al. Strain Elastography
- How To Do It? Ultrasound Int Open 2017;3:E137-E149.
11. Dietrich CF, Bibby E, Jenssen C, Saftoiu A, Iglesias-Garcia
J, Havre RF. EUS elastography: How to do it? Endosc Ul-
trasound 2018;7:20-28.
12. Dietrich C, Working Group on Sonography D. [Report of
the Working Group on Sonography DGVS]. Ultraschall
Med 2003;24:3-4.
13. Dietrich CF, Dong Y. Shear wave elastography with a new
reliability indicator. J Ultrason 2016;16:281-287.
14. Sarvazyan AP, Rudenko OV, Swanson SD, Fowlkes JB,
Emelianov SY. Shear wave elasticity imaging: a new ultra-
sonic technology of medical diagnostics. Ultrasound Med
Biol 1998;24:1419-1435.
15. Barr RG, Ferraioli G, Palmeri ML, et al. Elastography
Assessment of Liver Fibrosis: Society of Radiologists in
Ultrasound Consensus Conference Statement. Radiology
2015;276:845-861.
16. Barr RG, Cosgrove D, Brock M, et al. WFUMB Guide-
lines and Recommendations on the Clinical Use of Ultra-
sound Elastography: Part 5. Prostate. Ultrasound Med Biol
2017;43:27-48.
17. Moon JH, Hwang JY, Park JS, Koh SH, Park SY. Impact of
region of interest (ROI) size on the diagnostic performance
of shear wave elastography in differentiating solid breast
lesions. Acta Radiol 2018;59:657-663.
18. Skerl K, Vinnicombe S, Giannotti E, Thomson K, Evans
A. Influence of region of interest size and ultrasound lesion
size on the performance of 2D shear wave elastography
Med Ultrason 2022; 24(1): 95-106 103
(SWE) in solid breast masses. Clin Radiol 2015;70:1421-
1427.
19. Rominger MB, Kälin P, Mastalerz M, et al. Influencing Fac-
tors of 2D Shear Wave Elastography of the Muscle - An Ex
Vivo Animal Study. Ultrasound Int Open 2018;4:E54-e60.
20. Lipman SL, Rouze NC, Palmeri ML, Nightingale KR.
Evaluating the Improvement in Shear Wave Speed Image
Quality Using Multidimensional Directional Filters in the
Presence of Reflection Artifacts. IEEE Trans Ultrason Fer-
roelectr Freq Control 2016;63:1049-1063.
21. Deffieux T, Gennisson JL, Bercoff J, Tanter M. On the
effects of reflected waves in transient shear wave elas-
tography. IEEE Trans Ultrason Ferroelectr Freq Control
2011;58:2032-2035.
22. Bruce M, Kolokythas O, Ferraioli G, Filice C, O’Donnell
M. Limitations and artifacts in shear-wave elastography of
the liver. Biomed Eng Lett 2017;7:81-89.
23. Bouchet P, Gennisson JL, Podda A, Alilet M, Carrié
M, Aubry S. Artifacts and Technical Restrictions in 2D
Shear Wave Elastography. Ultraschall Med 2020;41:267-
277.
24. Dubinsky TJ, Shah HU, Erpelding TN, Sannananja B, Son-
neborn R, Zhang M. Propagation Imaging in the Demon-
stration of Common Shear Wave Artifacts. J Ultrasound
Med 2019;38:1611-1616.
25. Yu H, Wilson SR. Differentiation of benign from malignant
liver masses with Acoustic Radiation Force Impulse tech-
nique. Ultrasound Q 2011;27:217-223.
26. Dong Y, Wang WP, Xu Y, Cao J, Mao F, Dietrich CF.
Point shear wave speed measurement in differentiating
benign and malignant focal liver lesions. Med Ultrason
2017;19:259-264.
27. Dietrich CF, Shi L, Wei Q, et al. What does liver elas-
tography measure? Technical aspects and methodology.
Minerva Gastroenterol Dietol 2020;doi: 10.23736/S1121-
421X.20.02787-7.
28. Dong Y, Sirli R, Ferraioli G, et al. Shear wave elastogra-
phy of the liver - review on normal values. Z Gastroenterol
2017;55:153-166.
29. Barr RG, Wilson SR, Rubens D, Garcia-Tsao G, Ferraioli
G. Update to the Society of Radiologists in Ultrasound
Liver Elastography Consensus Statement. Radiology
2020;296:263-274.
30. Dietrich CF, Sirli R, Ferraioli G, et al. Current knowledge
in ultrasound-based liver elastography of pediatric patients.
Appl Sci 2018;8:944-963.
31. Ferraioli G, Barr RG, Dillman JR. Elastography for Pediat-
ric Chronic Liver Disease: A Review and Expert Opinion. J
Ultrasound Med 2020; doi: 10.1002/jum.15482.
32. Hall TJ, Milkowski A, Garra B, et al. RSNA/QIBA: shear
wave speed as a biomarker for liver fibrosis staging. In:
Ultrasonics Symposium (IUS), 2013 I.E. International;
2013;97-400.
33. Chang S, Kim MJ, Kim J, Lee MJ. Variability of shear wave
velocity using different frequencies in acoustic radiation
force impulse (ARFI) elastography: a phantom and normal
liver study. Ultraschall Med 2013;34:260-265.
104 Giovanna Ferraioli et al
34. Ling W, Lu Q, Quan J, Ma L, Luo Y. Assessment of impact
factors on shear wave based liver stiffness measurement.
Eur J Radiol 2013;82:335-341.
35. Ferraioli G, Lissandrin R, Zicchetti M, Bernuzzi S, Salvan-
eschi L, Filice C, Group ES. Ultrasound point shear wave
elastography assessment of liver and spleen stiffness: effect
of training on repeatability of measurements. Eur Radiol
2014;24:1283-1289.
36. D’Onofrio M, Gallotti A, Mucelli RP. Tissue quantification
with acoustic radiation force impulse imaging: Measure-
ment repeatability and normal values in the healthy liver.
AJR Am J Roentgenol 2010;195:132-136.
37. Bota S, Sporea I, Sirli R, Popescu A, Jurchis A. Factors
which influence the accuracy of acoustic radiation force
impulse (ARFI) elastography for the diagnosis of liver fi-
brosis in patients with chronic hepatitis C. Ultrasound Med
Biol 2013;39:407-412.
38. Ferraioli G, De Silvestri A, Reiberger T, et al. Adherence
to quality criteria improves concordance between tran-
sient elastography and ElastPQ for liver stiffness assess-
ment-A multicenter retrospective study. Dig Liver Dis
2018;50:1056-1061.
39. Ferraioli G, Maiocchi L, Lissandrin R, Tinelli C, De Sil-
vestri A, Filice C, Liver Fibrosis Study G. Accuracy of the
ElastPQ Technique for the Assessment of Liver Fibrosis
in Patients with Chronic Hepatitis C: a “Real Life” Single
Center Study. J Gastrointestin Liver Dis 2016;25:331-335.
40. Roccarina D, Iogna Prat L, Buzzetti E, et al. Establishing
Reliability Criteria for Liver ElastPQ Shear Wave Elastog-
raphy (ElastPQ-SWE): Comparison Between 10, 5 and 3
Measurements. Ultraschall Med 2019;doi: 10.1055/a-1010-
6052.
41. Fang C, Jaffer OS, Yusuf GT, et al. Reducing the Number
of Measurements in Liver Point Shear-Wave Elastography:
Factors that Influence the Number and Reliability of Meas-
urements in Assessment of Liver Fibrosis in Clinical Prac-
tice. Radiology 2018:172104.
42. Fang C, Jaffer OS, Yusuf GT, Konstantatou E, Quinlan
DJ, Agarwal K, Quaglia A, et al. Reducing the Number of
Measurements in Liver Point Shear-Wave Elastography:
Factors that Influence the Number and Reliability of Meas-
urements in Assessment of Liver Fibrosis in Clinical Prac-
tice. Radiology 2018;287:844-852.
43. Barr RG, Wilson SR, Rubens D, Garcia-Tsao G, Ferraioli
G. Update to the Society of Radiologists in Ultrasound
Liver Elastography Consensus Statement. Radiology
2020:192437.
44. Fraquelli M, Rigamonti C, Casazza G, et al. Reproducibil-
ity of transient elastography in the evaluation of liver fibro-
sis in patients with chronic liver disease. Gut 2007;56:968-
973.
45. Boursier J, Konate A, Gorea G, et al. Reproducibility of
liver stiffness measurement by ultrasonographic elastom-
etry. Clin Gastroenterol Hepatol 2008;6:1263-1269.
46. Afdhal NH, Bacon BR, Patel K, et al. Accuracy of fibros-
can, compared with histology, in analysis of liver fibrosis in
patients with hepatitis B or C: a United States multicenter
How to perform shear wave elastography. Part I
study. Clin Gastroenterol Hepatol 2015;13:772-779 e771-
773.
47. Guzmán-Aroca F, Reus M, Berná-Serna JD, et al. Repro-
ducibility of shear wave velocity measurements by acous-
tic radiation force impulse imaging of the liver: a study in
healthy volunteers. J Ultrasound Med 2011;30:975-979.
48. Friedrich-Rust M, Wunder K, Kriener S, et al. Liver fibro-
sis in viral hepatitis: noninvasive assessment with acoustic
radiation force impulse imaging versus transient elastogra-
phy. Radiology 2009;252:595-604.
49. Bota S, Sporea I, Sirli R, Popescu A, Danila M, Costaches-
cu D. Intra- and interoperator reproducibility of acoustic
radiation force impulse (ARFI) elastography--preliminary
results. Ultrasound Med Biol 2012;38:1103-1108.
50. Yoon JH, Lee JM, Han JK, Choi BI. Shear wave elastogra-
phy for liver stiffness measurement in clinical sonographic
examinations: evaluation of intraobserver reproducibility,
technical failure, and unreliable stiffness measurements. J
Ultrasound Med 2014;33:437-447.
51. Thiele M, Detlefsen S, Sevelsted Moller L, et al. Transient
and 2-Dimensional Shear-Wave Elastography Provide
Comparable Assessment of Alcoholic Liver Fibrosis and
Cirrhosis. Gastroenterology 2016;150:123-133.
52. Cassinotto C, Boursier J, De Ledinghen V, et al. Liver stiff-
ness in nonalcoholic fatty liver disease: A comparison of
Supersonic Shear Imaging, FibroScan and ARFI with liver
biopsy. Hepatology 2016;63:1817-1827.
53. Dietrich CF, Ferraioli G, Sirli R, et al. General advice in
ultrasound based elastography of pediatric patients. Med
Ultrason 2019;21:315-326.
54. Ferraioli G, De Silvestri A, Lissandrin R, et al. Evaluation
of Inter-System Variability in Liver Stiffness Measure-
ments. Ultraschall Med 2019;40:64-75.
55. Dietrich CF, Trenker C, Fontanilla T, et al. New Ultrasound
Techniques Challenge the Diagnosis of Sinusoidal Obstruc-
tion Syndrome. Ultrasound Med Biol 2018;44:2171-2182.
56. de Franchis R, Baveno VIF. Expanding consensus in portal
hypertension: Report of the Baveno VI Consensus Work-
shop: Stratifying risk and individualizing care for portal
hypertension. J Hepatol 2015;63:743-752.
57. Bazerbachi F, Haffar S, Wang Z, et al. Range of Normal
Liver Stiffness and Factors Associated With  Increased
Stiffness Measurements in Apparently Healthy Individuals.
Clin Gastroenterol Hepatol 2019;17:54-64.e51.
58. Hodge A, Lim S, Goh E, et al. Coffee Intake Is Associated
with a Lower Liver Stiffness in Patients with Non-Alcohol-
ic Fatty Liver Disease, Hepatitis C, and Hepatitis B. Nutri-
ents 2017;9:56.
59. Langdon JH, Elegbe E, Gonzalez RS, Osapoetra L, Ford
T, McAleavey SA. Measurement of Liver Stiffness Us-
ing Shear Wave Elastography in a Rat Model: Factors
Impacting Stiffness Measurement with Multiple- and Sin-
gle-Tracking-Location Techniques. Ultrasound Med Biol
2017;43:2629-2639.
60. Coco B, Oliveri F, Maina AM, et al. Transient elastography:
a new surrogate marker of liver fibrosis influenced by major
changes of transaminases. J Viral Hepat 2007;14:360-369.

61. Sagir A, Erhardt A, Schmitt M, Haussinger D. Transient elas-
tography is unreliable for detection of cirrhosis in patients
with acute liver damage. Hepatology 2008;47:592-595.
62. Arena U, Vizzutti F, Corti G, et al. Acute viral hepatitis in-
creases liver stiffness values measured by transient elastog-
raphy. Hepatology 2008;47:380-384.
63. Vigano M, Massironi S, Lampertico P, et al. Transient elas-
tography assessment of the liver stiffness dynamics during
acute hepatitis B. Eur J Gastroenterol Hepatol 2010;22:180-
184.
64. Millonig G, Reimann FM, Friedrich S, et al. Extrahepatic
cholestasis increases liver stiffness (FibroScan) irrespective
of fibrosis. Hepatology 2008;48:1718-1723.
65. Millonig G, Friedrich S, Adolf S, Fet al. Liver stiffness is
directly influenced by central venous pressure. J Hepatol
2010;52:206-210.
66. Colli A, Pozzoni P, Berzuini A, et al. Decompensated chron-
ic heart failure: increased liver stiffness measured by means
of transient elastography. Radiology 2010;257:872-878.
67. Huang R, Gao ZH, Tang A, Sebastiani G, Deschenes M.
Transient elastography is an unreliable marker of liver fi-
brosis in patients with portal vein thrombosis. Hepatology
2018;68:783-785.
68. Ferraioli G, Barr RG. Ultrasound liver elastography be-
yond liver fibrosis assessment. World J Gastroenterol
2020;26:3413-3420.
69. Krouskop TA, Wheeler TM, Kallel F, Garra BS, Hall T.
Elastic moduli of breast and prostate tissues under com-
pression. Ultrason Imaging 1998;20:260-274.
70. Cantisani V, David E, Barr RG, et al. US-Elastography for
Breast Lesion Characterization: Prospective Comparison of
US BIRADS, Strain Elastography and Shear wave Elastog-
raphy. Ultraschall Med 2020; doi: 10.1055/a-1134-4937.
71. Barr RG. Sonographic breast elastography: a primer. J Ul-
trasound Med 2012;31:773-783.
72. Barr RG. Comparison of Strain Elastography, Shearwave
Elastography, and Shearwave Elastography with a Qual-
ity MEasure in Evaluation of Breast Masses. In: Texas Uo,
editor. Eleventh International Tissue Elasticity Conference.
Deauville, France: University of Texas; 2012:3-4.
73. Barr RG, De Silvestri A, Scotti V, et al. Diagnostic Per-
formance and Accuracy of the 3 Interpreting Methods of
Breast Strain Elastography: A Systematic Review and Me-
ta-analysis. J Ultrasound Med 2019;38:1397-1404.
74. Barr RG, Zhang Z. Shear Wave Elastography of the Breast:
Value of a Quality Measure and Comparison to Strain Elas-
tography. Radiology 2015;275:45-53
75. Barr RG, Destounis S, Lackey LB, 2nd, Svensson WE,
Balleyguier C, Smith C. Evaluation of breast lesions using
sonographic elasticity imaging: a multicenter trial. J Ultra-
sound Med 2012;31:281-287.
76. Berg WA, Cosgrove DO, Dore CJ, et al. Shear-wave elastog-
raphy improves the specificity of breast US: the BE1 multi-
national study of 939 masses. Radiology 2012;262:435-449.
77. Cosgrove DO, Berg WA, Dore CJ, et al. Shear wave elas-
tography for breast masses is highly reproducible. Eur Ra-
diol 2012;22:1023-1032.
Med Ultrason 2022; 24(1): 95-106 105
78. Schafer FK, Hooley RJ, Ohlinger R, et al. ShearWave Elas-
tography BE1 multinational breast study: additional SWE
features support potential to downgrade BI-RADS(R)-3 le-
sions. Ultraschall Med 2013;34:254-259.
79. Maier AM, Heil J, Harcos A, et al. Prediction of pathologi-
cal complete response in breast cancer patients during neo-
adjuvant chemotherapy: Is shear wave elastography a use-
ful tool in clinical routine? Eur J Radiol 2020;128:109025.
80. Mendelson EB B-VM, Berg WA. ACR BI-RADS® Ultra-
sound in: ACR BI-RADS® Atlas, Breast Imaging Report-
ing and Data System. In: Reston, VA, American College of
Radiology; 2013.
81. Zheng X, Huang Y, Liu Y, et al. Shear-Wave Elastography
of the Breast: Added Value of a Quality Map in Diagnosis
and Prediction of the Biological Characteristics of Breast
Cancer. Korean J Radiol 2020;21:172-180.
82. Wojcinski S, Brandhorst K, Sadigh G, Hillemanns P, De-
genhardt F. Acoustic radiation force impulse imaging with
virtual touch tissue quantification: measurements of normal
breast tissue and dependence on the degree of pre-compres-
sion. Ultrasound Med Biol 2013;39:2226-2232.
83. Tozaki M, Fukuma E. Pattern classification of ShearWave
Elastography images for differential diagnosis between
benign and malignant solid breast masses. Acta Radiol
2011;52:1069-1075.
84. Cong R, Li J, Guo S. A new qualitative pattern classifica-
tion of shear wave elastograghy for solid breast mass evalu-
ation. Eur J Radiol 2017;87:111-119.
85. Barr RG. Breast Elastography: How to Perform and Inte-
grate Into a “Best-Practice” Patient Treatment Algorithm. J
Ultrasound Med 2020;39:7-17.
86. Tanter M, Bercoff J, Athanasiou A, et al. Quantitative as-
sessment of breast lesion viscoelasticity: initial clinical re-
sults using supersonic shear imaging. Ultrasound Med Biol
2008;34:1373-1386.
87. Dighe M, Barr R, Bojunga J, et al. Thyroid Ultrasound:
State of the Art. Part 2 - Focal Thyroid Lesions. Med Ultra-
son 2017;19:195-210.
88. Dighe M, Barr R, Bojunga J, et al. Thyroid Ultrasound:
State of the Art Part 1 - Thyroid Ultrasound reporting
and Diffuse Thyroid Diseases. Med Ultrason 2017;19:79-
93.
89. Jiang M, Li C, Tang S, et al. Nomogram Based on Shear-
Wave Elastography Radiomics Can Improve Preoperative
Cervical Lymph Node Staging for Papillary Thyroid Carci-
noma. Thyroid 2020;30:885-897.
90. Cantisani V, David E, Sidhu PS, et al. Parotid Gland Le-
sions: Multiparametric Ultrasound and MRI Features. Ul-
traschall Med 2016;37:454-471.
91. Fukuhara T, Matsuda E, Donishi R, et al. Clinical efficacy
of novel elastography using acoustic radiation force im-
pulse (ARFI) for diagnosis of malignant thyroid nodules.
Laryngoscope Investig Otolaryngol 2018;3:319-325.
92. Cosgrove D, Barr R, Bojunga J, et al. WFUMB Guide-
lines and Recommendations on the Clinical Use of Ultra-
sound Elastography: Part 4. Thyroid. Ultrasound Med Biol
2017;43:4-26.
106 Giovanna Ferraioli et al
93. Cantisani V, David E, Grazhdani H, et al. Prospective
Evaluation of Semiquantitative Strain Ratio and Quantita-
tive 2D Ultrasound Shear Wave Elastography (SWE) in As-
sociation with TIRADS Classification for Thyroid Nodule
Characterization. Ultraschall Med 2019;40:495-503.
94. Bhatia KS, Cho CC, Tong CS, Lee YY, Yuen EH, Ahuja
AT. Shear wave elastography of focal salivary gland le-
sions: preliminary experience in a routine head and neck
US clinic. Eur Radiol 2012;22:957-965.
95. Badarinza M, Serban O, Maghear L, Bocsa C, Fodor D.
Shear wave elastography as a new method to identify pa-
rotid lymphoma in primary Sjögren Syndrome patients:
an observational study. Rheumatology International
2020;40:1275-1281.
96. Caliskan E, Ozturk M, Bayramoglu Z, Comert RG, Adaletli
I. Evaluation of parotid glands in healthy children and ado-
lescents using shear wave elastography and superb micro-
vascular imaging. La Radiologia Medica 2018;123:1-9.
97. Badea AF, Tamas Szora A, Ciuleanu E, et al. ARFI quan-
titative elastography of the submandibular glands. Normal
How to perform shear wave elastography. Part I
measurements and the diagnosis value of the method in ra-
diation submaxillitis. Med Ultrason 2013;15:173-179.
98. Mandalia UY, Porte FN, Howlett DC. Salivary Gland: On-
cologic Imaging. Ultrasound Clinics 2014;9:99-113.
99. Herman J, Sedlackova Z, Vachutka J, Furst T, Salzman R,
Vomacka J. Shear wave elastography parameters of normal
soft tissues of the neck. Biomed Pap Med Fac Univ Palacky
Olomouc Czech Repub 2017;161:320-325.
100. Chen YL, Chang C, Zeng W, Wang F, Chen JJ, Qu N.
3-Dimensional shear wave elastography of breast lesions:
Added value of color patterns with emphasis on crater
sign of coronal plane. Medicine (Baltimore) 2016;95:
e4877.
101. Tozaki M, Isobe S, Fukuma E. Preliminary study of ul-
trasonographic tissue quantification of the breast using the
acoustic radiation force impulse (ARFI) technology. Eur J
Radiol 2011;80:e182-187.
102. Zhou J, Zhan W, Chang C, et al. Breast lesions: evaluation
with shear wave elastography, with special emphasis on
the “stiff rim” sign. Radiology 2014;272:63-72.
Pictorial essay Med Ultrason 2022, Vol. 24, no. 1, 107-113
DOI: 10.11152/mu-2690

Cystic renal diseases: role of ultrasound. Part II, genetic cystic renal
diseases
Adnan Kabaalioglu1, Nesrin Gunduz2, Ayse Keven3, Emel Durmaz3, Mine Aslan4, Ahmet Aslan4,
Serkan Guneyli1

1Department of Radiology, Koc University School of Medicine, Istanbul, Turkey, 2Department of Radiology, Istanbul
Medeniyet University, School of Medicine, Istanbul, Turkey, 3Department of Radiology, Akdeniz University School
of Medicine, Antalya, Turkey, 4Department of Radiology, King Hamad University Hospital, Muharraq, Bahrain

Abstract
Kidney cysts are quite common in adults. Though small simple renal cysts in an adult over 30-40 years of age are not too
unusual, however, if the same cysts are seen in a child, and especially if there are additional findings, then several diagnostic
possibilities may come to mind. The role of ultrasound, together with the help of intravenous contrast agents and Doppler
mode, are very critical in describing the morphologic features and follow-up of the complex or multiple and bilateral renal
cysts. These sonographic signs are occasionally specific for diagnosis, but in many cases sonographic clues should be evalu-
ated together with the other genetic and clinical data to reach diagnosis.
The first part of this pictorial essay included the introduction into the subject and the classification of non-genetic cystic
renal diseases. The key features for the non-genetic cystic renal diseases are illustrated. In the second part, eye-catching fea-
tures of genetic cystic renal diseases are demonstrated.
Keywords: cyst; cystic; Doppler; kidney; ultrasound (US)

Introduction Genetic cystic renal diseases include autosomal dom-

In this second part, eye-catching features of genetic
cystic renal diseases are demonstrated. In general, cysts
are probably the most frequent abnormal findings to be
reported during sonography due to their high prevalenc-
es. Actually, simple cysts, even if they are multiple, are
considered to be almost a normal developmental varia-
tion in the elderly. On the other hand, even a tiny simple
cyst in a child may be the strong evidence of a genetic
cystic renal disease.
Received 26.07.2020  Accepted 29.09.2020
Med Ultrason
2022, Vol. 24, No 1, 107-113
Corresponding author: Assoc. Prof. Dr. Serkan Guneyli, M.D.
Department of Radiology, Koc University
School of Medicine, Istanbul, Turkey
Phone: 0090 533 6981477
Fax: 0090 212 311 3410
E-mail: drserkanguneyli@gmail.com
inant polycystic kidney disease (ADPKD), autosomal
recessive polycystic kidney disease (ARPKD), juvenile
nephronophthisis / medullary cystic kidney disease com-
plex, and multiorgan syndromes with pluricystic kidneys.
Genetic cystic renal diseases
Autosomal dominant polycystic kidney disease
ADPKD is the most well-known and most frequent
hereditary cystic disease which usually manifests in the
third decade of life. However, some patients are inciden-
tally discovered in their sixth or seventh decades; the
genetic basis for these almost 15% of cases, have been
discovered in recent years. Chromosome 4 is responsi-
ble in these mild cases instead of chromosome 16. In the
elderly, multiple bilateral cysts due to aging, may some-
times create a suspicion of ADPKD, then kidney volume
and other clues may be required for a certain diagnosis.
108 Adnan Kabaalioglu et al Cystic renal diseases: role of ultrasound. Part II, genetic cystic renal diseases
Diagnosis with ultrasound (US) is easy in overt cas-
es, with bilateral large kidneys, full of multiple cysts that
compress and hide the parenchyma, with a variable func-
tion (fig 1). Liver and other organ cysts may accompany
the disease. Diagnosis may be difficult in children or
young adults when only one or few tiny cysts can be seen
and the kidney volume is normal or borderline (fig 2).
Fig 1. Typical ADPKD; bilateral enlarged kidneys with hun-
dreds of tiny cysts compressing the parenchyma.
Fig 2. a) 11-year-old boy. Only 2 cysts are seen by the convex
probe. b) Linear probe shows more than 5 tiny cysts (arrows)
in each kidney of the same patient with ADPKD, similar to the
findings of his mother.
Fig 3. a) A complex hemorrhagic cyst (arrow) in an adult with
ADPKD. b) A complex hemorrhagic cyst in another adult with
ADPKD.
Fig 4. Impact of twinkling artifact in color Doppler US mode,
to ease the diagnosis of nephrolithiasis in this 45-year-old male
with ADPKD.
Fig 5. a) A solid mass (arrow) in a 55-year-old male with AD-
PKD which was found to be a RCC after resection. b) Power
Doppler US shows the vascularity of the mass.
Fig 6. a) Multiple tiny echogenic foci in the liver of a 70-year-
old male with ADPKD. The lesions had low density on CT (not
shown). b) Multiple cysts were also present in the liver.
The high-resolution linear probes can certainly change
the diagnosis in the majority of children screened by US
because of a parent with known ADPKD.
Hemorrhage and infection rates of cysts in ADPKD
are interestingly higher than the simple usual cysts in the
normal population [1] (fig 3). Calcifications and stones
are also frequent and may be difficult to find out by US
alone in a big mass; Doppler US and CT may help (fig 4).
Renal cell carcinoma (RCC) incidence is not increased in
patients with ADPKD; but if they start having dialysis,
then the risk increases (fig 5). Rarely, biliary hamartomas
may be seen; cysts in ADPKD and biliary hamartomas
are thought to develop from the interrupted remodeling
 Med Ultrason 2022; 24(1): 107-113 109

Fig 7. a) Bilateral big and echogenic kidneys in a 6-year-old boy. b) Intense twinkling artifacts are seen.

of the ductal plates during the late phase of embryologic

development [1] (fig 6).
Autosomal recessive polycystic kidney disease
ARPKD is primarily a pediatric disease that is more
severe than ADPKD; almost half of the patients die either
in utero or soon after birth. The liver is always involved
with progression to end-stage disease [2]. The kidneys
get larger, starting prenatally, and individual cysts may
not be seen especially in the early phase; instead echo-
genic parenchyma is noted. The echogenic medullary re-
gion may be described as “nephrocalcinosis” and almost
Fig 8. a) Typical, big for age (130 mm) echogenic kidneys
all cases show strong twinkling artifacts (fig 7, fig 8). The with tiny cysts and dilated tubules in a 7-month-old girl with
tiny cysts or rather dilated tubules can be seen with high- ARPKD. b) Striking parenchymal twinkling artifacts occur by
resolution probes as fusiform structures (fig 9a). Another Doppler US mode.
frequent sonographic observation is the lateralization of
the gallbladder due to relative atrophy of the right liver Nephronophthisis / Medullary cystic kidney disease
lobe and hypertrophy of the left liver lobe (fig 9b). Be- complex
sides diffuse liver parenchymal heterogeneity, biliary This group of rare genetic diseases with already some
dilatations and biliary cysts can be diagnosed by US as controversies in classifications, have recently been split
Caroli disease accompanying ARPKD (fig 10). as “nephronophthisis” and “autosomal dominant tubu-
Differential diagnosis includes other genetic syn- lointerstitial kidney disease” (ADTKD) in a consensus
dromes which may have a similar kidney appearance statement of an international working group composed
with extrarenal findings and unique genetic features [3]. mainly of pediatric nephrologists and pediatric radiolo-

Fig 9. a) Dilated fusiform tubules (arrows) and tiny cysts in a 4-year-old boy with ARPKD. b) Heterogeneity in the liver and later-
alization of the gallbladder (arrow).
110 Adnan Kabaalioglu et al Cystic renal diseases: role of ultrasound. Part II, genetic cystic renal diseases

Fig 10. a) Irregularly dilated bile ducts (arrows) in a 9-year-old girl with ARPKD and Caroli syndrome. Axial CT (b) and coronal
MRI (c) of the same patient after liver transplantation showing the polycystic kidneys, transplanted left liver lobe, and an enlarged
spleen.

Fig 11. Two different cases followed up with a probable diagnosis of nephronophthisis. Bilateral echogenic small kidneys with tiny
medullary cysts (arrows) are shown in a 19-year-old (a) and a 10-year-old girl (b).

gists [3]. ADTKD includes the previous “uremic med- tomas in the heart, brain, and kidneys [5]. Although renal
ullary cystic kidney diseases” and some more genetic involvement is asymptomatic initially, more than 75%
cystic diseases. US or other imaging methods are not so die of renal failure. The disease can be suspected in utero
useful and specific in these diseases; the kidneys may be by the presence of cardiac rhabdomyomas (fig 12). The
normal or small in size, the parenchyma may be echogen- most frequent renal finding is angiomyolipoma (AML)
ic, and medullary cysts may not always be seen [3]. The with an incidence of 40-90 % (fig 13). Fat-poor AMLs
most important clue is the specific location of small cysts
(almost always less than 2 cm), in the cortico-medullary
junction (fig 11). Caroli disease and hepatic fibrosis may
accompany and should be checked. Genetic analysis is
mandatory for verification of diagnosis.
Multiorgan syndromes with pluricystic kidneys
In addition to ADPKD and ARPKD, cystic kidney
disease is a common feature of ciliopathies with extrare-
nal manifestations which require careful clinical workup
to identify the underlying genetic disorder, especially in
children. In the recent years, our understanding of the
basis of polycystic kidney disease has increased substan-
tially and more than 100 genes have been discovered to
be involved in these cystic kidney diseases with enor-
mous complexity [2,4].
The most common disease in this diverse group is tu-
berous sclerosis complex (TSC). This disorder is usually Fig 12. The prenatal US shows a cardiac echogenic mass re-
identified in infants and children based on characteristic ported as probable rhabdomyoma. The child is still on follow-
skin lesions, seizures, and cellular overgrowth or hamar- up with TSC diagnosis.
 Med Ultrason 2022; 24(1): 107-113 111

Fig 13. a-c) Multiple, bilateral, different-sized, and highly echogenic angiomyolipomas in an 18-year-old boy with TSC.

are not rare and they can be confusing, since RCCs may
also be rarely seen in TSC patients. In these patients, US
may not be sufficient for the diagnosis, and CT or mag-
netic resonance imaging (MRI) is required for the dif-
ferential diagnosis. Bilateral multiple cysts are seen in
15-40 % of patients (fig 14). The cysts and the kidneys
are bigger earlier in childhood in the TSC2/PKD1 CGS
(Tuberous Sclerosis Complex 2/Polycystic Kidney Dis-
ease 1 Contagious Gene Syndrome) form of the disease
[6] (fig 15). AMLs larger than 3-4 cm should be observed
and evaluated for the need of embolisation against the Fig 14. Multiple, bilateral, and tiny cysts (arrows) in a 9-year-
old boy with TSC.
bleeding risk (fig 16).
Von Hippel-Lindau disease, which is a rare disorder
is characterized by cysts, cystic and hypervascular vis-
ceral neoplasms. Renal lesions, including renal cysts and
RCC are seen in 30–75% of cases [7]. Pancreatic cysts
and cystadenomas, neuroendocrine tumors and pheo-
chromocytomas may be also seen (fig 17).
In Meckel-Gruber syndrome, most cases are prenatal-
ly detected (90%) and die in utero or soon after birth [8].
Although many multiorgan anomalies are seen, the triad
of polycystic kidneys, encephalocele, and polydactyly is
the most common finding [8] (fig 18).
The diagnosis of HNF1B-associated disease can not
be made with imaging alone and requires genetic con- Fig 15. Multiple big cysts in big kidneys (16 cm) of a 4-year-
firmation [3]. It is considered to belong to the ADTKD old boy with Type 2 TSC.

Fig 16. a) Multiple AMLs (arrows) in the right kidney of a 17-year-old boy with TSC. One of them was a huge (14 cm in diameter)
fat-poor AML and embolised. This lesion (arrow) is shown on US (b) and CT (c).
112 Adnan Kabaalioglu et al Cystic renal diseases: role of ultrasound. Part II, genetic cystic renal diseases

Fig 17. A 28-year-old female with von Hippel-Lindau disease: a) Complex right renal cystic mass of 8 cm; b) Doppler US revealing
hypervascularity of cystic RCC which was bilateral; c) Pancreatic cysts are shown on US; d) CT shows both the cystic RCC (arrow)
in the right kidney and multiple pancreatic cysts (arrowheads).

group and is the most common cause of hyperechoic kid-

neys on the prenatal US. However, usually the renal US
findings are nonspecific; may mimic ARPKD, may be
unilateral, may reflect dysplasia, and may be normal. As-
sociated anomalies, especially female genital anomalies
should be screened with US [3].
McKusick-Kaufman syndrome may have similar
findings with HNF1B disease, and polydactyly may ac-
company genital anomalies and renal cysts (fig 19).
Bardet-Biedl syndrome, Beckwith–Wiedemann syn- Fig 18. a) Big and polycystic kidneys and b) encephalocele in a
drome, glomerulocystic kidney disease, Zellweger syn- fetus with anhydramnios.
drome, Joubert syndrome, Oral-facial-digital syndrome,
and Jeune syndrome are relatively well-known hereditary are known. Doppler and CEUS features are especially
multiorgan syndromes with cystic renal changes. How- important for differential diagnosis of neoplastic cystic
ever, the whole family is very large and includes multiple renal masses from other non-neoplastic cysts.
syndromes. Without genetic testing, diagnosis and differ-
ential diagnosis is not easy. US or MRI is most often not Acknowledgement: The authors thank Gokce Ak-
specific and not helpful in the initial diagnostic phase, gunduz Annac for creative illustrations in fig 1 in the first
and sonography is almost always used during follow-up. part.

Conclusion Conflict of interest: None.

Cystic renal diseases are a large spectrum contain- References

ing many genetic and non-genetic conditions, in which
sonographic signs might be very helpful in diagnosis if 1. Kabaalioglu A, MacLennan GT. Cystic Diseases of the
the meaning, sensitivity, and specificity of these US signs Kidney. In: Dogra VS, MacLennan GT (eds.). Genitouri-

Fig 19. Abdominal mass measuring 8 cm in a 10-month-old girl with polydactyly of both hands and feet was diagnosed as hydromet-
rocolpos due to vaginal atresia. Transverse (a) and longitudinal (b) sonograms show the huge abdominal midline cystic mass with
internal echoes. c) Both kidneys are enlarged and appear polycystic (arrows).

nary Radiology: Kidney, Bladder and Urethra. Springer-
Verlag London, 2013:95-119.
2. Bergmann C. Genetics of Autosomal Recessive Polycystic
Kidney Disease and Its Differential Diagnoses. Front Pedi-
atr 2018;5:221.
3. Gimpel C, Avni EF, Breysem L, et al. Imaging of Kidney
Cysts and Cystic Kidney Diseases in Children: An Inter-
national Working Group Consensus Statement. Radiology
2019;290:769-782.
4. Müller RU, Benzing T. Cystic Kidney Diseases From the
Adult Nephrologist’s Point of View. Front Pediatr 2018;6:65.
5. Randle SC. Tuberous Sclerosis Complex: A Review. Pedi-
atr Ann 2017;46:e166-e171.
Med Ultrason 2022; 24(1): 107-113 113
6. Back SJ, Andronikou S, Kilborn T, Kaplan BS, Darge
K. Imaging features of tuberous sclerosis complex with
autosomal-dominant polycystic kidney disease: a con-
tiguous gene syndrome.  Pediatr Radiol 2015;45:386-
395.
7. Katabathina VS, Kota G, Dasyam AK, Shanbhogue AK,
Prasad SR. Adult renal cystic disease: a genetic, biological,
and developmental primer.  Radiographics 2010;30:1509-
1523.
8. Barisic I, Boban L, Loane M, et al. Meckel-Gruber Syn-
drome: a population-based study on prevalence, prenatal
diagnosis, clinical features, and survival in Europe.  Eur J
Hum Genet 2015;23:746-752.
Case report Med Ultrason 2022, Vol. 24, no. 1, 114-116
DOI: 10.11152/mu-3325

Primary splenic leiomyosarcoma – case report and literature review

Elena Simona Ioanițescu1,2, Mugur Grasu2,3, Letiția Toma1,2

1Department of Internal Medicine, Fundeni Clinical Institute, 2”Carol Davila” University of Medicine and Pharmacy,
3Department of Interventional Radiology, Fundeni Clinical Institute, Bucharest, Romania

Abstract
Primary tumors of the spleen are rarely encountered in clinical practice and their diagnosis often requires invasive proce-
dures (splenectomy). Leiomyosarcomas are rare tumors originating from smooth muscle cells or their precursor mesenchymal
cells and as such can arise in any organs, most typically abdominal ones. Only a few cases of leiomyosarcomas of the spleen
have been described in literature. We present the case of a 69 year-old, a previously healthy patient, with non-specific symp-
toms, diagnosed on CT scan with multiple splenic, hepatic and bone tumors. Biopsy from one of the liver tumors revealed the
diagnosis of leiomyosarcoma. Due to characteristic aspects on contrast-enhanced ultrasonography and CT scan we concluded
that the primary tumor was located in the spleen, while the others represented metastases.
Keywords: leiomyosarcoma; spleen; contrast-enhanced ultrasonography

Introduction Leiomyosarcomas are rare tumors originating in

Malignant tumors of the spleen can be classified as
lymphoid, non-lymphoid and metastases, originating
most frequently from melanomas, breast and lung cancers
[1]. Primary involvement of the spleen in lymphomas is
much rarer than splenic infiltration during the course of
the disease [2] and the most frequently encountered are
non-Hodgkin’s lymphomas originating in B cells [3].
Non-lymphoid malignant tumors are rare and include
sarcomas, angiosarcomas or malignant teratomas [1].
A clear diagnosis regarding focal splenic lesions is
difficult to achieve by imaging means, particularly due
to imaging similarities and the scarcity of particular le-
sions [4–6]. Ultrasonography is the first step in diagnosis
and evaluation of splenic tumors, and contrast-enhanced
ultrasonography (CEUS) can add important information
regarding the characteristics of focal splenic lesions [7].
Received 22.06.2021  Accepted 11.09.2021
Med Ultrason
2022, Vol. 24, No 1, 114-116
Corresponding author: Elena Simona Ioanițescu
Department of Internal Medicine,
Fundeni Clinical Institute, 258 Fundeni Street,
Bucharest, Romania, 022328
E-mail simona.ioanitescu@gmail.com
Phone: +40722494586
smooth muscle cells; therefore, they may affect any organ
with a predilection for abdominal organs and the uterus.
They are so rare that they have not been included in the
latest WHO classification of splenic tumors [8]. To our
knowledge, there have been only three cases of primary
leiomyosarcomas of the spleen described in literature.
Case report
A 69 year-old female patient was admitted for dif-
fuse abdominal pain and general malaise, progressively
accentuated over the past two weeks. The patient had
no previously diagnosed conditions and had no chronic
medication. She denied constipation or diarrhea, as well
as melena or hematochezia. She was a non-smoker and
had no history of alcohol or illicit drug use. On admis-
sion, the patient signed an informed consent for scientific
and research purposes.
Clinical evaluation revealed a pale, dehydrated pa-
tient, with sclera jaundice, oxygen saturation of 97% in
ambient air, blood pressure 99/69 mmHg and 56 beats/
minute heart rate, bilateral lower limb edema. Abdominal
exam revealed a diffusely tender abdomen, with enlarged
liver (5 cm below lower rib) and spleen (4 cm below low-
er rib); there were no signs of acute abdomen. Neurologic
 Med Ultrason 2022; 24(1): 114-116 115

Fig 1. Liver CEUS examination: a) and c) arterial phase revealed iso-hypoenhancement followed by early wash out; d) and e) portal
and late phase showing progressive and marked wash out; b) reference image.

Fig 2. CEUS examination of the spleen: a) and b) arterial phase revealed multiple cystic lesions with slightly hypoenhancement of
septae and walls in some lesions, the majority isoenhancing; c) and d) venous and late phase; e) reference image.

evaluation revealed a disoriented patient, without other
neurologic deficits.
Blood tests were suggestive for liver insufficiency
(hypoalbuminemia, hypocholesterolemia) as well as
signs of kidney injury (creatinine 2.02 mg/dL), increased
levels of uric acid (11.7 mg/dL), severe anemia (6.6 g/
dL) and thrombocytopenia. Serum markers for chronic
viral hepatitis as well as for autoimmune disorders of the
liver were negative and the patient did not have proteinu-
ria. The patient was negative for HIV.
Abdominal ultrasonography revealed moderate as-
cites, enlarged lymph nodes in the upper abdomen, en-
larged liver with multiple hypoechoic tumors, measuring
up to 5 cm, important splenomegaly, with a long axis of 21
cm, with multiple focal lesions apparently similar to those
from the liver. CEUS was performed revealing arterial
iso-hypoenhacement of the liver lesions with rapid wash-
out, highly suggestive for liver metastases (fig 1). CEUS
of the spleen showed multiple cystic lesions, some of
them with septae, located in almost the entire spleen. The
lesions present arterial peripheral rim enhancement with
slightly or no wash-out (fig 2). Upper abdominal lymph
nodes were non-enhancing, suggesting necrosis (fig 3).
Upper endoscopy and colonoscopy had no pathologic
findings. Thoraco-abdominal CT scan revealed bilateral
pleural fluid, enlarged mediastinum, abdominal lymph
nodes and multiple hypodense liver and splenic lesions,
the largest located at the inferior splenic pole (93 mm)
(fig 4). A lytic lesion to the L1 vertebral body was also
revealed on CT-scan. Due to accessibility, we performed
a CT-guided biopsy of the liver lesions. Histopathologi-
cal analysis showed a mesenchymal proliferation of fusi-
form and epithelioid cells. Immune-histochemical test
were negative for CD117, CD34, WT1, DOG1 (exclud-
ing gastro-intestinal stromal tumor), and positive for Vi-
mentin (confirming mesenchymal origin) and SMA (spe-
cific for myofibroblasts).
We concluded that the diagnosis was primary leio-
myosarcoma of the spleen associated with liver, bones
and lymph nodes metastases. Due to the poor clinical sta-
tus of the patient, as well as the extension of disease, the
patient was referred for palliative treatment.
Fig 3. CEUS examination, late phase. Non-enhancing celiac
lymph node, suggesting necrosis (asterisc). Arrow indicates
a liver metastasis.
116 Elena Simona Ioanițescu et al
Fig 4. Abdominal CT scan (parenchymal phase) revealing mul-
tiple liver and splenic lesions, predominantly in the spleen.
Discussion
Splenic sarcomas are the rarest primary tumors aris-
ing in the spleen with only three cases of leiomyosarcoma
reported so far [8–10]. Two of these were in young wom-
en (49 and 54 years old) while the third was described in
an 87 year old male patient. Previous reports have based
their diagnosis on histological evaluation after splenec-
tomy, but in our case the intervention was considered to
have an increased risk and biopsy from metastases was
elected. Two of the previously reported cases described
hemorrhage secondary to the splenic lesions; in our case,
the patient’s anemia was corrected by blood transfusions
and ultrasonography and CT evaluation did not reveal
signs of bleeding. Another important aspect in our case
is the extent of the disease; the previously reported cases
had only splenic involvement, while our patient presented
liver, bones and lymph nodes metastases.
Primary splenic leiomyosarcoma – case report and literature review
Performing CEUS in our patient was an important
step in assessing the hepatic and splenic lesions, leading
to the diagnosis of primary tumor of the spleen. CEUS
proved the cystic appearance of the splenic lesions and
suggests for the first time the spleen as origin of the malig-
nancy. However, extensive immunohistochemical analy-
sis of the lesions was required for a positive diagnosis.
References
1. Fotiadis CI, Georgopoulos I, Stoidis C, Patapis P. Primary
tumors of the spleen. Int J Biomed Sci 2009;5:85–91.
2. Spier CM, Kjeldsberg CR, Eyre HJ, Behm FG. Malignant
lymphoma with primary presentation in the spleen. A study
of 20 patients. Arch Pathol Lab Med 1985;109:1076–1080.
3. Iliescu L, Mercan-Stanciu A, Ioanitescu ES, Toma L. Hepa-
titis C-Associated B-cell Non-Hodgkin Lymphoma: A Pic-
torial Review. Ultrasound Q 2018;34:156–166.
4. Ioanitescu ES, Copaci I, Mindrut E, et al. Various aspects
of Contrast-enhanced Ultrasonography in splenic lesions - a
pictorial essay. Med Ultrason 2020;22:2521.
5. Sidhu PS, Cantisani V, Dietrich CF, et al. The EFSUMB
Guidelines and Recommendations for the Clinical Practice
of Contrast-Enhanced Ultrasound (CEUS) in Non-Hepatic
Applications: Update 2017 (Long Version). Ultraschall
Med 2018;39:e2–e44.
6. Trenker C, Görg C, Freeman S, et al. WFUMB Position
Paper—Incidental Findings, How to Manage: Spleen. Ul-
trasound Med Biol 2021;47:2017-2032.
7. Omar A, Freeman S. Contrast-enhanced ultrasound of the
spleen. Ultrasound 2016 Feb;24:41–49.
8. Farah BL, Chee Y, Ching S, Tan C. Human Pathology: Case
Reports. Primary splenic leiomyosarcoma as an exception-
ally rare cause of ruptured splenomegaly – A case report
and review of primary splenic sarcomas. Hum Pathol Case
Reports 2020;22:200452.
9. Piovanello P, Viola V, Costa G, et al. Locally advanced leio-
myosarcoma of the spleen. A case report and review of the
literature. World J Surg Oncol 2007;5:135.
10. Daudia AT, Walker S, Morgan B, Lloyd DM. Images in surgery
Leiomyosarcoma of the spleen. Surgery 2001;130:893-894.
Case report Med Ultrason 2022, Vol. 24, no. 1, 117-119
DOI: 10.11152/mu-2677

Vomiting-induced costal cartilage fracture: a case report

Eleni Drakonaki1, Ioannis Karageorgiou2, Stamatios Kokkinakis2, Neofytos Maliotis2, Rania
Spyridaki3, Emmanouil K Symvoulakis2

1Department of Anatomy, European University of Cyprus Medical School, Nicosia, Cyprus, 2Clinic of Social and
Family Medicine, School of Medicine, University of Crete, Heraklion, Greece, 3Private Hospital Creta Interclinic,
Heraklion, Greece

Abstract
The use of ultrasonography as a first line imaging test in cases of possible costal cartilage fracture can be pivotal. In this
case report, we present the case of a patient with a suspected atraumatic vomiting-induced costal cartilage fracture. The costal
cartilage fracture was non-displaced and incomplete, thus not visible in a Computed Tomography scan. When Ultrasound
imaging was employed at the area of tenderness, soft tissue edema and hematoma around the cartilage were visualized. High
level of suspicion for a cartilage fracture in this case revealed a subtle osseous injury.
Keywords: costal cartilage; fractures, cartilage; vomiting; ultrasonography

Introduction
Thoracic cage injuries present a wide range of potential
diagnoses. Costal cartilage (CC) injuries pose a diagnos-
tic challenge, often being missed in the initial assessment
involving a chest radiograph. Costal cartilage fractures
should be considered, especially in cases of blunt trauma
involving contact sports athletes or high-energy trauma,
such as motor vehicle accidents [1]. These fractures are
often underdiagnosed, while atraumatic cases have rarely
been reported in the literature [2]. In this article, a case of
vomiting-induced costal cartilage fracture is presented.
Case report
A 53-year-old woman presented at a primary care ser-
vice institution due to significant pain at the left hemitho-
rax. The pain was more intense during deep breathing
Received 17.06.2020  Accepted 25.09.2020
Med Ultrason
2022, Vol. 24, No 1, 117-119
Corresponding author: Emmanouil K Symvoulakis
Assistant Professor of Primary Health Care
Clinic of Social and Family Medicine,
School of Medicine, University of Crete, Greece
Voutes, 71003, Heraklion, Greece
Phone: +302810394621
Email: symvouman@yahoo.com
and coughing. The patient reported that the pain had ap-
peared suddenly six days earlier when standing against
a hard surface at the toilet while vomiting. She also re-
ported that she had several episodes of vomiting in the
last week, due to an episode of gastroenteritis.
Clinical examination revealed no significant abnor-
mality. In palpation, there was an area of significant
tenderness over the left costal margin near the left ster-
nocostal synchondrosis. An ultrasonography (US) of the
area of pain was ordered, which revealed the findings de-
scribed in figure 1.
A CT scan was ordered to rule out a rib fracture. CT
images of the thorax identified no rib fracture. At axial
CT images there was a tiny low-density area at the sev-
enth costal cartilage and swelling of the adjacent soft tis-
sue. As the CT was non-diagnostic, further investigation
with MRI was performed (fig 2). Interestingly, the MRI
revealed a hyperintense linear area at the superior part
of the seventh costal cartilage that corresponded to the
tiny low density area identified in the CT. The diagno-
sis of a vertical incomplete fracture of the seventh costal
cartilage with associated soft tissue edema secondary to
injury or significant vomiting was made.
Analgesics and rest were recommended. At follow up
four weeks later, U/S showed that the soft tissue edema/
hematoma had partially resolved and the patient was
feeling significantly better.
118 Eleni Drakonaki et al
Fig 1. Ultrasound images over the painful area at the LT cos-
tal margin. Axial (a) and longitudinal (b) B-mode US images
showing the seventh and sixth left costal cartilage seven days
after injury. There is an irregular hypoechoic area (asterisk) sur-
rounding the seventh costal cartilage and the intercostal space
in keeping with hematoma. The subcutaneous tissue overlying
the hematoma is hyperechoic, which is probably secondary to
inflammation or hemorrhagic infiltration from the hematoma.
No fracture line is identified. Axial (c) and Longitudinal (d)
B-mode US images showing the seventh and sixth left costal
cartilage 30 days after injury. The hypoechoic area overlying
the costal cartilage has diminished in size, suggesting that the
hematoma has partially resolved. Doppler imaging (d) shows
limited vascularity at the subcutaneous tissue overlying the
area of injury and hematoma, suggesting an ongoing healing
procedure.
Discussion
After a brief literature review, we found that the most
well described mechanism of injury that approaches
vomiting is vigorous coughing. According to a recent
study, costal cartilage fracture due to violent coughing
was reported in two cases [2]. In our case, recurrent vom-
iting induced a costal cartilage fracture. We speculate
that this may be another potential cause of CC injury that
has not been previously reported.
The natural history of costal cartilage fracture remains
unclear. Observing the cartilage and soft-tissue abnormal-
ities of the anterior chest wall after injury is often difficult
with the radiographs due to their insensitivity. However,
eligible techniques that are more sensitive in revealing
these types of injuries include CT, MRI and US [3-5].
Treatment depends on the location and severity of the
fracture. Management in cases of single non-displaced
CC fractures is primarily conservative. As far as patients
with more complex trauma are concerned, operative
management may be beneficial in specific patient sub-
groups although this is not certain [6].
Vomiting-induced costal cartilage fracture: a case report
Even though the usefulness of CT in diagnosing cos-
tal cartilage fractures is well established [7] the same is
not true for MRI. There are some reports comparing find-
ings between diagnostic modalities [8], but are of limited
value, due to the small number of cases they examine.
The same is true for the role of US [9]. What the authors
seem to agree on is that costal cartilage fractures are un-
derdiagnosed [7] due to the lack of a reliable and readily
accessible imaging technique.
Our case was special, for two main reasons. Firstly,
patient history and presentation was atypical. Costal car-
tilage fractures occur either as sports injuries in contact
sports [4] or in case of high-energy blunt force trauma
such as motor vehicle accidents [3]. As a result, costal
cartilage fractures tend to occur in middle-aged males
[7]. Secondly, because the fracture was not displaced and
incomplete, a CT scan diagnosis was difficult to estab-
lish. In our case, MRI was more sensitive in detecting the
high signal intensity fracture line.
Finally, we would like to use this case as an oppor-
tunity to point out that US can be used as the first line
imaging test guided by the area of tenderness and pain
and showing soft tissue edema and hematoma around the
Fig 2. Coronal (a), (c) fat-suppressed T2-weighted images at
the painful left costosternal area reveals a hyperintense linear
area at the superior part of the seventh costal cartilage that
corresponds to a vertical incomplete fracture (arrow at c), ac-
companied by hyperintense soft tissue edema/hematoma at the
adjacent intercostal spaces (c) and the superficial soft tissues
(a). Axial (b) and corresponding coronal (d) CT images of the
thorax at the area of injury reveals a tiny low density area at the
seventh costal cartilage (arrows at b and d) that corresponds to
the incomplete costal cartilage fracture. Significant soft-tissue
edema/hematoma superficial and deep to the sixth and seventh
costal cartilage due to contusion injury is also present.

cartilage. If the US is positive, CT or MRI can be consid-
ered, depending on patient history, physical findings, and
clinical suspicion.
In conclusion, to the best of our knowledge, no vom-
iting-induced costal cartilage fracture has previously
been reported in the literature. A high level of suspicion
is required to make a correct diagnosis in atraumatic
cases. In non-displaced fractures, MRI seems to be su-
perior compared to chest CT and more studies regarding
its value are required. Ultrasound examination of the af-
fected area can either lead to a diagnosis in complicated
cases or imply that further imaging is needed, as in our
case.
References
1. Kani KK, Mulcahy H, Porrino JA, Chew FS. Thoracic Cage
Injuries. Eur J Radiol 2019;110:225-232.
2. Daniels SP, Kazam JJ, Yao KV, Xu HS, Green DB.
Cough-induced costal cartilage fracture. Clin Imaging
2019;55:161-164.
3. Nummela MT, Bensch FV, Pyhältö TT, Koskinen SK. Inci-
dence and Imaging Findings of Costal Cartilage Fractures
in Patients with Blunt Chest Trauma: A Retrospective Re-
Med Ultrason 2022; 24(1): 117-119 119
view of 1461 Consecutive Whole-Body CT Examinations
for Trauma. Radiology 2018;286:696-704.
4. McAdams TR, Deimel JF, Ferguson J, Beamer BS,
Beaulieu CF. Chondral rib fractures in professional
American football: two cases and current practice pat-
terns among NFL team physicians. Orthop J Sports Med
2016;4:2325967115627623.
5. Subhas N, Kline MJ, Moskal MJ, White LM, Recht MP.
MRI evaluation of costal cartilage injuries. AJR Am J
Roentgenol 2008;191:129-132.
6. Yuan SM. Isolated costal cartilage fractures: the radio-
graphically overlooked injuries. Folia Morphol (Warsz)
2017;76:139-142.
7. Malghem J, Vande Berg B, Lecouvet F, Maldague B. Cos-
tal Cartilage Fractures as Revealed on CT and Sonography.
AJR Am J Roentgenol 2001;176:429-432.
8. Tomas X, Facenda C, Vaz N, et al. Thoracic wall trau-
ma—misdiagnosed lesions on radiographs and usefulness
of ultrasound, multidetector computed tomography and
magnetic resonance imaging. Quant Imaging Med Surg
2017;7:384-397.
9. Griffith JF, Rainer TH, Ching AS, Law KL, Cocks RA,
Metreweli C. Sonography Compared With Radiography
in Revealing Acute Rib Fracture. AJR Am J Roentgenol
1999;173:1603-1609.
Letter to the Editor Med Ultrason 2022, Vol. 24, no. 1, 120-128

Ultrasonographic diagnosis and guided treatment of erector spinae

aponeurosis enthesopathy

I-Chun Liu1, Mathieu Boudier-Revéret2, Min Cheol Chang3, Ming-Yen Hsiao1,4

1Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan,
2Department of Physical Medicine and Rehabilitation, Centre hospitalier de l’Université de Montréal, Montreal,
Canada, 3Department of Rehabilitation Medicine, College of Medicine, Yeungnam University, Daegu, Republic of
Korea, 4Department of Physical Medicine and Rehabilitation, College of Medicine, National Taiwan University,
Taipei, Taiwan

To the Editor,

A 36-year-old female visited our outpatient clinic

with chronic low back pain for 7 months. Her pain was
mostly localized at the right lumbosacral area with oc-
casional referred pain at right gluteal region and was
aggravated by postural changes, particularly when bend-
ing and standing up after prolonged sitting. She had no
known systemic diseases or trauma history. Physical
examination demonstrated local tenderness on the right
posterior superior iliac spine (PSIS) and erector spine
muscle (ESM) without limitation of lumbar range of mo-
tion. No sensory deficit or focal weakness was found. Ra-
diographs of the lumbar spine were unremarkable.
Ultrasonography (US) revealed marked swollen and
hypoechoic erector spinae aponeurosis (ESA) at the en-
thesis near right PSIS, with a small linear calcification
and regional hetero-echogenicity, corresponding to the
most tender area (fig 1). Under the impression of ESA en-
thesopathy, US-guided injection with 1 ml 50% dextrose
and 1.5 ml 1% Xylocaine at right ESA enthesis was per-
formed. The patient reported a significant pain reduction
in posture changes immediately post-injection (50%) and
2 weeks after the 1st injection (more than 80%). There-
fore, a series of 3 injections at 3 to 4-week interval was

Received 27.01.2022  Accepted 07.02.2022
Med Ultrason
2022, Vol. 24, No 1, 120-121, DOI: 10.11152/mu-3596,
Corresponding author: Ming-Yen Hsiao, MD, PhD
Department of Physical Medicine and
Rehabilitation, College of Medicine,
National Taiwan University, Taipei, Taiwan
7, Zhongshan S. Rd., Zhongzheng Dist.,
Taipei City 100, Taiwan
Email: myhsiao@ntu.edu.tw
Phone: +886-23123456 ex 67316
Fig 1. Axial views of ESA (arrows) of left (asymptomatic) and
right (symptomatic) sides. Focal hypoechoic and swollen ESA
(arrowheads) at the right enthesis (A). A small linear calcifi-
cation (crosses, B) and regional hetero-echogenicity were also
noted (void arrows, C). The corresponding probe positions at
different levels of posterior medial iliac crest (D). US-guided
injection of the ESA enthesis (E).
arranged and the patient reported complete resolution of
symptoms subsequently.

ESA overlies the ESM dorsally in the lumbar region
and fuses with the thoracolumbar fascia caudally, attach-
ing to the iliac crest and sacrum [1]. Although ESA has
been proposed as a pain generator of lower back [2], im-
age-based pathological findings are rarely reported.
However, US-guided injection of the ESA enthesis
has been proposed as an effective method in treating iliac
crest pain syndrome [3]. On US examination, the ESA
can be visualized between the thoracolumbar fascia su-
perficially and ESM deeply, as a band-like hyperechoic
structure [4]. Our case demonstrated typical sonograph-
ic findings of ESA enthesopathy and its potential role in
non-specific low back pain.
Ultrasound guidance may be beneficial for localizing the atrophied
muscles in electromyography
Wei-Chen Huang, Yi-Hsiang Chiu, Kuo-Chang Wei
Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan
Dear Editor,
A 70-year-old man received a total endovascular re-
pair of thoracic aortic aneurysm with left axillary arte-
rial catheterization. Inability to flex left thumb, index
and middle fingers was noted immediately after surgery.
Additionally, he experienced numbness over left index
and middle fingers and visited the rehabilitation clinic.
Physical examination revealed poor left thumb and in-
dex finger flexion with atrophied left thenar and forearm
muscles. Tinel sign was elicited at the left axillary region.
Three months after surgery, nerve conduction studies
revealed prolonged distal motor latency with decreased
Received 27.11.2021  Accepted 23.01.2022
Med Ultrason
2022, Vol. 24, No 1, 121-122, DOI: 10.11152/mu-3548,
Corresponding author: Kuo-Chang Wei, MD.
Department of Physical Medicine and
Rehabilitation, National Taiwan University
Hospital, Taipei, Taiwan.
46, Gongyuan Rd., Zhongzheng Dist.,
Taipei City 100, Taiwan (R.O.C.).
E-mail: gordon80446@gmail.com
Phone: 886-2-23123456-66635
Med Ultrason 2022; 24(1): 120-128 121
References
1. Daggfeldt K, Huang QM, Thorstensson A. The visible hu-
man anatomy of the lumbar erector spinae. Spine (Phila Pa
1976) 2000;25:2719-2725.
2. Creze M, Soubeyrand M, Nyangoh Timoh K, Gagey O. Or-
ganization of the fascia and aponeurosis in the lumbar par-
aspinal compartment. Surg Radiol Anat 2018;40:1231-1242.
3. Ricci V, Ozcakar L. Ultrasound-guided injection of the
erector spinae enthesis for iliac crest pain syndrome. J Res
Med Sci 2019;24:69.
4. Todorov P, Nestorova R, Batalov A. The sonoanatomy of
lumbar erector spinae and its iliac attachment - the poten-
tial substrate of the iliac crest pain syndrome, an ultrasound
study in healthy subjects. J Ultrason 2018;18:16-21.
compound motor action potential and nearly absent sen-
sory nerve action potential in the left median nerve. EMG
needle was inserted into pronator teres (PT) using surface
anatomy for localization but failed to precisely locate
flexor pollicis longus (FPL) as the expected increased
spontaneous activities were not observed. Ultrasound
(US)-guided muscle samplings were applied and there
were marked spontaneous activities in FPL and flexor
digitorum superficialis (FDS) (fig 1a,b). Denervation
signs, including hyperechoic texture and decreased vol-
ume, were observed in multiple median-innervated fore-
arm muscles (fig 1c,d). US nerve tracking revealed focal
swelling of the median nerve with loss of typical honey-
comb appearance next to the left axillary artery. These
findings were indicative of proximal median neuropathy
at axillary level.
US guidance has been significantly important for pre-
cise EMG samplings of atrophied muscles because it not
only provide precise localization but also prevents un-
necessary neurovascular injuries. Here, accidental radial
artery puncture might have occurred without US guid-
ance during FPL sampling, considering their adjacency.
US is useful in detecting muscles’ denervation changes,
with respect to their nerve innervation. The denervation
122 Wei-Chen Huang et al Ultrasound guidance may be beneficial for localizing the atrophied muscles in electromyography

Fig 1. Ultrasound-guided electromyography needle sampling of left (a) flexor pollicis longus (FPL) (blue area) and (b) flexor digito-
rum superficialis (FDS) muscles (red area). Radial artery (RA) lied closely above the atrophied FPL muscle (blue area). Denervation
signs, including hyperechoic texture and decreased volume, of median-innervated muscles were noted at (c) proximal and (d) distal
forearm levels. As muscles atrophied (left), median nerve (arrow) became more superficial. Dashed arrow, electromyography needle;
open arrow, pronator quadratus muscle; arrowhead, flexor digitorum superficialis muscle; R, radius; U, ulna.

pattern of muscles may infer the nerve injury site [1]. In
our case, atrophied left forearm flexors appeared hyper-
echoic under US, indicating median nerve injury level
was at or proximal to elbow level. Since all median-in-
nervated muscles lie within the forearm and hand, injury
level cannot be determined only by EMG when the me-
dian nerve is injured above the elbow. US nerve tracking
provides additional morphological information on the
injured nerve when incorporated into electrodiagnostic
studies [2]. In conclusion, we suggest combining US and
EMG which could make electrodiagnostic studies more
precise and safer.
References
1. Gunreben G, Bogdahn U. Real-time sonography of acute
and chronic muscle denervation. Muscle Nerve 1991;14:
654-664.
2. Domkundwar S, Autkar G, Khadilkar SV, Virarkar M. Ul-
trasound and EMG-NCV study (electromyography and
nerve conduction velocity) correlation in diagnosis of nerve
pathologies. J Ultrasound 2017;20:111-122.

Imaging findings of a tall cell variant of papillary breast carcinoma

Maohua Pang1, Mingyuan Yuan1, Min Yu2

1Department of Imaging Medicine, 2Department of Ultrasound, Shanghai University of Medicine and Health Sciences
Affiliated Zhoupu Hospital, Shanghai, P.R. China

Received 20.01.2022  Accepted 23.01.2022 To the Editor,

Med Ultrason
2022, Vol. 24, No 1, 122-123, DOI: 10.11152/mu-3592,
Corresponding author: Mingyuan Yuan, MD
Department of Imaging Medicine, Shanghai
University of Medicine and Health Sciences
Affiliated Zhoupu Hospital,
1500 Zhouyuan Road, Pudong New Area,
Shanghai, 201318, P.R. China
Phone/fax: +86 15221353727
+86 02168139123
E-mail: zp_yuanmy@sumhs.edu.cn
A 63-year-old female with a left breast mass that ap-
peared 5 years ago was hospitalized, due to nipple bleed-
ing for 1 week. Occasionally, a little milk-like or light-yel-
low liquid flowed out of the nipple. We performed breast
ultrasound (US), magnetic resonance imaging (MRI)
and molybdenum target imaging for diagnostic purpose.
A cystic solid nodule was seen in the left breast at 12
o’clock, with a size of approximately 18×10 mm, clear

Fig 1. a) Breast ultrasound: A cystic solid nodule, clear bound-
ary, angular edges and uneven internal echo; b) Breast MRI:
The scanning lesions are unevenly enhanced, and the time sig-
nal curve is flat; c) Breast molybdenum target: Circular equal
density mass shadow, irregular contour; d) Histological mani-
festations: The nuclei of neoplastic columnar epithelium are
reversed in a polar direction.
boundary, angular edges and uneven internal echo on
breast US (fig 1a). Breast MRI revealed a mass of ap-
proximately 1.5×1.4×1.4 cm in the left outer upper quad-
rant, with a slightly low signal on T1WI, high signal on
T2WI lipid suppression sequence, high signal on DWI,
and no decrease in ADC signal. After enhancement, the
scanning lesions were unevenly enhanced, and the time
the signal curve was flat (fig 1b). Breast molybdenum tar-
get imaging depicted circular equal density mass shadow
in the upper quadrant of the left outer breast, with fuzzy
edge, irregular contour, length, and diameter of approxi-
mately 1.5 cm (fig 1c). The left breast was scattered with
circular and spotted calcifications, and no obvious calci-
fication was found in the right breast.
Our patient underwent left breast-conserving and
sentinel lymph node biopsy. Postoperative pathology re-
vealed a high cell subtype papillary carcinoma in the left
breast (fig 1d). Immunohistochemistry revealed that PR,
Med Ultrason 2022; 24(1): 120-128 123
CerbB-2, p63, calponin, Syn, CD56, CgA, TG, TTF-1,
TTF1, and TG were negative, while CK7, ER, EMA,
GCDFP-15, GATA3 and CK19 were positive.
The tall cell variant of papillary breast carcinoma is
a rare subtype of invasive breast cancer, characterized
by high polarity columnar cells with nuclear polarity re-
versal. Eusebi et al [1] reported five cases of malignant
tumours with primary breast histomorphology similar to
thyroid high cell subtype papillary carcinoma.
US is the first choice of imaging examination for
breast tumors, with the advantages of convenience and
non-invasiveness. However, sometimes, doctors may be
inexperienced in diagnosing such rare breast cancer sub-
types, leading to misdiagnosis. We believe that for the
preoperative evaluation of breast tumours, breast ultra-
sound, MRI and molybdenum target imaging should be
performed at the same time to achieve diagnostic accu-
racy.
The clinical course of high cell subtype papillary
breast carcinoma is slow and it is a low-grade malignant
tumor [2]. Our patient’s left breast tumour of 5 years also
suggested that the tumour grew slowly. We believe that
breast-conserving surgery plus anterior sentinel lymph
node biopsy should be considered, followed by close
follow-up. After surgery, radiotherapy, chemotherapy, or
targeted therapy may be performed based on personal-
ized comprehensive evaluation of pathological results.
Acknowledgements: This study was supported by
the 2020 New Interdisciplinary Construction Project Of
Shanghai Pudong New Area Health (PWXx2020-06).
References
1. Eusebi V, Damiani S, Ellis IO, Azzopardi JG, Rosai J.
Breast tumor resembling the tall cell variant of papillary
thyroid carcinoma: report of 5 cases. Am J Surg Pathol
2003;27:1114-1148.
2. Foschini MP, Asioli S, Foreid S, et al. Solid papillary breast
carcinomas resembling the tall cell variant of papillary thy-
roid neoplasms: A unique invasive tumor with indolent be-
havior. Am J Surg Pathol 2017;41:887-895.
124 Maohua Pang et al Imaging findings of a spindle epithelial tumour with thyroid thymoid differentiation

Imaging findings of a spindle epithelial tumour with thyroid thymoid

differentiation

Maohua Pang1, Mingyuan Yuan1, Min Yu2

1Department of Imaging Medicine, 2Department of Ultrasound, Shanghai University of Medicine and Health Sciences
Affiliated Zhoupu Hospital, Shanghai, P.R. China

To the Editor,

A 81-year-old male who conducted partial thyroidec-

tomy 16 years ago (we have no information about the spe-
cific operation method and postoperative pathology), was
diagnosed with a spindle epithelial tumour with thymus-
like differentiation (SETTLE) in the left residual thyroid.
Thyroid ultrasonography (US) demonstrated a sub-
stantial mixed echo lump under the left residual thyroid,
32×26×45 mm, ovalize, uneven internal echo with inter-
nal dot strong echo, smooth and clean edge, no obvious
acoustic halo and no evident change in the posterior echo
(fig 1a). Colour Doppler showed a blood flow signal in
the internal and edge portion (fig 1b).
The tumour showed biphasic differentiation, which
was a typical component of spindle cells with different
degrees of fibrosis. Spindle cells were arranged in a stag-
gered beam, small cytoplasm, long and thin nucleus, fine
chromatin, unclear nucleolus, few nuclear fission, visible
epithelial components, and the cells were arranged in a
tubular, papillary, striate, and glandular pattern. In im-
munohistochemistry, adenoids of the tumour CK broad
spectrum and EMA expression were strong (fig 1c) and
spindle cell cytoplasm CK, vimentin and P63 were ex-
pressed (fig 1d). The Thyroglobulin, TTF1, Desmin,
CD5, CD117, Calcitonin and CD34 did not express.
SETTLE is a dramatically rare and low-degree malig-
nant tumour in the thyroid. In 1991, Chan and Rosai [1]
Received 08.01.2022  Accepted 23.01.2022
Med Ultrason
2022, Vol. 24, No 1, 124-125, DOI: 10.11152/mu-3575,
Corresponding author: Mingyuan Yuan, MD
Department of Imaging Medicine, Shanghai
University of Medicine and Health Sciences
Affiliated Zhoupu Hospital,
1500 Zhouyuan Road, Pudong New Area,
Shanghai, 201318, P.R. China
Phone/fax: +86 15221353727
+86 02168139123
E-mail: zp_yuanmy@sumhs.edu.cn
Fig 1. a) Thyroid ultrasound demonstrated a mixed echo lump
under the left thyroid; b) colour Doppler aspect of the mass;
c) immunohistochemical CK tumour cells turned to be posi-
tive; d) immunohistochemical P63 tumour cells were found to
be positive.
reported about SETTLE for the first time and found that
ectopic hamartoma type thymoma, ectopic thymoma,
Spindle epithelial tumour with thymus-like differentia-
tion and thyroid carcinoma with adenoid differentiation.
In addition, they listed a series of tumour pedigree con-
cepts that indicated the thymus characteristics on the
neck successively from benign to malignant. As far as
we know, less than 50 cases were published, mainly in
children, young people and rarely in elders [2].
To our knowledge our case is the oldest SETTLE
patient. After complete surgical excision, the post-op-
eration follow-up after 28 months reported a favourable
prognosis. SETTLE has low degree malignancy, long-
term occurrence and the distant metastasis, in particular
the blood bank transfer, appear in an advanced stage.
It is usually transferred into the lung with possible lo-
cal lymph node metastasis, but even after metastasis, the
majority of patients have demonstrated a relatively long
lifetime [3,4].
In our case, the patient underwent left thyroidectomy
16 years ago and it was impossible to identify whether
her SETTLE was a recurrent case. The postoperative

follow-up after 28 months did not show recurrence or
metastasis, but further observation is needed.
Acknowledgements: This study was supported by
the 2020 New Interdisciplinary Construction Project Of
Shanghai Pudong New Area Health (PWXx2020-06).
References
1. Chan JK, Rosai J. Tumors of the neck showing thymic or
related branchial pouch differentiation: a unifying concept.
Hum Pathol 1991;22:349-367.
Med Ultrason 2022; 24(1): 120-128 125
2. Misra RK, Mitra S, Yadav R, Bundela A. Spindle epi-
thelial tumor with thymus-like differentiation: a case
report and review of literature. Acta Cytol 2013;57:303-
308.
3. Erickson ML, Tapia B, Moreno ER, McKee MA, Kowalski
DP, Reyes-Múgica M. Early metastasizing spindle epithe-
lial tumor with thymus-like differentiation of the thyroid.
Pediatr Dev Pathol 2005;8:599-606.
4. Abrosimov AY, Li Volsi VA. Spindle epithelial tumor
with thymus-like differentiation of the thyroid with neck
lymph node metastasis: a case report. Endocr Pathol 2005;
16:139-143.

Ultrasound and clinical findings of hyalinizing trabecular tumor of

the thyroid

Yong Jun Xing1*, Jing Zhang2*, Bi Shun Yi1

* the authors share the first authorship

1Department
of General Surgery, Lishui People’s Hospital, Lishui, 2Department of Ministry of Women’s Health,
Dongyang Maternal and Child Care Hospital, Dongyang, Zhejiang, P.R. China

To the Editor

A 47-year-old female incidentally found a lump,

about the size of a pecan, in her right neck. At clinical
exam a 3-cm diameter nodule in the right anterior neck,
non-tender, with clear border and mobile with swallow-
ing was found.
Thyroid ultrasound (US) evidenced in the right
thyroid a 30×20×45 mm hypoechoic mass with clear
boundary (fig 1a), aspect ratio <1, uneven internal echo,
abundant blood flow signal, resistance index of 0.64 and
TI-RADS class of 4a.
Gross inspection during a right thyroidectomy showed
that the nodules had approximately 4.0×3.0 cm and a clear
border. The intraoperative freezing report demonstrated a
right thyroid follicular tumor. The primary consideration
was hyalinizing trabecular tumor (HTT). Postoperative
Received 30.01.2022  Accepted 01.02.2022
Med Ultrason
2022, Vol. 24, No 1, 125-126, DOI: 10.11152/mu-3601,
Corresponding author: Bi Shun Yi
Department of General Surgery,
Lishui People’s Hospital
15 Dazhong street, Liandu District,
Lishui, Zhejiang, 323000, P.R.China
Phone/fax: +86 18957091991
+86 05782780253
E-mail: ybs18957091991@163.com
Fig 1. Thyroid ultrasonography evidenced a low-echo nodule
in the right gland, bounded and irregular, with uneven internal
echoes. b) Microscopically, the tumor has a hyalinizing fibrosis
around the blood vessels (hematoxylin and eosin stain ×200).
pathology confirmed the right thyroid HTT diagnosis (fig
1b). Immunohistochemistry: CK(+), Thyrogloblin(+),
TTF-1(+), CK19(-), Galectin-3(-), Calcitonin(-), CGA(-),
Syn(-). The postoperative recovery was good.
HTT is an extremely rare tumor of follicular origin,
named also paraganglioma, transversely trabecular ad-
enoma or papillary carcinoma. It is thought to originate
from thyroid follicular epithelium, but its etiology re-
mains unclear. In a small number of patients, the condition
is associated with radiation exposure, but some scholars
speculate that some patients with HTT had the condi-
tion in relation to chronic lymphocytic thyroiditis [1].
US shows poor specificity in identifying HTT that
can be easily misdiagnosed as thyroid adenoma, papil-
126 Yong Jun Xing, Jing Zhang et al Ultrasound and clinical findings of hyalinizing trabecular tumor of the thyroid
lary carcinoma or other tumors [2], being round or oval,
solid, hypoechoic and clear. Few nodules have unclear
borders, characteristics closely resembling that of papil-
lary thyroid carcinoma but without calcification. Intra-
operative frozen section can sometimes help in diagnos-
ing HTT and patients can avoid total thyroidectomy [3].
Although most of them are benign tumors, there are still
a few with malignant potential, recurrence, or even me-
tastasis risk. Therefore, patients with malignant potential
need close and regular follow-ups post-operation.
US is the first choice for the imaging examination
of thyroid nodules as it is a non-invasive, radiation-free,
convenient and fast tool. However, rare thyroid tumors
may be easily misdiagnosed as thyroid cancer.
Comment on “Usefulness of lung ultrasound in the early identification
of severe COVID-19: results from a prospective study”
Rujittika Mungmunpuntipantip1, Viroj Wiwanitkit2
1Private Academic Consultant, Bangkok Thailand, 2Dr DY Patil University, Pune, India
Dear Editor,
We would like to share ideas on “Usefulness of lung
ultrasound in the early identification of severe COVID-
19: results from a prospective study [1].” Hernández-
Píriz et al noted that “The combination of the ultrasound
score and the presence of respiratory failure can easily
identify patients with a higher risk to present complica-
tions [1].” We agree that ultrasound might be useful for
early detection of severe COVID-19. Results in this re-
port are concordant with those reported by Chardoli et
al [2]. However, Chardoli et al observed that ultrasound
finding on anterior lung field has little clinical value
for prediction of clinical severity [2]. Nevertheless, the
Received 12.11.2021  Accepted 09.01.2022
Med Ultrason
2022, Vol. 24, No 1, 126, DOI: 10.11152/mu-3494,
Corresponding author: Rujittika Mungmunpuntipantip
Private Academic Consultant,
Bangkok Thailand
Email: rujitttika@gmail.com
References
1. You TK, Jang KY, Moon WS, et al. Fine-needle aspiration
cytology of hyalinizing trabecular adenoma of the thyroid
in a patient with Hashimoto’s thyroiditis: A case report.
Acta Cytol 2012;56:448-452.
2. Lee S, Han BK, Ko EY, Oh YL, Choe JH, Shin JH. The
ultrasonography features of hyalinizing trabecular tumor
of the thyroid are more consistent with its benign be-
havior than cytology or frozen section readings. Thyroid
2011;21:253-259.
3. Sung SY, Shen HY, Hsieh CB, et al. Hyalinizing trabecular
tumor of thyroid: Does frozen section prevent unnecessar-
ily aggressive operation? Six new cases and a literature re-
view. J Chin Med Assoc 2014;77:573-577.
prognosis of the patients also depends on other factors
including to underlying concurrent medial disorder and
appropriate treatment of the illness. Additionally, ultra-
sound tool might not be available in remote area of poor
countries. In case that ultrasound tool is required, experi-
ence of user is important. A good training and proficiency
control are required to maintain the usefulness of ultra-
sound for early diagnosis of severe COVID-19 [3].
References
1. Hernández-Píriz A, Tung-Chen Y, Jiménez-Virumbrales D,
et al. Usefulness of lung ultrasound in the early identifica-
tion of severe COVID-19: results from a prospective study.
Med Ultrason. 2021 . doi: 10.11152/mu-3263.
2. Chardoli M, Sabbaghan Kermani S, Abdollahzade Man-
qoutaei S, et al. Lung ultrasound in redicting COVID-19
clinical outcomes: A prospective observational study..J Am
Coll Emerg Physicians Open 2021;2:e12575.
3. Cid X, Wang A, Heiberg J, et al. Point-of-care lung ultra-
sound in the assessment of patients with COVID-19: A tu-
torial. Australas J Ultrasound Med 2020;23:271-281.

Comment on “Usefulness of lung ultrasound in the early identification
of severe COVID-19: results from a prospective study”
Domenica Di Costanzo1, Mariano Mazza1, Antonio Esquinas2
1Department of Respiratory Disease, AORN Sant’Anna e San Sebastiano di Caserta, Caserta, Italy, 2Intensive Care
Unit, Hospital Morales Meseguer, Murcia, Spain
To the Editor,
We have read the paper of Hernández-Píriz et al [1]
with great interest, where a pattern of correlation was
established between lung ultrasound (LUS) findings and
the degree of respiratory failure and prognosis in patients
with COVID-19. However, we think that some aspects of
this study could have been better analyzed.
In this study the timing from hospital admission to
LUS evaluation is unclear. Particularly, it is not specified
if patients were receiving any non-invasive ventilatory
support while performing LUS since they were trans-
ferred to the hospital on average of 2.51±3.95 days from
arrival at the emergency department (ED). This aspect
may be relevant since the use of High-Flow Nasal Can-
nula Oxygen Therapy (HFNCOT) could improve lung
aeration as well as Non-Invasive Ventilation (NIV) in-
duces alveolar recruitment thus affecting LUS evaluation
[2].
Also, the authors did not incorporate diaphragm ul-
trasonography, which may be an interesting parameter to
evaluate respiratory compromise.
About methodology, criteria for initiating NIV have
not been defined. Also, to assess the relation between
LUS findings and degree of respiratory failure, other
endpoints could be considered: need to increase standard
oxygen therapy [3] or use of HFNCOT [4]. Moreover,
pulse-oximetric saturation (SpO2) was used when arte-
rial blood gas analysis was not available: controversies
exist about the Severinhause-Ellis equation since SpO2
is dependent on variables related to patient (local perfu-
Received 20.01.2022  Accepted 09.02.2022
Med Ultrason
2022, Vol. 24, No 1, 127, DOI: 10.11152/mu-3591,
Corresponding author: Antonio Esquinas
Intensive Care Unit,
Hospital Morales Meseguer,
Murcia, Spain
E-mail: antmesquinas@gmail.com
Med Ultrason 2022; 24(1): 120-128 127
sion, temperature) and therefore not as reliable as blood
gas analysis.
About patient selection, LUS findings in COVID-19
can occur in other pathological processes such as bac-
terial pneumonia, left ventricular failure, atelectasis and
fibrosis. For this reason, studies about this topic have
considered different exclusion criteria: diagnosis of car-
diogenic acute pulmonary oedema and interstitial lung
disease [5], bacterial pulmonary superinfection [4], lung
malignancy or lobectomy [6]. Furthermore, it is possi-
ble that some patients with mild disease may have been
discharged from ED and excluded from the study thus
affecting the prevalence of LUS findings.
Further clinical trials are required to evaluate the cor-
relation between LUS and outcome in COVID-19.
References
1. Hernández-Píriz A, Tung-Chen Y, Jiménez-Virumbrales D,
et al. Usefulness of lung ultrasound in the early identifica-
tion of severe COVID-19: results from a prospective study.
Med Ultrason 2021. doi:10.11152/mu-3263.
2. Volpicelli G, Lamorte A, Villén T. What’s new in lung ul-
trasound during the COVID-19 pandemic. Intensive Care
Med 2020;46:1445-1448.
3. Rubio-Gracia J, Sánchez-Marteles M, Garcés-Horna V, et
al. Multiple Approaches at Admission Based on Lung Ul-
trasound and Biomarkers Improves Risk Identification in
COVID-19 Patients. J Clin Med 2021;10:5478.
4. Gil-Rodríguez J, Martos-Ruiz M, Peregrina-Rivas JA, et al.
Lung Ultrasound, Clinical and Analytic Scoring Systems as
Prognostic Tools in SARS-CoV-2 Pneumonia: A Validating
Cohort. Diagnostics (Basel) 2021;11:2211.
5. Biasucci DG, Buonsenso D, Piano A, et al. Lung ultrasound
predicts non-invasive ventilation outcome in COVID-19
acute respiratory failure: a pilot study. Minerva Anestesiol
2021;87:1006-1016. 
6. Chardoli M, Sabbaghan Kermani S, Abdollahzade Man-
qoutaei S, et al. Lung ultrasound in predicting COVID-19
clinical outcomes: A prospective observational study. J Am
Coll Emerg Physicians Open 2021;2:e12575.
128 Alba Hernández-Píriz et al Author’s Reply

Author’s Reply

Alba Hernández-Píriz1, Yale Tung Chen2,3, David Jiménez-Virumbrales4, Ibone Ayala-

Larrañaga1, Raquel Barba-Martín5, Jesús Canora-Lebrato1,6, Antonio Zapatero-Gaviria1,6,
Gonzalo García De Casasola-Sánchez7

1Department of Internal Medicine, Hospital Universitario Fuenlabrada Fuenlabrada, Madrid, Spain, 2Department of
Internal Medicine, Hospital Universitario Puerta de Hierro, Majadahonda, Madrid, 3Department of Medicine, Univer-
sidad Alfonso X El Sabio, Madrid, 4Department of Internal Medicine, Hospital Universitario Severo Ochoa, Leganés,
5Department of Internal Medicine, Hospital Rey Juan Carlos, Móstoles, Madrid, 6Department of Medicine, Universi-

dad Rey Juan Carlos, Madrid, 7Department of Internal Medicine, Hospital Infanta Cristina, Madrid, Spain

Dear Editor,
We read with interest the comments sent by Di Cos-
tanzo et al. The authors agree on the potential role of lung
ultrasound (LUS) and the degree of respiratory failure
and prognosis in patients with COVID-19 [1] but raise
some methodological questions.
First, we would like to highlight that the study was
carried out in a field hospital, during the first wave of the
pandemic in Spain, to decongest hospitals. That is why
we did not have the facilities used in conventional hos-
pitals. Patients were admitted to a conventional Internal
Medicine ward. Only 5 of the 107 patients had received
positive ventilatory support [2] before LUS (2 of them
in the first 10 hours), this is unlikely to have an impact.
None had High-Flow Nasal Cannula Oxygen Therapy
(HFNCOT). The timing from admission to LUS evalu-
ation was 7.64±4.13 days and we wanted to emphasize
the number of days from symptom onset as an important
marker to develop adult respiratory distress syndrome
(ARDS).
Regarding diaphragmatic ultrasound we agree it is an
interesting parameter worth to study, however, our study
focused on alterations of the lung parenchyma and the se-
verity of ARDS and, at that moment, was not considered.
We agree that the use of partial pressure of oxygen is
more reliable than oxygen saturation by pulse oximetry,
however, we did not have access to a 24-hour laboratory
to perform it.
Although many patients with mild disease were not
admitted, our aim was to correlate the ultrasound find-
ings with respiratory failure, and in other studies it has
been shown that these patients with mild disease have
little sonographic abnormalities [3].
The possibility of other processes was also studied,
since an echocardiogram was performed along with
LUS, with no evidence suggesting heart failure, bacterial
superinfection and none of our patients had a history of
pulmonary fibrosis at the time of inclusion in the study.
The criteria for starting NIV were severe dysp-
nea, tachypnoea over 30 bpm, PaO2/FiO2 <200 (or the
need for FiO2 greater than 0.4 to achieve an SpO2 of at
least 92%) or acute ventilatory failure (pH <7.35 with
PaCO2 >45 mm Hg).
We appreciate the comments and agree that it is nec-
essary new well-designed studies to provide new insights
on the correlation between LUS and prognosis of COV-
ID-19.
References
1. Hernández-Píriz A, Tung-Chen Y, Jiménez-Virumbrales D,
et al. Usefulness of lung ultrasound in the early identifica-
tion of severe COVID-19: results from a prospective study.
Med Ultrason 2021 Doi: 10.11152/mu-3263.
2. Mateos-Rodríguez A, Ortega-Anselmi J, Candel-González
FJ, et al. Alternative CPAP methods for the treatment of
secondary serious respiratory failure due to pneumonia by
COVID-19, Med Clin (Barc) 2021;156:55-60.
3. Tung-Chen Y, Martí de Gracia M, Díez-Tascón A, et al. Cor-
relation between Chest Computed Tomography and Lung
Ultrasonography in Patients with Coronavirus Disease
2019 (COVID-19). Ultrasound Med Biol 2020;46:2918-
2926.
In memoriam
A mai plecat dintre noi un DOCTOR MARE,
Mircea Leonid Stamate
A absolvit în 1966 Facultatea de Pediatrie la U.M.F.
„Carol Davila”. A devenit medic specialist pediatru în
1971 și a obținut doctoratul în medicină în 1983, apoi a
devenit medic primar în 1984. S-a specializat în cardi-
ologie pediatrică în 1990. A făcut stagii de specializare în
ecografie generală la Spitalul „Robert Debre”, Paris și la
Spitalul din Ebry, Franța, în 1991 precum și un stagiu de
pediatrie generală la University of Louisville, Kentucky,
SUA, în 1994.
Cu o activitate profesională prodigioasă, a fost mem-
bru al Comitetului Director al Societății Române de Pedi-
atrie și șef de secție la Spitalul Maria Sklodowska Curie
din București.
A început să practice ecografia încă de la începutul
anilor ’80. După înființarea Societății Române de Ultra-
sonografie în Medicină și Biologie (SRUMB) a fost
membru activ al societății, cu numeroase participări la
conferințele și congresele naționale din țară și străinătate.
Autor și coautor al unui număr mare de articole
științifice, și al unor tratate de specialitate, DOCTORUL
Mircea Leonid Stamate a fost unanim apreciat de toți cei
care l-au cunoscut, pentru calitățile sale umane și profe-
sionale.
L-am cunoscut pe Mircea Stamate la începutul anilor
’90 la conferințele și congresele SRUMB din țară și am
participat împreună la congrese în străinătate. Am fost
cucerit de simțul umorului, de voioșia și pofta de viață
dar și de profesionalismul desăvârșit de care dădea do-
vada. Apoi l-am rugat în țară să participe mai mulți ani
la rând în Comisia de examen pentru obținerea atesta-
tului de studii în Ecografie generală, organizată la Spita-
lul Clinic Fundeni. Întotdeauna a răspuns cu entuziasm
și bunăvoință.
Vocea caldă, calmul și blândețea care l-au caracteri-
zat, sfaturile bune pe care le dădea candidaților la exame-
nele de competență, au contribuit la formarea mai multor
generații de specialiști în ecografie și la tot ce a însemnat
pași înainte în ecografia din România.
Prin plecarea sa dintre noi, mult prea repede, DOCTO-
RUL Mircea Leonid Stamate, lasă un gol mare în inimile
noastre, cei care l-am cunoscut, iar medicina românească
suferă o mare pierdere.
Dumnezeu să-l odihnească.
Dan A. Stănescu
Institutul Clinic Fundeni, București
 Guidelines for Authors
Medical Ultrasonography is the official publication of the
Romanian Society for Ultrasonography in Medicine and Biol-
ogy (SRUMB).
The journal aims to promote ultrasound diagnosis by
publishing papers that deal with fundamental and clinical
research, scientific reviews, clinical case reports, progress in
ultrasound physics or in the field of medical technology and
equipment, as well as educational papers, special reports and
letters to the editor.
The journal is published quarterly and papers are accepted
for publication in English language.
The official title abbreviation is Med Ultrason. It should
be used in citations, footnotes, legends for figures and all bib-
liographic references.
1. Copyright
Submitting a scientific paper to Medical Ultrasonography
for publishing is subject to compliance with the following
statements:
• the paper is original and has not been published in other
journals or books;
• the paper has not been sent or is not under consideration for
publication elsewhere;
• all authors agree upon publication of the paper.
The statements can be downloaded from the journal web-
site.
All these statements should be included in a formal dec-
laration signed by all the authors. The filled in, signed and
scanned images of these two forms have to be uploaded at the
last step of the submission process (Step 4. Uploading Sup-
plementary Files).
In cases where the paper is accepted for publication, copy-
right will be transferred to Medical Ultrasonography and the
“Iuliu Hatieganu” Medical Publishing House. Authors must
agree to undertake all responsibility for the scientific content
and originality of the paper; Medical Ultrasonography will
take no responsibility whatsoever in this respect.
2. Preparing the manuscript
2.1 General
a. Style & language
Manuscripts should be prepared according to the style of
the journal. They should be written in concise and grammati-
cally correct US English. Authors should ask for assistance if
not writing in their native language.
Papers not conforming to the style of the journal, incom-
prehensible, written in inappropriate English, or not submit-
ted strictly according to the journal guidelines will be re-
turned to the authors for revision, without peer reviewing.
If the manuscript is not complying to accepted standards
of English usage, the authors may be required to bear the cost
of English supervision/ translation.
b. Text and page layout
The papers submitted for publication must be write in
Microsoft Word program, in 12p Times New Roman font, 1.5
line spacing. Page size should be A4, margins must be normal
(2.54 mm in top, bottom, right and left). Page numbering must
begin with the title page. No numbering of the lines is needed.
c. Text length, abbreviations and measurement units
The length of the manuscripts: maximum 5000 words
for original paper, 6000 for reviews, 2000 for pictorial essay,
1500 for case report, 500 for letter to editor (including the
abstract, keywords, references, legend and tables). For letter
to editor only one imagine or table is accepted.
The editors reserve the right of condensing any paper sub-
mitted.
Use only standard abbreviations. Avoid abbreviations in
the title and abstract. Explain every abbreviation before to use
it.
Measurements of length, height, weight, and volume
should be reported in metric units. All hematological and clin-
ical chemistry measurements should be reported in the metric
system in terms of the International System of Units (SI).
d. Images and tables
Imagines (figures) and tables should be grouped in a dis-
tinct section. They must be numbered according to the order in
which they appear in the text- the imagines (figures) in Arabic
numeral (ex. fig 1) and the tables in Roman numeral (ex. tab I).
Do not send them in PDF format!
The captions for figures (images) must be typed on a sepa-
rate page entitled “Legend for figures”. Each table must have
a title. Images or tables should not appear in the text; the de-
sired location for insertion should be indicated by means of a
paragraph, such as: (location for figure no….) or (location for
table no…)
Only high quality images will be accepted for publica-
tion. File formats: BMP or TIFF for images, 300 dpi, image
width 8 cm (single column) or 16.7 cm (double column), for
color images color mode should be CMYK.
The tables (design: Table Grid) are accepted as word doc-
uments included at the end of the main document (see below).
In cases where reproduction of previously published im-
ages is intended, it is necessary to attach the written consent of
the author and of the publishing house where it was formerly
published. All prospective or experimental papers involving
human subjects or experimental animals must include the
agreement granted by the medical ethics commission of the
institution where the research was conducted.
If the manuscript reports medical research involving hu-
man subjects, authors must include a statement confirming
that informed consent was obtained from all subjects, accord-
ing to the World Medical Association Declaration of Helsinki,
revised in 2000, Edinburgh.
3. Structure of the submission files
The manuscript has to be submitted in this form:
 a) in step 2 of submission process, the word document (the
original file) that is uploaded has to include the following: title
of the manuscript, abstract and keywords, main manuscript,
references, tables, legend. Do not insert here the name of the
authors and affiliation in order to ensure a blind review.
b) in step 4 of submission process the supplementary files
that must be uploaded are: title page, images, submission let-
ter, declaration of conflict of interest.
Title page  includes: title of the paper, full names of the
authors, department and institution(s) where the study was
conducted, postal code, city, district, phone and/or fax number
and/or e-mail address for contacting the first author and cor-
responding author, full postal address for correspondence and
ordering reprints.
Abstract  (on a separate page) preceding the text body.
In the case of original papers, abstracts should not exceed
250 words and should have the following structure: 1) aims;
2) material and methods; 3) results; 4) conclusions. Abstracts
for literature reviews and educational papers should not
exceed 200 words. For case reports, the abstract must not
exceed 120 words and must underline the following: 1) pur-
pose of the presentation; 2) peculiarities of the case; ranking
of the issues approached within the general knowledge of the
respective condition.
Three to five keywords must be selected for every paper
from the Index Medicus (http://www.ncbi.nlm.nih.gov/sites/
entrez?db=mesh); the key words should be inserted after the
abstract and separated by semi-colon (term1; term2; term3).
The main document has to be structured as follows:
Introduction  – should define the topic of the paper and
present the status of current knowledge in the field.
Material and methods  – should describe the equipment
employed, the group of patients studied and the methodology.
We recommend specification of the type of ultrasound
equipment employed. The statistical analysis methodology
used must also be described.
Results  – should present the obtained data, in a concise
manner, preferably in tables and diagrams.
Discussions – should present the interpretation of the re-
searcher’s results from the perspective of relevant literature
data.
Conclusions of the paper must be clearly stated in the end.
Acknowledgements – should be made only to those who
have made a substantial contribution to the study
References must include only papers that are quoted in the
text and that have been published. References must be num-
bered in Arabic numerals in the order in which they appear
in the text (where they should be inserted between square
brackets [ ]) and listed in numerical order. Titles of medical
journals must be abbreviated according to the Index Medicus.
All authors must be quoted for an article, if they are up to
six. Over seven authors, only the first three will be quoted,
followed by the “et al” indication. References should be listed
according to the following format (examples):
a) Article:
• Marks WM, Filly RA, Callen PW. Real-time evaluation of
pleural lesions: new observations regarding the probability
of obtaining free fluid. Radiology 1982;142:163-164.
b) Papers published only with DOI numbers:
• Guerriero S, Alcazar JL, Pascual MA, Ajossa S, Olartecoe-
chea B, Hereter L. The pre-operative diagnosis of metastatic
ovarian tumors is related to the origin of the primary tumor.
Ultrasound Obstet Gynecol 2011, doi: 10.1002/uog.10120.
c) Book:
• Talano JV, Gardin JM. Textbook of two dimensional echo-
cardiography. London: Gruene & Stratton, 1983.
d) Book chapter:
• Brooks M. The Liver. In: Goldberg BB, Pettersson H (eds).
Ultrasonography. Oslo, The Nicer Year Book 1996:55-82.
Tables should be added at the end of the main document,
in the same word document.
4. Editorial policy
Medical Ultrasonography promotes evaluation of all the
scientific papers by independent reviewers. The editor-in-
chief or one of the editors evaluates each manuscript and, in
1-3 weeks from reception, decides upon their priority level
(sent to review, rejected without being sent for review or re-
turned to authors with suggestions for improvement before
submitting to review). The editors reserve the right to request
any changes they may consider appropriate, in the title, struc-
ture or body of the paper.
The papers are submitted to two blind reviewers with
expertise in ultrasonography. Based on the reviewers’ recom-
mendations, the editors decide whether a paper is published
or not. In case of marked discrepancy between the decisions
of the two reviewers, the editor may send the manuscript to
another arbitrator for additional comments and a recommend-
ed decision. The full decisional process may last 6-8 weeks.
Failure of the authors to comply with the editorial revision
requests may induce publication rejection.
A 150 euro processing fee (100 euro for the first author if
SRUMB member) is charged for each accepted paper starting
from 1st March 2021. The fee must be paid ahead of publication,
at the time when the authors give the authorization for publi-
cation of the final proof. Instructions for online payment are
available on the journal site. No free reprints of the published
paper are supplied. No additional reprints may be ordered.
5. Submitting manuscripts for publication
To submit a manuscript for publication, one author
should register on our website, and follow the five steps of the
submission process. The journal does not accept email or
CD submissions of articles. The editorial office will send a
confirmation e-mail to the correspondence address.
Remark: Medical Ultrasonography cannot be held respon-
sible for losing or damaging the files delivered through the
Internet.