Professional Documents
Culture Documents
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
family of
ultrasound systems
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 Med Ultrason 2022, Vol. 24, no. 1, 5-6
DOI: 10.11152/mu-3616
Department Allgemeine Innere Medizin (DAIM), Kliniken Hirslanden, Beau Site, Salem und Permanence, Bern,
Switzerland
The World Federation for Ultrasound in Medicine been described in Med Ultrason [5,7]. Contrast-enhanced
and Biology (WFUMB) is dedicated to the advancement ultrasound (CEUS) ([11] and also to a lesser degree elas-
of ultrasound by encouraging research, promoting inter- tography [12] have expanded the roles of LUS. The chal-
national cooperation, disseminating scientific informa- lenges of CEUS in pediatric patients have been recently
tion and improving communication and understanding reviewed [13-15].
in the world community using ultrasound in medicine The lack of superficial adipose and bone tissue pro-
and biology. One mission of WFUMB is to bring sus- vides favorable acoustic windows in children compared
tainable ultrasound programs to the underserved areas of to the more often more difficult situation in adults due to
the world to improve global healthcare through collabo- the presence of a bony thorax makes ultrasound the first
ration, communication and education (www.wfumb.org). line of investigation for evaluation of pleural and chest
Medical Ultrasonography (formerly Revista Româna de wall abnormalities in pediatric patients. In a meta-anal-
Ultrasonografie from 1999 to 2008) is the official publi- ysis consisting of 1510 children, chest ultrasound shows
cation of the Romanian Society for Ultrasonography in significantly better sensitivity and similar specificity in
Medicine and Biology (SRUMB), published in English, detecting pneumonia in children compared to chest radi-
quarterly. SRUMB and Med Ultrason are active members ography [16]. LUS has been proposed as a reasonable al-
in WFUMB and the European Federation of Societies for ternative first-line investigation for diagnosing suspected
Ultrasound in Medicine and Biology (EFSUMB) [1]. community-acquired pneumonia [17].
In the current WFUMB paper series, three consecu- The first article covers the technical requirements, ex-
tive papers are published in Med Ultrason describing the amination technique, normal sonographic appearance of
examination technique, applications and practical use of the pleural space, identifying and determining the type
chest, diaphragmatic, pleura, mediastinal and lung ultra- and volume of pleural effusion, pleuritis and diffuse pleu-
sound in children with congenital and acquired diseases. ral thickening, solid pleural lesions including benign and
The advantages of ultrasound in pediatric patients are ob- malignant pleural tumors and fibrogenic changes of the
vious: its high spatial and temporal resolution, real-time pleura (fibrothorax) [2]. The most important pathologies
imaging, and lack of ionizing radiation and bedside avail- of the pleural cavity in pediatric patients are pleural effu-
ability at the point of care [2-4]. In addition, new ultra- sion and pneumothorax [2].
sound technologies have been introduced into chest and In the second paper the use of ultrasound in the lung
lung ultrasound applications including higher resolution in pediatric patients is described including the intersti-
transducers and harmonic imaging allowing direct visu- tial syndrome, bacterial pneumonia and viral infections,
alization of structures being less dependent on artifacts COViD findings, atelectasis, lung consolidation, bron-
[5-8]. The use and controversies of lung artefacts have chiolitis and congenital diseases of the respiratory sys-
tem including congenital pulmonary airway malforma-
Received Accepted
tion (CPAM) and sequester [3]. Severe acute respiratory
Med Ultrason
2022, Vol. 24, No 1, 5-6 syndrome (SARS) coronavirus-2 disease (COViD) has
Corresponding author: Prof. Dr. med. Christoph F. Dietrich, MBA been a challenge since 2019. Lung ultrasound allows the
Department of Internal Medicine (DAIM) identification of typical sonographic signs in the course
Kliniken Hirslanden Bern, Beau Site,
Salem and Permanence
of COVID-19 infections including the triad of ultrasound
Schänzlihalde 11, 3031 Bern, Switzerland features of the pleura and pleural space represented by
E-mail: c.f.dietrich@googlemail.com irregularities, findings of interstitial pneumonia repre-
6 Christoph F Dietrich Lung ultrasound for ever
sented by subpleural consolidation and B-line artefacts. 10. Safai Zadeh E, Görg C, Dietrich CF, Görlach J, Alhyari
In addition, contrast enhanced ultrasound depicts typical A, Trenker C. Contrast-enhanced ultrasound for evalua-
perfusion defects in COViD-patients compared to other tion of pleural effusion: a pictorial essay. J Ultrasound Med
2022;41:485-503.
entities [9,11]. Extrapulmonary manifestations can be
11. Safai Zadeh E, Westhoff CC, Keber CU, et al. Perfusion
also detected and characterized using ultrasound [18].
Patterns of Peripheral Organizing Pneumonia (POP) Using
In the third article, the use of ultrasound for chest Contrast-Enhanced Ultrasound (CEUS) and Their Correla-
wall, mediastinum, diaphragmatic diseases, trachea, in- tion with Immunohistochemically Detected Vascularization
terventions and artifacts in pediatric patients are summa- Patterns. Diagnostics (Basel) 2021;11:1601.
rized [4]. 12. Dietrich CF, Ferraioli G, Sirli R, et al. General advice in
In conclusion, the change in attitude and growing ultrasound based elastography of pediatric patients. Med
awareness of the diagnostic possibilities has led to lung Ultrason 2019;21:315-326.
ultrasound being accepted as a valuable point of care 13. Dietrich CF, Averkiou M, Nielsen MB, et al. How to per-
method and is supported by international societies in- form Contrast-Enhanced Ultrasound (CEUS). Ultrasound
cluding EFSUMB and WFUMB [19-22]. Int Open 2018;4:E2-E15.
14. Sidhu PS, Cantisani V, Deganello A, et al. Role of Con-
References trast-Enhanced Ultrasound (CEUS) in Paediatric Prac-
tice: An EFSUMB Position Statement. Ultraschall Med
1. Piscaglia F, Stefanini F, Cantisani V, et al. Benefits, Open 2017;38:33-43.
questions and Challenges of the use of Ultrasound in the 15. Dietrich CF, Augustiniene R, Batko T, et al. European Fed-
COVID-19 pandemic era. The views of a panel of world- eration of Societies for Ultrasound in Medicine and Biology
wide international experts. Ultraschall Med 2020;41:228- (EFSUMB): An Update on the Pediatric CEUS Registry on
236. Behalf of the “EFSUMB Pediatric CEUS Registry Working
2. Jaworska J, Buda N, Ciuca IM, et al. Ultrasound of the Group”. Ultraschall Med 2021;42:270-277.
pleura in children, WFUMB review paper. Med Ultrason 16. Balk DS, Lee C, Schafer J, et al. Lung ultrasound compared
2021;23:339-347. to chest X-ray for diagnosis of pediatric pneumonia: A me-
3. Dietrich CF, Buda N, Ciuca IM, et al. Lung ultrasound in ta-analysis. Pediatr Pulmonol 2018;53:1130-1139.
children, WFUMB review paper (part 2). Med Ultrason 17. Stadler JAM, Andronikou S, Zar HJ. Lung ultrasound for
2021;23:443-452. the diagnosis of community-acquired pneumonia in chil-
4. Fang C, Jaworska J, Buda N, et al. Ultrasound of the chest dren. Pediatr Radiol 2017;47:1412-1419.
and mediastinum in children, interventions and artefacts. 18. Dehmani S, Penkalla N, Jung EM, et al. Scoping Review:
WFUMB review paper (part 3). Med Ultrason 2022;24: Sonographic evidence of intraabdominal manifestations of
65-76. COVID-19. Med Ultrason 2022. doi:10.11152/mu-3538
5. Mathis G, Horn R, Morf S, et al. WFUMB position paper 19. Dietrich CF, Mathis G, Cui XW, Ignee A, Hocke M, Hirche
on reverberation artefacts in lung ultrasound: B-lines or TO. Ultrasound of the pleurae and lungs. Ultrasound Med
comet-tails? Med Ultrason 2021;23:70-73. Biol 2015;41:351-365.
6. Yue Lee FC, Jenssen C, Dietrich CF. A common misunder- 20. Dietrich CF, Goudie A, Chiorean L, et al. Point of Care Ul-
standing in lung ultrasound: the comet tail artefact. Med trasound: A WFUMB Position Paper. Ultrasound Med Biol
Ultrason 2018;20:379-384. 2017;43:49-58.
7. Dietrich CF, Mathis G, Blaivas M, et al. Lung artefacts and 21. Soldati G, Smargiassi A, Inchingolo R, et al. Proposal for
their use. Med Ultrason 2016;18:488-499. International Standardization of the Use of Lung Ultra-
8. Dietrich CF, Mathis G, Blaivas M, et al. Lung B-line arte- sound for Patients With COVID-19: A Simple, Quantitative,
facts and their use. J Thorac Dis 2016;8:1356-1365. Reproducible Method. J Ultrasound Med 2020;39:1413-
9. Safai Zadeh E, Beutel B, Dietrich CF, et al. Perfusion Pat- 1419.
terns of Peripheral Pulmonary Lesions in COVID-19 Pa- 22. Volpicelli G, Elbarbary M, Blaivas M, et al. International
tients Using Contrast-Enhanced Ultrasound (CEUS): A evidence-based recommendations for point-of-care lung ul-
Case Series. J Ultrasound Med 2021;40:2403-2411. trasound. Intensive Care Med 2012;38:577-591.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 7-13
DOI: 10.11152/mu-3238
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
is associated with even higher exposure to ionizing radia- study and deemed this study exempt as per current fed-
tion and higher health care costs [14,16-22]. Critically ill eral guidelines.
and hemodynamically unstable patients may especially In preparation of scanning, all cadavers were posi-
benefit from a portable imaging modality, such as point- tioned in reverse Trendelenburg at 30 degrees for ten
of-care ultrasound (PoCUS) performed by trained emer- minutes and pre-scanned by two trained separate EPs, to
gency physicians (EPs), to expedite diagnosis and man- rule out the presence of any artifacts or irregularities at
agement prior to definitive imaging. the peritoneal area including sonographic artifacts con-
Prior investigations by non-EP imaging experts have sistent with pneumoperitoneum. All participating emer-
shown that ultrasound (US) can detect small quantities gency physicians had prior extensive US experience in
of intraperitoneal air. As early as 1982, Seitz and Reising PoCUS including pneumoperitoneum diagnosis.
reported in a proof of concept study that the accuracy of The two coordinators injected 15 fresh human ca-
US is compatible with conventional chest x-ray imaging, davers with various predefined quantities of free intra-
the gold standard imaging test in 1982 for free air. These peritoneal air, ranging from 0-10 mL. Each cadaver was
highly trained experts detected as little as 1 mL of free in- injected 3 times throughout the study and scanned three
traperitoneal air with 100% accuracy [23]. In their large- times. The injection/scanning order was based upon a
scale follow up study of about 4,000 consecutive patients pre-designed randomized order and matrix (Table I, fig
presenting to a single ED with non-traumatic acute ab- 1). Figure 2 illustrates a positive and negative US finding
dominal pain, the same authors demonstrated 90% sensi- for free intraperitoneal air. The matrix specified the quan-
tivity and 100% specificity of US for pneumoperitoneum tity of air to be injected, the location of injection and the
[23]. Smaller subsequent studies performed by highly order of the cadavers to be scanned. A cadaver was rand-
trained physicians further supported these initial find- omized to either the mid-epigastric (ME) injection (1 cm
ings, including in blunt trauma and non-trauma patients. cephalad to the umbilicus) or left lower quadrant (LLQ)
All showed high accuracy and superior sensitivity when abdominal injection, which was performed exactly equi-
compared to plain radiography [24-28]. However, given distant between umbilicus and left anterior superior iliac
the trend of increasing use of PoCUS in emergency medi- spine (ASIS). Once a cadaver was randomized to a spe-
cine education, there is little knowledge about whether cific injection site, all three injections for this cadaver
EPs who train in PoCUS can reliably identify pneumop- were administered at the same anatomical location and
eritoneum [29,30]. It would additionally be of interest to resulted in cumulative amounts of free air.
learn if a cadaver model with induced pneumoperitone- Of the 15 cadavers, eight were injected x3 in the
um can serve as a reliable teaching model for emergency midepigastric area, seven x3 into the left lower quadrant
physicians learning to diagnose pneumoperitoneum with as described above. This created 45 different scanning
PoCUS. scenarios: 24 for cadavers with ME free air and 21 for
If EPs trained in PoCUS can accurately detect free LLQ free air (fig 3). As each cadaver was scanned by
intraperitoneal air, especially small amounts of free in- both sonographer A and B, this resulted in a total of 48
traperitoneal air, management of critically ill patients encounters for ME cadavers and 42 observations in LLQ
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.
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. Fig 1. Workflow of air injection and randomized scanning of
The Institutional Review Board reviewed this research 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
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.
Results
Despite the creation of 45 blinded encounters, this 9. MacKersie AB, Lane MJ, Gerhardt RT, et al. Nontrau-
was a small study limited by the number of cadavers matic acute abdominal pain: unenhanced helical CT com-
available and needs to be repeated on a larger scale, ide- pared with three-view acute abdominal series. Radiology
2005;237:114-122.
ally on live patients.
10. Butler J, Martin B. Towards evidence based emergency
medicine: best BETs from the Manchester Royal Infirmary.
In conclusion, this study demonstrated that two Detection of pneumoperitoneum on erect chest radiograph.
trained emergency physicians detected small amounts Emerg Med J 2002;19:46-47.
of free air ≥2 mL reliably when injected into the ME in 11. Gupta H, Dupuy DE. Advances in imaging of the acute ab-
this human cadaver model. PoCUS appeared feasible for domen. Surg Clin North Am 1997;77:1245-1263.
detecting small amounts of pneumoperitoneum in this 12. Billittier AJ, Abrams BJ, Brunetto A. Radiographic imag-
model with this approach, and, if clinically validated, ing modalities for the patient in the emergency department
could expedite the diagnosis and management of patients with abdominal complaints. Emerg Med Clin North Am
suspected of having pneumoperitoneum in the acute care 1996;14:789-850.
setting. 13. Ahn SH, Mayo-Smith WW, Murphy BL, Reinert SE, Cro-
nan JJ. Acute nontraumatic abdominal pain in adult pa-
A future prospective study looking at PoCUS detec-
tients: abdominal radiography compared with CT evalua-
tion of pneumoperitoneum among live human beings tion. Radiology 2002;225:159-164.
would be needed to improve external validity and pro- 14. Stapakis JC, Thickman D. Diagnosis of pneumoperitone-
vide clinical value. um: abdominal CT vs. upright chest film. J Comput Assist
Tomogr 1992;16:713-716.
Acknowledgements: We would like to thank 15. Greene CS. Indications for plain abdominal radiography in the
Dr. Richard Levitan and Mr. Ronn Wade from the Uni- emergency department. Ann Emerg Med 1986;15:257-260.
versity of Maryland, Baltimore Anatomical Services Di- 16. National Research Council. Health risks from exposure to
vision for permitting us to utilize their human cadavers low levels of ionizing radiation: BEIR VII phase 2. Nation-
and cadaver lab facilities. We would also like to thank al Academies Press 2006. Available from: http://www.nap.
edu/catalog/11340.html
Dr. Ilene Staff for her assistance with the statistical analy-
17. Preston DL, Ron E, Tokuoka S, et al. Solid cancer inci-
sis of the data collected. Funding for this study was from
dence in atomic bomb survivors: 1958-1998. Radiat Res
the Hartford Hospital New Investigator Grant. 2007;168:1-64.
18. Pearce MS, Salotti JA, Little MP, et al. Radiation exposure
Rerefences from CT scans in childhood and subsequent risk of leukae-
mia and brain tumours: a retrospective cohort study. Lancet
1. Langell JT, Mulvihill SJ. Gastrointestinal perforation and 2012;380:499-505.
the acute abdomen. Med Clin North Am 2008;92:599-625. 19. Mathews JD, Forsythe AV, Brady Z, et al. Cancer risk in
2. McCaig LF, Burt CW. National Hospital Ambulatory Medi- 680,000 people exposed to computed tomography scans in
cal Care Survey: 2002 emergency department summary. childhood or adolescence: data linkage study of 11 million
Adv Data 2004;(340):1-34. Australians. BMJ 2013;346:f2360.
3. Court-Brown CM, McQueen MM, Patterson-Brown S, 20. Lumbreras B, Donat L, Hernandez-Aguado I. Incidental
Nixon SJ. Emergency surgical care in Scotland. Surgeon findings in imaging diagnostic tests: a systematic review.
2007;5:72-75. Br J Radiol 2010;83:276-289.
4. Schietroma M, Cappelli S, Carlei F, Pescosolido A, Lygida- 21. Thompson RJ, Wojcik SM, Grant WD, Ko PY. Inciden-
kis NJ, Amicucci G. “Acute abdomen”: early laparoscopy tal Findings on CT Scans in the Emergency Department.
or active laparotomic-laparoscopic observation? Hepato- Emerg Med Int 2011;2011:624847.
gastroenterology 2007;54:1137-1141. 22. Medicare Payment Advisory Commission MPA. Report to
5. Ordonez CA, Puyana JC. Management of peritonitis in the the Congress: improving incentives in Medicare. Available
critically ill patient. Surg Clin North Am 2006;86:1323- from: http://www.medpac.gov/docs/default-source/reports/
1349. Jun09_EntireReport.pdf Accessed 2016.
6. Wittmann DH, Schein M, Condon RE. Management of sec- 23. Seitz K, Reising KD. Ultrasound detection of free air in the
ondary peritonitis. Ann Surg 1996;224:10-18. abdominal cavity. Ultraschall Med 1982;3:4-6.
7. Feldman M, Friedman LS, Sleisenger MH. (Eds.). Sleisen- 24. Chang-Chien CS, Lin HH, Yen CL, Lee CM, Lin SM. So-
ger and Fordtran’s gastrointestinal and liver disease: Patho- nographic demonstration of free air in perforated peptic ul-
physiology, diagnosis, management. 7th ed. Philadelphia: cers: comparison of sonography with radiography. J Clin
WB Saunders, 2004. Ultrasound 1989;17:95-100.
8. Stoker J, van Randen A, Lameris W, Boermeester MA. 25. Chen SC, Wang HP, Chen WJ, et al. Selective use of ultra-
Imaging patients with acute abdominal pain. Radiology sonography for the detection of pneumoperitoneum. Acad
2009;253:31-46. Emerg Med 2002;9:643-645.
Med Ultrason 2022; 24(1): 7-13 13
26. Muradali D, Wilson S, Burns PN, Shapiro H, Hope-Simp- moperitoneum via point-of-care ultrasound. Ultrasound J
son D. A specific sign of pneumoperitoneum on sonog- 2020;12:52.
raphy: enhancement of the peritoneal stripe. AJR Am J 30. Nazerian P, Tozzetti C, Vanni S, et al. Accuracy of abdomi-
Roentgenol 1999;173:1257-1262. nal ultrasound for the diagnosis of pneumoperitoneum in
27. Butcher CH, Dooley RW, Levitov AB. Detection of sub- patients with acute abdominal pain: a pilot study. Crit Ul-
cutaneous and intramuscular air with sonography: a sensi- trasound J 2015;7:15.
tive and specific modality. J Ultrasound Med 2011;30:791- 31. Hoffmann B, Nürnberg D, Westergaard MC. Focus on ab-
795. normal air: diagnostic ultrasonography for the acute abdo-
28. Moriwaki Y, Sugiyama M, Toyoda H, et al. Ultra- men. Eur J Emerg Med 2012;19:284-291.
sonography for the diagnosis of intraperitoneal free 32. Lee DH, Lim JH, Ko YT, Yoon Y. Sonographic detection
air in chest-abdominal-pelvic blunt trauma and criti- of pneumoperitoneum in patients with acute abdomen. AJR
cal acute abdominal pain. Arch Surg 2009;144:137- Am J Roentgenol 1990;154:107-109.
141. 33. Karahan OI, Kurt A, Baykara M, Coskun A. Detectability
29. Taylor MA, Merritt CH, Riddle PJ Jr, DeGennaro CJ, Bar- of intraperitoneal free air by ultrasonography. Tani Girisim
ron KR. Diagnosis at gut point: rapid identification of pneu- Radyol 2003;9:60-62.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 14-18
DOI: 10.11152/mu-3163
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
with fetus with type IIIa and type IIIb (30.8±5.1 years vs. features of the case 17 that had type IIIb combined with
32.6±7.9 years, p=0.65). Ultrasonographic findings of type IIIa are presented in figure 2.
the shunt in each case are presented in Table I. Several associated anomalies were detected, the
All the 12 type IIIa cases had a shunt between the most-common one being intrauterine growth restriction
inferior left portal vein (LPVi) and the left hepatic vein (IUGR) followed by polyhydramnios and increased car-
(LHV) and a representative sonography showing such a diothoracic ratio. There was no significant difference in
shunt is presented in figure 1 (case 1). Three type IIIb the incidence of associated anomalies between type IIIa
cases had a combination of another shunt (2 with type I and type IIIb patients (p=0.82). Prenatal karyotyping was
and one with type IIIa) (Table I). The ultrasonographic 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 5. Francois B, Gottrand F, Lachaux A, Boyer C, Benoit B, De
vein systems that are in line with the definition of the type Smet S. Outcome of intrahepatic portosystemic shunt diag-
IIIa shunt under the Achiron and Kivilevitch classifica- nosed prenatally. Eur J Pediatr 2017;176:1613-1618.
6. Han BH, Park SB, Song MJ, et al. Congenital portosystem-
tion. Their follow up results showed 5 cases had spon-
ic shunts: prenatal manifestations with postnatal confirma-
taneous closure at approximately 1 year after birth [6].
tion and follow-up. J Ultrasound Med 2013;32:45-52.
7. Delle Chiaie L, Neuberger P, Von Kalle T. Congenital in-
Conclusion trahepatic portosystemic shunt: prenatal diagnosis and pos-
sible influence on fetal growth. Ultrasound Obstet Gynecol
This is the largest series of Type III shunt prenatally 2008;32:233-235.
analyzed by ultrasonography using the new in-utero UP- 8. Gorincour G, Droulle P, Guibaud L. Prenatal diagnosis of
SVS classification. We conclude that the majority of type umbilicoportosystemic shunts: report of 11 cases and review
III UPSVS cases belong to type IIIa with good outcome of the literature. AJR Am J Roentgenol 2005;184:163-168.
and the lack of integrate intrahepatic portal vein system 9. Francois B, Lachaux A, Gottrand F, De Smet S. Prenatally
diagnosed congenital portosystemic shunts. J Matern Fetal
is the main reason for TOP in type IIIb cases. These data
Neonatal Med 2018;31:1364-1368.
suggest the UPSVS classification is a useful tool for prog-
10. Yagel S, Kivilevitch Z, Cohen SM, et al. The fetal venous
nosis prediction of fetal PSS and prenatal counseling. system, part I: normal embryology, anatomy, hemodynam-
ics, ultrasound evaluation and Doppler investigation. Ultra-
Acknowledgements: This study was supported by sound Obstet Gynecol 2010;35:741-750.
the Provincial Key Research and Development Fund of 11. Achiron R, Kivilevitch Z. Fetal umbilical-portal-systemic
Shandong Province, China (Grant #: 2015GSF118081 venous shunt: in-utero classification and clinical signifi-
and 2016GSF201141). cance. Ultrasound Obstet Gynecol 2016;47:739-747.
12. Wu H, Tao G, Cong X, et al. Prenatal sonographic char-
Conflict of interest: none. acteristics and postnatal outcomes of umbilical-portal-sys-
temic venous shunts under the new in-utero classification: A
References retrospective study. Medicine (Baltimore) 2019;98:e14125.
13. Blanc T, Guerin F, Franchi-Abella S, et al. Congenital
1. Sokollik C, Bandsma RH, Gana JC, van den Heuvel M, portosystemic shunts in children: a new anatomical clas-
Ling SC. Congenital portosystemic shunt: characteriza- sification correlated with surgical strategy. Ann Surg
tion of a multisystem disease. J Pediatr Gastroenterol Nutr 2014;260:188-198.
2013;56:675-681. 14. Matsuura T, Takahashi Y, Yanagi Y, et al. Surgical strategy
2. Franchi-Abella S, Branchereau S, Lambert V, et al. Com- according to the anatomical types of congenital portosys-
plications of congenital portosystemic shunts in children: temic shunts in children. J Pediatr Surg 2016;51:2099-2104.
therapeutic options and outcomes. J Pediatr Gastroenterol 15. Yagel S, Cohen SM, Valsky DV, Shen O, Lipschuetz M,
Nutr 2010;51:322-330. Messing B. Systematic examination of the fetal abdominal
3. Chocarro G, Amesty MV, Encinas JL, et al. Congenital precordial veins: a cohort study. Ultrasound Obstet Gy-
portosystemic shunts: clinic heterogeneity requires an in- necol 2015;45:578-583.
dividual management of the patient. Eur J Pediatr Surg 16. Park JH, Cha SH, Han JK, Han MC. Intrahepatic portosys-
2016;26:74-80. temic venous shunt. AJR Am J Roentgenol 1990;155:527-
4. Yagel S, Kivilevitch Z, Cohen SM, et al. The fetal venous 528.
system, Part II: ultrasound evaluation of the fetus with con- 17. Dong X, Wu H, Zhu L, et al. Prenatal ultrasound analy-
genital venous system malformation or developing circula- sis of umbilical-portal-systemic venous shunts concurrent
tory compromise. Ultrasound Obstet Gynecol 2010;36:93- with trisomy 21. J Ultrasound Med 2020. doi:10.1002/
111. jum.15507.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 19-26
DOI: 10.11152/mu-3124
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
sis may be challenging. Fever and cough in presenting peutic procedures. The following outcomes were pur-
pictures suggest pneumonia, but non-specific symptoms, sued: needing continuous thoracic drainage, requiring
such as abdominal pain and nuchal rigidity, may create surgical interventions consisting of either segmentecto-
significant difficulties [1]. The etiological diagnosis is my or video-assisted thoracoscopic surgery decortication
difficult and neither clinical nor radiologic features dis- and demanding prolonged hospitalization over 10 days.
tinguish between bacterial, atypical bacterial and viral
pneumonia [2,4]. Pleural effusion, empyema, necrotizing Material and methods
pneumonia, lung abscess and pneumatocele may compli-
cate CAP in children [5]. Study design
Although not recommended routinely for diagnosis in We conducted a retrospective, observational, unicen-
children, chest radiography (chest X-ray) is the most used tric study. Data were collected from electronic medical
investigation for detecting CAP lesions. The technique records of pediatric patients with the diagnosis of CAP
has certain limitations as non-severe pneumonia cases requiring hospitalization between 2017-2019 in “Marie
may show normal chest X-ray or just perihilar changes S. Curie” Emergency Children’s Hospital, Bucharest,
while minimal pleural effusion may go undetected [1,3]. Romania. All patients’ records and clinical information
Few studies have shown that radiological findings are as- were analyzed anonymously. The study was conducted
sociated with the severity of CAP in pediatric patients. with the approval of the local Ethics Committee.
Despite that, the prognostic role of chest radiography in Patients’ medical files were studied for demographic
children with CAP has not been established [6,7]. information, clinical and paraclinical data, treatment,
Thereby, the clinicians’ interest shifted towards lung clinical course, and discharge summary. The diagnosis
ultrasonography (LUS), considered in time a safe and ac- of CAP was established according to the British Tho-
cessible option for diagnosing pneumonia and its com- racic Society guidelines [1]. Clinical findings (fever,
plications [8]. Recent systematic reviews have verified cough, chest pain, respiratory distress, gastrointestinal
the high accuracy of LUS for diagnostic of pneumonia, complaints and other symptoms, respiratory rate, oxygen
concluding that it is a useful imaging alternative to chest saturation), laboratory tests (complete blood count, C-
radiography inclusive in pediatric CAP [9,10]. Even if reactive protein - CRP and procalcitonin level) and imag-
ultrasound’s usefulness is limited in evaluating a well- istic investigations (chest X-ray and LUS) at admission
aerated lung due to the imperfect transmission of the and during hospitalization were noted. All patients were
sound waves, it is an essential diagnostic tool in the pres- hospitalized and a 5th day LUS control was performed,
ence of consolidations, allowing the evaluation of lung while control blood investigations were carried out only
parenchyma [11]. LUS can detect small pulmonary le- when considered by the physician. The chest X-rays and
sions not visible on chest X-ray, suggesting that this is LUS clips or images were digitally archived in the hos-
the preferred approach for assessing patients with small pital informatic system and they were reanalyzed for this
consolidations or small parapneumonic pleural effusion study by an experienced senior pediatric radiologist.
[6,9,12-15]. Fluid or aerial bronchogram may be well de- Performance of LUS
scribed, suggesting a consolidation process. LUS is also LUS was performed soon after complicated CAP was
fairly useful in evaluating the pleural effusion, providing suspected, using the GE LogiqS8 ultrasound system,
the best detail of the fluid nature, quantity, consistency, with convex (C1-6) and linear (L3-12) probes. Anterior,
echogenicity. It can demonstrate the presence of internal lateral and posterior intercostal spaces were examined in
septations, cellular debris, honeycombing, pleural thick- longitudinal and transverse sections with patients in su-
ening, or the lack of free movements with gravity, sug- pine and sitting positions.
gesting complications that cannot be well characterized All the enrolled patients underwent first LUS at ad-
on static X-ray or computer tomography images [6,7,11]. mission and the second LUS 5 days after the beginning
Besides, LUS can help guide any invasive diagnostic or of treatment. Per center’s capabilities, the second LUS
therapeutic procedures [11]. was performed by the same person that made the first ex-
This study aimed at evaluating the LUS character- amination – this approach reduced possible interobserver
istics of complicated CAP in hospitalized children at variability. Findings were reported as normal or LUS
baseline and 5 days after beginning the antibiotic treat- features of pneumonia were recorded including: the pres-
ment and to assess the LUS value for predicting the clini- ence of pulmonary parenchymal lesions (consolidation/
cal outcome in these children. Therefore, we examined atelectasis), number, size and localization of lesions, the
which LUS findings were risk factors for a poor outcome presence of bronchogram and its characteristics (air or
in complicated CAP, requiring specific invasive thera- fluid bronchogram), the presence and aspect of pleural
Med Ultrason 2022; 24(1): 19-26 21
effusion (echogenicity, homogeneity, thickness, the pres- Severe CAP was considered according to the British
ence of septa). Thoracic Society guidelines, including persistent fever,
Two examples of comparative evolutive findings on tachypnea, respiratory distress, cyanosis, grunting respi-
the chest X-ray and LUS are presented in figure 1 and 2. rations, hypoxemia, tachycardia, prolonged capillary re-
Inclusion criteria fill time, signs of dehydration or intense positively acute
We included pediatric patients (age under 18 years phase reactants (leukocytosis, CRP, procalcitonin) [1].
old) with admission diagnosis of complicated CAP, based We defined as complicated CAP those patients presenting
on history, initial clinical examination and chest radiog- with clinical, biological and chest X-ray features sugges-
raphy, who underwent LUS at admission and 5 days after tive for local development of pleural effusion or empy-
the beginning of antibiotic therapy. 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
follow-up. However, more extensive prospective studies 12. Claes AS, Clapuyt P, Menten R, Michaux N, Dumitriu D.
are necessary to support our findings regarding the as- Performance of chest ultrasound in paediatric pneumonia.
sociation between LUS and clinical outcome in pediatric Eur J Radiol 2017;88:82-87.
13. Esposito S, Papa SS, Barzani I, et al. Performance of lung
patients with complicated CAP.
ultrasonography in children with community-acquired
pneumonia. Ital J Pediatr 2014;40:37.
Conflict of interests: none 14. Shah VP, Tunik MG, Tsung JW. Prospective Evaluation
of Point-of-Care Ultrasonography for the Diagnosis of
References
Pneumonia in Children and Young Adults. JAMA Pediatr
1. Harris M, Clark J, Coote N, et al. British Thoracic Soci- 2013;167:119-125.
ety guidelines for the management of community-acquired 15. Reali F, Sferrazza Papa GF, Carlucci P, et al. Can lung ultra-
pneumonia in children: update 2011. Thorax 2011;66:ii1- sound replace chest radiography for the diagnosis of pneu-
ii23. monia in hospitalised children? Respiration 2014;88:112-
2. Man SC, Fufezan O, Sas V, Schnell C. Performance of 115.
lung ultrasonography for the diagnosis of community-ac- 16. Musolino AM, Tomà P, Supino MC, et al. Lung ultrasound
quired pneumonia in hospitalised children. Med Ultrason features of children with complicated and noncomplicated
2017;19:276-281. community-acquired pneumonia: A prospective study.
3. Everard ML. Paediatric respiratory infections. Eur Respir Pediatr Pulmonol 2019;54:1479-1486.
Rev 2016;25:36-40. 17. Pereda MA, Chavez MA, Hooper-Miele CC, et al. Lung
4. Heiskanen-Kosma T, Korppi M, Jokinen C, et al. Etiology Ultrasound for the Diagnosis of Pneumonia in Children: A
of childhood pneumonia: serologic results of a prospective, Meta-analysis. Paediatrics 2015;135:714-722.
population-based study. Pediatr Infect Dis J 1998;17:986- 18. Ambroggio L, Sucharew H, Rattan MS, et al. Lung Ul-
991. trasonography: A Viable Alternative to Chest Radiogra-
5. Mbata G, Chukwuka C, Onyedum C, Onwubere B, Aguwa phy in Children with Suspected Pneumonia? J Pediatr
E. The Role of Complications of Community-Acquired 2016;176:93-98.e7.
Pneumonia on the Outcome of the Illness: A Prospective 19. Saraya S, El Bakry R. Ultrasound: Can it replace CT in the
Observational Study in a Tertiary Institution in Eastern Ni- evaluation of pneumonia in paediatric age group? Egypt J
geria. Ann Med Health Sci Res 2013;3:365-369. Radiol Nucl Med 2017;48:687-694.
6. Reissig A, Gramegna A, Aliberti S. The role of lung ul- 20. Kurian J, Levin TL, Han BK, Taragin BH, Weinstein
trasound in the diagnosis and follow-up of community- S. Comparison of Ultrasound and CT in the Evaluation
acquired pneumonia. Eur J Intern Med 2012;23:391-397. of Pneumonia Complicated by Parapneumonic Effu-
7. Chen IC, Hsu JH, Wu JR, Dai ZK. Updated Guidelines for sion in Children. AJR Am J Roentgenol 2009;193:1648-
Childhood Pneumonia Management: A Promising Role for 1654.
Lung Ultrasound. Pediatr Neonatol 2015;56:363-364. 21. Chen IC, Lin MY, Liu YC, et al. The role of transthoracic
8. Weinberg B, Diakoumakis EE, Kass EG, Seife B, Zvi ZB. ultrasonography in predicting the outcome of community-
The air bronchogram: sonographic demonstration. AJR Am acquired pneumonia in hospitalised children. PLoS One
J Roentgenol 1986;147:593-595. 2017;12:e0173343.
9. Heuvelings CC, Bélard S, Familusi MA, Spijker R, Gro- 22. Haggag YI, Mashhour K, Ahmed K, Samir N, Radwan
busch MP, Zar HJ. Chest ultrasound for the diagnosis of W. Effectiveness of Lung Ultrasound in Comparison with
paediatric pulmonary diseases: a systematic review and Chest X-Ray in Diagnosis of Lung Consolidation. Open
meta-analysis of diagnostic test accuracy. Br Med Bull Access Maced J Med Sci 2019;7:2457-2461.
2019;129:35-51. 23. Sperandeo M, Carnevale V, Muscarella S, et al. Clinical ap-
10. Xin H, Li J, Hu HY. Is Lung Ultrasound Useful for Diag- plication of transthoracic ultrasonography in patients with
nosing Pneumonia in Children? A Meta-Analysis and Sys- pneumonia. Eur J Clin Invest 2011;41:1-7.
tematic Review. Ultrasound Q 2018;34:3-10. 24. Byington CL, Spencer LY, Johnson TA, et al. An epidemio-
11. Light M, Blaisdell CJ, Homnick DN, Schechter MS, Wein- logical investigation of a sustained high rate of paediatric
berger MW. (Eds.). Paediatric Pulmonology. American parapneumonic empyema: risk factors and microbiological
Academy of Pediatrics, 2011. 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
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 CT, one can reduce the radiation exposure during child-
hood or the need for sedation which is necessary in time
Contrast enhanced voiding urosonography (ceVUS) consuming MRI examinations [2-4].
and contrast enhanced ultrasound (CEUS) are adjunctive CeVUS and CEUS have a diagnostic efficiency
techniques to conventional ultrasound. These techniques comparable to conventional imaging methods. Studies
are radiation free imaging modalities, which are impor- demonstrated that ceVUS has a sensitivity of 57-100 %
tant especially for children [1]. Using ceVUS and CEUS and specificity of 85-100 % in comparison to voiding
modalities as alternatives to radiography, fluoroscopy or 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-
Received 08.04.2021 Accepted 14.07.2021
Med Ultrason
sitivity and specificity in comparison to MRI or CT and is
2022, Vol. 24, No 1, 27-32 superior to fundamental B-mode sonography dependent
Corresponding author: Hans-Joachim Mentzel on the indication [7,8]. Until now there is no study evalu-
Institute of Diagnostic and Interventional ating the acceptance of parents for CEUS examinations
Radiology, University Hospital Jena
Am Klinikum 1, 07740 Jena, Germany
in their children.
E-mail: hans-joachim.mentzel@med.uni-jena.de For ceVUS the ultrasound contrast agent (UCA) is
Phone: 00493641 - 9 328 501 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® Exclusion criteria included patients aged >18 years,
(Bracco Imaging, Italy) is approved in Europe for the a known sensitivity of sulfur hexafluoride or other com-
intravesical application in children since 2017, but the ponents of SonoVue®, cardiopulmonary disorders [17],
intravenous application is only possible with off-label acute urinary tract infection for ceVUS or the lack of in-
use. Whereas, in the USA Lumason® (=SonoVue) is also formed consent of the legal guardians.
approved for intravenous use since 2016 [1]. SonoVue® Ultrasound was performed by two certified paediatric
consists of stabilized sulfur-hexafluoride microbubbles radiologists with experience in CEUS for more than ten
[10] and is well tolerated [11,12]. Intravesically applied years (each >300 ceVUS, >100 CEUS). The indication
SonoVue® is eliminated by micturition whereas intrave- for ceVUS or CEUS was made in an interdisciplinary
nously applied SonoVue® is eliminated by the lungs [10]. consultation in coordination with the legal guardians.
In consequence, SonoVue® can also be used in case of ceVUS examination
renal failure [9]. CeVUS was performed using a 9-3 MHz convex
There are few studies evaluating the risks of intra- probe on a ZS3 ultrasound machine (Mindray, China).
vesical or intravenous use of SonoVue® in children. In Baseline pre-contrast ultrasound of the urinary tract
intravesical use the minority of children shows adverse (bladder, ureter, kidneys) was conducted in supine and
events caused by catheterisation, mostly minor or moder- prone position. Afterwards, the bladder was catheterized
ate and non-serious events [6,13,14]. In intravenous use under standardized aseptic conditions using a 6 CH feed-
few paediatric cases showed adverse events, mostly mi- ing tube with two lateral eyes (B. Braun, Germany). One
nor or moderate. There were only two cases of serious urine tube was sampled for laboratory examination and
adverse events in children reported in the literature up to the bladder was emptied. Then, the catheter was linked
now [15,16]. to a three-way stopcock; one line (direct way) was con-
The aim of this study was to evaluate parents’ ac- nected to the UCA and the other to the saline solution
ceptance of contrast enhanced sonography in their chil- bag. SonoVue® was always prepared in accordance with
dren and to ask if they would prefer ceVUS to VCUG the manufacturer’s recommendations and applied in a
or CEUS to CT or MRI for a possible next examination. sterile manner. 0.1 ml of SonoVue® was applied into the
Furthermore, this study was designed to evaluate the bladder. After the administration of the UCA the bladder
safety profile of SonoVue®. was filled with prewarmed saline solution by drop infu-
sion (70 cm table height) until the estimated age-related
Materials and methods maximum bladder capacity was reached or the child
started to micturate. Under real-time ultrasound guidance
The prospective study was approved by the local in- the distribution of the UCA in the bladder was observed
ternal Ethics Review Board. and the retrovesical and proximal ureters as well as the
Patient Selection ureteropelvine junction and the kidneys were examined
Over a one-year study period, 55.4 % (56/101) of the continuously during filling and voiding. During voiding
parents of all ceVUS examinations and 54.5 % (30/55) the urethra was explored by perineal positioned probe.
of the parents of all CEUS examinations could be inter- Filling and micturition were repeated up to four times in
viewed. The other parents or legal guardians rejected the each patient (minimum two times). To minimize destruc-
participation in the survey because of time constraints or tion of the bubbles, the mechanical index was turned to a
no interest. level maximum of 0.10.
Before ceVUS and also before CEUS examinations CEUS examination
the parents were informed about the aim and asked to For UCA administration application in the cubital
participate in the study. All legal guardians were in- vein was preferred. In small infants, other positions (e.g.
formed about the off-label use of SonoVue®, proce- scalp veins) were used. Baseline pre-contrast ultrasound
dures and alternative imaging modalities for ceVUS and examination was tailored to the specific clinical query
CEUS. Alternative procedures, such as VCUG, CT and/ 9-3 MHz probe on a ZS3 ultrasound machine (Mindray,
or MRI were explained in detail including information China) or a 6-1 MHz convex abdominal probe ACUSON
about advantages and disadvantages of these methods. S2000 (Siemens Healthineers, Germany). The amount
Some parents were familiar with the alternative imag- of SonoVue® was calculated using the formula 0.1 ml x
ing modalities due to prior examinations. Their informed age in years as mentioned in the ESPR guidelines [18].
written consent was obtained prior to the examination. SonoVue® was always prepared according to the manu-
Their knowledge about the procedures as well as the ad- facturer’s recommendation and applied in a sterile man-
vantages and disadvantages were not analysed. ner in the direct way of a three‑way cock. Single UCA
Med Ultrason 2022; 24(1): 27-32 29
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 Fig 1. In a 2-month-old female baby, diagnosed with hydro-
any adverse events (AE). nephrosis during fetal ultrasonography and postnatal positive
urothel sign, the ceVUS showed high grade reflux (IV) with
Telephone survey dilated and elongated ureter
Three days after the examination the parents were
contacted for a standardized telephone survey. At first, For ceVUS 96 % (54/56) of the parents would repeat
parents were asked if they would be willing to repeat ceVUS if necessary and 4 % (2/56) of the parents refused
the examination with the same procedure (ceVUS or ceVUS for further examinations. The main reason report-
CEUS) if necessary. Secondly, they were asked if they ed for rejection was the stress children are put under due
would prefer VCUG to ceVUS for the next time or if they to catheterisation which is needed also in the alternative
would prefer CT or MRI to CEUS. Furthermore, the par- examination VCUG. Concerning CEUS examination,
ents were asked about a list of possible AE. For ceVUS 100% (30/30) of parents would agree with performing
these were skin rash, pruritus, wheals, fever, urinary tract CEUS again. ceVUS was favour to VCUG by 92.9%
infection, respiratory problems and for CEUS these were (52/56) parents. The main reported reason was in 84.6%
skin rash, pruritus, wheals, fever, respiratory problems, (44/52) of cases the lack of radiation exposure. Two par-
taste disorders, vomiting and pain. ents preferred VCUG to ceVUS because they felt VCUG
Statistical analysis is less complicated and less stressful for their child. In
Descriptive statistics was used to report the results. one case, every further examination with catheterization
was refused and in another case the parents did not give
Results any statement.
AE after ceVUS were reported by the parents in 3.6
CeVUS % (2/56) of the cases. These events were mild fever and
All ceVUS examinations were of diagnostic quality skin rash, which each occurred once and were both self-
and answered all diagnostic questions. No child needed limited.
further imaging diagnostics such as VCUG following CEUS
ceVUS. In the study, the indications for ceVUS were as- After CEUS the therapy of one child could be finished,
sessment of vesicoureteric reflux (VUR) (fig 1), bladder one child needed a change in the therapy and 86.7 %
rupture or urogenital malformation as recommended in (26/30) needed a follow-up. For 93.3% (28/30) of the
the guidelines of the ESPR/ESUR [19]. Patient charac- children CEUS was adequate. Two children needed MRI
teristics for both groups are shown in Table I. and an endoscopic retrograde cholangiopancreatography
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
Torsion of the testicular appendages is often over- sion seen on the ultrasound device. Tissue elasticity was
looked by the clinician, leading to unnecessary surgical calculated in real time and displayed as a color-coded
scrotal exploration because of erroneously suspected overlay on the B-mode. Red was encoded as soft, green
testicular torsion, thus increasing operation-related com- as intermediate and blue as hard tissue stiffness.
plications and causing additional costs for the hospital. The standard protocol of pediatric scrotal US ex-
Ultrasound (US) represents a widely available and rec- amination at our institution includes: 1) transverse
ommended diagnostic tool for the evaluation of acute gray-scale and Doppler evaluation of both testicles for
scrotal pain in all age groups [7]. Sonographic appear- initial comparison; 2) gray-scale scans of each testicle
ances on gray-scale US and color Doppler US of torsion and epididymis in both transverse and sagittal planes;
of the testicular appendages have already been presented 3) color Doppler scans of all structures; 4) derivation of
in several studies [8-13]. Doppler shift for arterial and venous intratesticular blood
The aim of this retrospective analysis was to expand flow; and 5) strain elastography of suspect findings.
those already established techniques with strain elastog- All multiparametric US examinations were per-
raphy and thus present typical appearances of torsion of formed using a Logic E9 unit with a linear probe (ML,
the testicular appendages on multiparametric US (gray- 6–15 MHz; GE Healthcare) or a HI Vision Ascen-
scale US, color Doppler US and strain elastography) to dus unit with a linear probe (EUP-L74M, 5–13 MHz;
be able to reliably confirm the diagnosis of torsion of the Hitachi).
testicular appendages in order to prevent unnecessary Torsion of testicular appendages
surgical scrotal exploration. on multiparametric US
Retrospective viewing of the patients’ images was
Material and methods performed by two radiologists (A.F., L.G.) and evaluated
in consensus. Based on previous publications [8-13] and
Patients our experience, we defined the following findings as the
This study was approved by the Institutional Review typical appearance of torsion of the testicular appendages
Board of the Medical University Innsbruck. All patients’ on multiparametric US: Gray-scale US - 1) heterogenous
data included in this retrospective study were handled ac- round lesion with close proximity to the upper pole of
cording to the norms of the Declaration of Helsinki and the testis/epididymis (fig 1a,b), 2) scrotal skin thicken-
its amendments. ing (fig 1c) and 3) concomitant hydrocele (fig 1d); Color
Retrospective analysis of all patients presented to the Doppler US - 1) avascularity of the round lesion found
urological department with an acute scrotum between on gray-scale US (fig 2a,b) and 3) reactive hyperemia
January 2018 and July 2020 identified 32 patients dis- of the associated epididymis (fig 2c); Strain Elastogra-
charged with the diagnosis of torsion of testicular ap- phy - increased tissue stiffness (encoded as blue on strain
pendages. Of these, eleven patients 6-17 years old (mean, elastography; fig 3b,d) of the round lesion found on gray
11.1 years) were examined with a high-end ultrasound scale US (fig 3a,c).
device at the radiological department of our institution. Statistical analysis
US evaluation Descriptive statistics were performed using SPSS
At our institution, US examinations in patients with Version 22 (SPSS Inc., Chicago, Illinois). Data are ex-
acute scrotal pain are performed by the Department of pressed as total numbers, mean and range.
Radiology during daytime working hours and by the
urologist on duty himself during off hours. All examina- Results
tions of our study cohort were performed at the Depart-
ment of Radiology using a high-end ultrasound device Results are shown in Table I. In all eleven cases of
and conducted or supervised by a radiologist (A.F.) with our study cohort the torsed appendix was adjacent or in
more than 10 years of experience in scrotal ultrasound. close proximity to the upper pole of the testis and to the
Standardized presets were used for gray-scale US epididymis. A differentiation between testicular/epididy-
examination. Color Doppler US examination was per- mal appendix was not possible. Seven torsions (63.3%)
formed with the highest signal gain setting possible with- occurred on the right side. A heterogenous round lesion
out the appearance of background noise to maximize sen- was observed in all cases and had a predominantly hyper-
sitivity to slow flow velocities. Strain elastography was echogenic texture in 9 (81.8%), and a predominantly hy-
performed with repeated compression and decompres- poechogenic texture in 2 (18.2%) cases. These round le-
sion of the testis, with the pressure applied to the testicles sions had a maximal diameter ranging from 4 to 11.1 mm
adjusted according to the visual indicator for compres- (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).
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
Postoperative analgesia was prescribed around the minimum sample size required to have an 80% prob-
clock in the form of paracetamol 1 g po 6 hourly and ability of detecting a 30% decrease in onset time (level
diclofenac 75 mg po 12 hourly. Oxycodone 10 mg orally of significance 0.05) was 7 patients per group. We re-
4-6 hourly was administered as rescue analgesia. Post- cruited 10 patients per group to account for potential
operatively, sensory and motor function of each nerve dropouts.
was assessed every 15 mins. Sensory and motor dura- Statistical analysis was performed using SPSS ver-
tion was measured separately for each nerve from T0 to sion 24 (IBM, Armonk, New York). The Shapiro-Wilk
return of sensation to cold and motor power to >3, re- test was used for normality testing. Continuous, normally
spectively. Overall sensory and motor block offset was distributed data are presented as mean (SD), and non-
defined as return of sensation to cold and motor power normally distributed data as median (interquartile range
(score ≥3) respectively, in any one nerve distribution [IQR]). Comparison between groups were analysed us-
area. ing the unpaired Student’s t test for normally distribut-
The primary outcome was overall surgical anaesthe- ed data and the Mann-Whitney U test for nonparamet-
sia onset time, which was defined as the time elapsed ric data. Categorical variables were compared between
from conclusion of block (T0) until attainment of sur- groups using Pearson’s or Fischer’s exact test. All tests
gical anaesthesia in all nerves distribution areas. Sec- were two-tailed, and P < 0.05 was considered statistically
ondary outcome measures included overall duration of significant.
sensory and motor block, as well as sensory and motor
onset times and durations of individual blocks. Over- Results
all duration of surgical anaesthesia was defined as time
elapsed from T0 to return of sensation and motor power Twenty patients (10 in each group) were recruited to
(score ≥3) respectively in any one nerve distribution the study from September 2014 to August 2015. All pa-
area. tients completed the study (fig 1) and none of the patients
Block performance parameters were recorded such as required rescue block, conversion to general anaesthesia
imaging time (defined as time elapsed from placement of or intraoperative opioid analgesia. There were no adverse
US probe on the patient to acquisition of a satisfactory events noted in either group. The patient demographic
image of the axillary artery and surrounding nerves) and characteristics were similar between the groups (Ta-
needling time (defined as the time interval between in- ble II). Table III details onset times. The median [IQR]
sertion and removal of block needle). Thus, performance overall onset time of surgical anaesthesia was shorter
time was defined as the sum of imaging and needling in Group 1% compared to Group 2%. The overall onset
times. The number of needle passes were recorded. The time of sensory but not motor block was also shorter in
initial needle pass was considered as the first pass and Group 1%. Onset times of individual nerves were similar
any subsequent needle advancement preceded by retrac- in the two groups, with the exception of median sensory
tion of 1 cm counted as an additional pass. Incidences of onset time which was shorter in Group 1%. Table IV de-
vascular puncture and paraesthesia were also noted. picts block durations. The mean (SD) overall duration of
Sample size and statistical analysis surgical anaesthesia was shorter in Group 1% compared
In the absence of data from previous studies using to Group 2%, reflective of overall motor block duration.
20 ml of lidocaine 2% with epinephrine for ultrasound Individual sensory and motor block durations were simi-
guided axillary brachial plexus block, sample size was lar, with median motor block duration shorter in Group
calculated based on our pilot study of 10 patients. We 1%. Figure 2 shows the primary outcome measure, over-
found a mean ± SD onset time of 11.25±2.3 min. The all onset of surgical anaesthesia.
Table II. Patient characteristics
Group 2% (n=10) Group 1% (n=10) p value
Age, y 46.8±18.2 48±15.7 0.88
Sex, M/F, n 7/3 8/2 0.60
BMI, Kg/m2 25.3±3.2 24.5±3.6 0.62
ASA grade (I/II/III), n 7/3/0 4/6/0 0.18
Duration of surgery, min 62±10.8 58.5±14.9 0.56
Site of surgery (wrist/hand), n 5/5 6/4
Continuous variables are presented as means ± SD, categorical variables as counts
Med Ultrason 2022; 24(1): 38-43 41
Group 1% had a shorter needling time, performance Discussion
time and fewer needle passes when compared to Group
2%. No difference was found between the groups with In this single centre randomised controlled trial, we
respect to imaging time, incidence of vascular puncture observed that when using 400 mg of lidocaine with epi-
or paraesthesia (Table V). nephrine, increasing the volume of injectate by dilution
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
In still images the LPSL thickness was measured in Associations between categorical variables (individual
its long axis at three reference points: at its iliac enthesis sono-lesions) were examined through Chi-square test
(at a point where the deep margin of the ligament meets and/or Fisher’s exact test. Odds ratios were calculated to
the iliac bone), at its sacral enthesis (at a point where the establish the odds for the occurrence of a certain sono-
deep margin of the ligament meets the sacrum) and at lesion in patients with SIP. Pearson correlation r was
its middle part (ligament body). In addition, the ligament employed to examine the relationship between continu-
body thickness and width were measured in the short ously measured and normally distributed variables and
axis. As a final value at a given point, the mean of three Spearman rank-order correlation was performed when
consecutive measurements was recorded, as studies show those conditions were not observed. The intra-rater and
that repetitive measurements provide better reliability inter-rater agreement were established through Cohen’s
[13]. In addition, LPSL was assessed for possible so- kappa (95% confidence intervals).
nopathological features: 1. hypoechogenicity or altered
fibrillar structure; 2. presence of well-defined anechoic Results
zones within the ligament (partial-thickness tears); 3. an
anechoic rim along the entire or at least one third of the LPSL in healthy individuals
ligament length (periligamentous edema). All of these The LPSL was assessed bilaterally in 32 healthy in-
sono-markers were recorded as dichotomous variables, dividuals (64 ligaments), 22 females, 10 males, mean
i.e. present/absent. age of 41.97±12.87 years and mean BMI of 23.21±3.52.
Reference images of the LPSL were saved in a JPG We were able to visualize the ligament with the above-
format. They were used to test the intra-reader reliability described protocol in all subjects.
of the ultrasonographer (PT) and an inter-reader agree- In the longitudinal axis, the LPSL was seen as a hy-
ment with a novice sonographer with only a several perechoic, slightly concave fibrillar structure, with well-
months experience (LM). Before testing, the novice so- defined margins and predominantly uniform thickness
nographer received a special training in LPSL anatomy that spans over the SIJ and attaches to the ilium (just in-
and US appearance. For the reliability testing, 10 ran- ferior to the PSIS) and the lateral sacrum (to the sacral
domly selected sets of images, each set consisting of two tubercle, a structure of variable size in different subjects)
reference images of a single ligament, were used. The (fig 2b). In the short axis, the LPSL exhibited a flat to
intra-reader testing was performed about 6 months after oval shape, layered structure, well-defined upper margin,
the initial scan, blinded to the previous results and the and less conspicuous inferior margin (fig 2c). Medially,
identity of patients/controls. The sonopathological le- the ligament merged with the lateral portion of ESA, a
sions reported are relative to the first analysis. finding also shown previously [14]. The gluteus maxi-
Statistical analysis mus muscle was seen to override the ligament from a
The statistical software used to perform the analysis lateral direction, while the inferior gluteal aponeurosis
included IBM SPSS version 26 (2018), Minitab version attached to the LPSL’s lateral part. Thus, the LPSL ap-
19 (2019) and MedCalc version 19.4.1 (2020). We exam- peared as a link between the thicker and more stretched
ined the data for normality through the Shapiro-Wilk’s ESA medially and the thinner and looser GMA laterally.
test and took into consideration the values of skewness. Beneath the ligament, a fibro-adipose tissue of variable
Continuously measured and normally distributed vari- amount and echogenicity was seen. This tissue occupied
ables were described through the mean values and stand- the posterior part of the SIJ and expanded to the sacral
ard deviations (±SD). Categorical data were processed foramina and over the ilium.
in frequencies and percentages. An independent-samples The mean length of the LPSL was 32.32±4.29 mm,
t-test was used for two-group comparisons on normally and its thickness at the iliac entheses, the middle part
distributed continuous variables. When the assumption (ligament body) and the sacral entheses was 2.05±0.55
of equal variances was not observed (Levene’s statis- mm, 1.51±0.42 mm and 1.64±0.41 mm, respectively. The
tic p<0.05), we reported the results for equal variances width of the LPSL body measured in the transverse plane
not observed (uv). For multiple comparisons, Bonfer- was 8.14±1.28 mm.
onni correction was applied to control for Type I error. No significant associations were observed between
In such cases, statistical significance was accepted if the subjects’ age, sex, the side of the body and LPSL
p<0.0125. measurements.
ROC curve analysis was used to establish the di- LPSL in patients with SIP
agnostic potential, optimal criterion values and the as- For the second part of the study, 40 SIP patients were
sociated sensitivity and specificity of LPSL thickness. 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.
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 94.74%, positive predictive value = 91.19%, negative
pathology predictive value = 89.98%.
Thickening of the ligament body was found to be The second model (ligament body thickening plus
associated with the largest AUC among all three points periligamentous edema) was also significant (Chi-square
of measurement. Periligamentous edema and hypoecho- = 80.49, p<0.001) with AUC = 0.964 (95%CI: 0.931 to
genicity/altered fibrillar structure were the two most 0.996), sensitivity = 100%, specificity = 84.75%, positive
frequently observed sonopathological markers. Accord- predictive value = 81.26%, negative predictive value =
ingly, we tested the cumulative diagnostic accuracy of th 100%.
ligament body thickening in conjunction with either hy-
poechogenicity or periligamentous edema through binary Discussion
logistic regression and ROC curve analysis.
The first model (ligament body thickening plus hy- There is growing evidence that soft tissue structures
poechogenicity/altered structure) was significant (Chi- posterior to the SIJ (including the LPSL) could be respon-
square = 76.217, p<0.001), with AUC = 0.953 (95%CI: sible for a significant proportion of SIP [1,3,5,15,16]. For
0.911 to 0.995), sensitivity = 83.78%, specificity = example, it is frequently reported that there is no major
Med Ultrason 2022; 24(1): 44-51 49
difference in the efficacy of SIJ injections, regardless of In the majority of ligaments in the SIP study group
whether they are delivered strictly intraarticularly or the (72.5%), we observed a periligamentous edema. It was
infiltration is done periarticularly [17,18]. Indeed, the US seen as well-defined anechoic rim superiorly or, less
evaluation of the LPSL, performed in our study, revealed frequently, inferiorly along the LPSL body. This phe-
ligamentous sonopathological lesions in more than two nomenon has been previously reported in traumatic and
thirds of the patients with SIP. This is in line with the data degenerative ligamentous pathology [24,25], but its his-
of Murakami et al, who advocate that the first interven- tological nature remains unclear. In the case of LPSL
tion in SIP should be an injection in its ligamentous area, it may represent a pathological process similar to para-
as four of five SIP patients would have benefit from this tenonitis that involves the ESA and/or GMA.
approach [19]. The role of sacroiliac ligaments as pain In assessing the fibrillar structure and the echogenici-
generators was further confirmed empirically by Saun- ty of the LPSL, we found diffuse or local hypoechogenic-
ders et al. The authors evaluated the effect of US guided ity/altered fibrillar structure significantly more often in
injections in the LPSLs in patients with SIJ dysfunction ligaments from the SIP patients group. Hypoechogenic-
and pain [20] and reported significant improvement in ity of ligaments is considered to be caused by the disor-
all scores and performance indicators that lasted through- ganization of the collagen fibrils and edema of the clefts
out the entire 12-month follow-up period. In addition, a between the fibrils and the ligament matrix [22]. Thus,
recent study, performed by SPECT-CT, shows increased hypoechogenicity may indicate a pathological transfor-
uptake in the region of LPSL in patients complaining of mation of the ligamentous tissue, especially when this
SIP and a clinical diagnosis of SIJ dysfunction [21]. Fur- finding is accompanied by a thickening. In addition to
thermore, the intensity of the uptake could be well quan- this, the altered fibrillar structure may represent focal or
tified and possibly used as a follow-up tool. However, more global dehiscence between the ligament’s layers.
this imaging method utilizes an ionizing radiation, which Less frequently, partial intrasubstance tears of the
would limit its use in the daily clinical practice. LPSL were detected. This might be a rare injury in a tight
In the present study we used US, a safe and widely ligament such as the LPSL that supports an intrinsically
available imaging method. US was first used to assess stable joint as the sacroiliac [1,2], but nevertheless, no
the LPSL in healthy volunteers by Moore et al and later tears were detected among the 64 ligaments in the control
by LeGoff et al [10,11]. In addition to these studies, we group.
provide data on the normal thickness and width of LPSL The analysis of the cumulative occurrence of patho-
as well as more detailed description of its sonoanatomy logical sono-markers (other than thickening) in the LPSL
and relations to the surrounding structures. We have of patients vs. controls revealed that the majority of the
measured the LPSL thickness at three easily reproduc- painful ligaments (84%) exhibited two or all three patho-
ible points, namely at its iliac and sacral entheses and logical sono-markers. The combination of ligament’s
at its middle part. These sites were chosen on the basis body thickening and the presence of hypoechogenic-
of the morphology data, showing a different structure ity/altered fibrillar structure or periligamentous edema
of the ligament’s body in comparison to its entheses yielded the highest values for sensitivity and specificity
[4]. in identifying ligamentous injury in SIP patients.
Thickening is one of the major US features of liga- In our study we used a single clinical test to diag-
mentous injury [22]. Our results show that thickness was nose patients as having probable SIP (Fortin finger test
increased at all three sites, the ligament body being most [8]), while most of the studies on SIJ pathologies use a
pronounced. Comparing to its entheses, the LPSL body composite of tests [9,16,20]. Anyway, our target was the
has a more complex structure, consisting of layers and PSIL, which is an extraarticular and relatively superficial
incorporating the ESA and GMA. Thus, ligament body structure [3]. Such extraarticular and superficial source
thickening could in fact reflect the thickening of either of pain would be expected to be associated with a well-
or both of these aponeuroses – a finding similar to the described locus of pain [1,6] that the patents would be
thickening of the thoracolumbar fascia in patients with able to point out – like in the Fortin finger test. In addi-
chronic low back pain, described by Langevin et al [23]. tion, a similar test was used by other authors under the
However, this suggestion should be confirmed by his- name of a one-finger test resulting in a more accurate
tological proof. This ligament thickening could also be identification of the site of pain in the sacroiliac region
the morphological basis for the compression of the lat- [26].
eral branches of S2 and S3 dorsal rami and entrapment Our study has certain limitations. First and foremost,
neuropathy which was proposed by different authors as a findings were not compared to a “gold standard” or an-
potential reason for SIP [6,7,11]. 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 dorsal sacroiliac ligament: its implication for understanding
SIP would not normally need an operation. Thus, a con- low back pain. Spine (Phila Pa 1976) 1996;21:556-562.
firmation of the sonographic findings at surgery as a gold 4. McGrath C, Nicholson H, Hurst P. The long posterior sac-
roiliac ligament: a histological study of morphological rela-
standard is virtually impossible. Two other possible im-
tions in the posterior sacroiliac region. Joint Bone Spine
aging modalities to assess the LPSL would be MRI and
2009;76:57-62.
SPECT-CT. MRI has certain limitations as ligaments 5. Borowsky CD, Fagen G. Sources of sacroiliac region pain:
are relatively avascular structures. Besides MRI is still insights gained from a study comparing standard intra-ar-
an expensive option. SPECT-CT seems promising for ticular injection with a technique combining intra- and peri-
depicting LPSL pathology, but it involves a consider- articular injection. Arch Phys Med Rehabil 2008;89:2048-
able amount of ionizing radiation, and the images have a 2056.
lower anatomical resolution. Secondly, patients and con- 6. McGrath MC, Jeffery R, Stringer MD. The dorsal sacral
trol subjects in the study were with normal BMI while in rami and branches: Sonographic visualization of their vas-
more obese patients, the examination of the LPSL could cular signature. Int J Osteopath Med 2012;15:3-12.
be more difficult. However, as the ligament presents a 7. Zhou L, Schneck CD, Shao Z. The Anatomy of Dorsal Ra-
mus Nerves and Its Implications in Lower Back Pain, Neu-
well-defined hyperechoic fibrillar structure, it should be
rosci Med 2012;3:192-201.
possible to assess it even in these cases. Thirdly, through- 8. Fortin JD, Falco FJ. The Fortin finger test: an indicator of
out the course of data collection, the researcher undertak- sacroiliac pain. Am J Orthop (Belle Mead NJ) 1997;26:477-
ing the US assessment was not blinded to the group of 480.
each participant, and as such, there was a potential for 9. 9. Bogduk N. Pain provocation tests for the assessment of
observer bias. However, to minimize subjectivity, a strict sacroiliac joint dysfunction. J Spinal Disord 1999;12:357-
scanning protocol was followed in each subject and the 358.
sonopathological lesions were well- and predefined and 10. Moore AE, Jeffery R, Gray A, Stringer MD. An anatomical
repeated measurements undertaken. Lastly, we did not ultrasound study of the long posterior sacro-iliac ligament.
use diagnostic periarticular SIJ injections to differenti- Clin Anat 2010;23:971-977.
11. Le Goff B, Berthelot JM, Maugars Y. Ultrasound assess-
ate patients with ligamentous pain. Yet, periarticular SIJ
ment of the posterior sacroiliac ligaments. Clin Exp Rheu-
injections are still not that well standardized (as the in-
matol 2011;29:1014-1047.
traarticular ones) and one problem to their diagnostic use 12. Arnbak B, Hendricks O, Hørslev-Petersen K, et al. The
might be the unpredictable spread of the injectate in the discriminative value of inflammatory back pain in pa-
periarticular tissues. tients with persistent low back pain. Scand J Rheumatol
2016;45:321-328.
Conclusion 13. Fede C, Gaudreault N, Fan C, Macchi V, De Caro R, Stecco
C. Morphometric and dynamic measurements of muscular
Our study confirms that the LPSL could be visualized fascia in healthy individuals using ultrasound imaging: a
by US in details and its structure and relations - accurate- summary of the discrepancies and gaps in the current litera-
ly assessed. In addition, in patients with noninflamma- ture. Surg Radiol Anat 2018;40:1329-1341.
14. Todorov P, Nestorova R, Batalov A. The sonoanatomy of
tory SIP, the US evaluation of the LPSL could reveal its
lumbar erector spinae and its iliac attachment - the poten-
thickening as well as sonopathological transformation: tial substrate of the iliac crest pain syndrome, an ultrasound
hypoechogenicity/altered fibrillar structure and periliga- study in healthy subjects. J Ultrason 2018;18:16-21.
mentous edema. Further studies are required in order to 15. Kurosawa D, Murakami E, Ozawa H, et al. A Diagnostic
confirm the value of these findings for the clinical practice. Scoring System for Sacroiliac Joint Pain Originating from
the Posterior Ligament. Pain Med 2017;18:228-238.
Conflict of interest: none 16. Vleeming A, Mooney V, Stoeckart R. Movement, stability
and lumbopelvic pain. Integration of research and therapy.
References 2nd Edition. Churchill Livingstone Elsevier, 2007.
17. Nacey NC, Patrie JT, Fox MG. Fluoroscopically Guided
1. Bogduk N. Clinical anatomy of the lumbar spine and sa- Sacroiliac Joint Injections: Comparison of the Effects
crum. 4th Edition. Elsevier, 2005. of Intraarticular and Periarticular Injections on Immedi-
2. Vleeming A, Schuenke MD, Masi AT, Carreiro JE, Dan- ate and Short-Term Pain Relief. AJR Am J Roentgenol
neels L, Willard FH. The sacroiliac joint: an overview of 2016;207:1055-1061.
its anatomy, function, and potential clinical implications. J 18. Murakami E, Tanaka Y, Aizawa T, Ishizuka M, Kokubun S.
Anat 2012;221:537-567. Effect of periarticular and intraarticular lidocaine injections
3. Vleeming A, Pool-Goudzwaard AL, Hammudoghlu D, for sacroiliac joint pain: prospective comparative study. J
Stoeckart R, Snijders CJ, Mens JM. The function of the long Orthop Sci 2007;12:274-280.
Med Ultrason 2022; 24(1): 44-51 51
19. Murakami E, Kurosawa D, Aizawa T. Treatment strategy human subjects with chronic low back pain. BMC Muscu-
for sacroiliac joint-related pain at or around the posterior su- loskelet Disord 2009;10:151.
perior iliac spine. Clin Neurol Neurosurg 2018;165:43-46. 24. Shahabpour M, Staelens B, Van Overstraeten L, et al. Ad-
20. Saunders J, Cusi M, Hackett L, Van der Wall H. An explora- vanced imaging of the scapholunate ligamentous complex.
tion of ultrasound-guided therapeutic injection of the dorsal Skeletal Radiol 2015;44:1709-1725.
interosseous ligaments of the sacroiliac joint for mechanical 25. Ward R, Mendoza J, Wong E, et al. MCL Periligamen-
dysfunction of the joint. JSM Pain Manag 2016;1:1003-1007. tous Edema: Sprain or Nontraumatic Meniscal Pathology?
21. Cusi M, Saunders J, Van der Wall H, Fogelman I. Meta- Radiological Society of North America 2010 Scientific
bolic disturbances identified by SPECT-CT in patients with Assembly and Annual Meeting, November 28 - Decem-
a clinical diagnosis of sacroiliac joint incompetence. Eur ber 3, 2010, Chicago IL. Available from: http://archive.
Spine J 2013;22:1674-1682. rsna.org/2010/9012278.html, Accessed September 28,
22. Hodgson RJ, O’Connor PJ, Grainger AJ. Tendon and liga- 2020.
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 Uozumi H. Diagram specific to sacroiliac joint pain site in-
evidence of altered lumbar connective tissue structure 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
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
Med Ultrason 2022; 24(1): 52-57 53
is compressed, RA blood flow increases, which indicates Catheterization technique
a functional continuity between radial and ulnar circula- The right forearm was abducted and placed on the
tion in the hand [5]. catheterization table with hyperextension of the wrist.
The first invasive stage of PCI procedures is the cath- After a local anesthetic injection, arterial puncture was
eterization of the RA or UA. In many studies, despite achieved. Following palpation of the site of maximal
conflicting results [6], changes in blood flow parameters pulse prominence, using the Seldinger technique, a 0.025-
and diameters of the RA and UA after reciprocal com- inch straight guidewire was passed through a needle. The
pression/occlusion have been investigated in the early or needle was removed and a 5F or 6F hydrophilic sheath
late periods of the procedure using Doppler ultrasonog- was placed over the guidewire. A spasmolytic cocktail
raphy (DUS) [7-10]. However, blood flow characteristics (100-200 µg of Perlinganit and 12.5 µg of diltiazem, if
and diameter changes in both arteries (catheterized and required) and heparin (5000 IU bolus in case of diagnos-
non-catheterized) before and after RA or UA catheteriza- tic catheterization or 70-100 IU/kg in case of PCI) were
tion for CAG and/or PCIs have not been examined or infused intra-arterially through the sheath. For diagnostic
compared during the early period. purposes, conventional 5F Tiger catheters (Terumo Corp.,
Our study hypothesis focusing on RA and UA cath- Tokyo, Japan), and for PCI 5F or 6F (rarely, in complex
eterization (including cannulation, decannulation, sheath PCI) extra back-up, Judkins, or Amplatz guiding cath-
insertion, sheath removal and 4 hours of gradually de- eters were used. After CAG or PCI, the arterial sheath
creasing compression) was that this procedure will cause was removed immediately following cardiac catheteriza-
changes in the structure and hemodynamics of both arter- tion completion. Local haemostasis was achieved using a
ies. For this reason, the aim of our study was to evaluate pressure bandage (Terumo, Tokyo, Japan) inflated at the
the structural and hemodynamic changes in RA and UA puncture site with 15 to 20 mL of air. The pressure used
after catheterization by using DUS. for UA compression was higher than that for the RA, as
there is no bone structure to compress the UA at the bot-
Materials and methods tom and the UA is a deeper vessel. These bandages were
removed after 4 hours and the patients were discharged
Study population within 48 hours of procedure completion. Catheterization
Between January 2018 and August 2020, a total of procedures for both arteries were performed by the same
328 patients who underwent CAG and/or PCI were in- operator (H.G.) with more than 10 years of experience
cluded in the study: 171 patients via a transradial ap- and over 7000 catheterizations.
proach (TRA), 157 patients via a transulnar approach Ultrasonographic evaluation
(TUA). All procedures were performed through the right All DUS examinations were performed by the same
UA and RA. DUS was carried out before procedure radiologist (Y.G) with more than 20 years of experience,
(maximum 48 hours before catheterisation) and 5-48 in DUS using an Aplio 400 Doppler USG system (Toshi-
hours after compression bandage was removed. ba, Tokyo, Japan) with a 7.2-14 MHz multi-frequency
The main indication for TUA or TRA was a wider linear array transducer.
and easily palpable pulse. The demographic and clinical The lumen diameters, peak systolic velocity (PSV),
characteristics (age, gender, systolic and diastolic blood end-diastolic velocity (EDV), pulsatility index (PI) and
pressure, clinical diagnoses, fasting blood glucose, lipid resistive index (RI) were measured at the wrist level. The
profile, history of diabetes mellitus, hypertension, smok- sagittal diameter of the RA and UA were measured at
ing, chronic kidney disease, and drug use) were recor- the diastolic phase (from intima to intima) in a short-axis
ded. section. The ultrasound examination was performed at
Study exclusion criteria were as follows: <18 or >80 room temperature (22 and 23°C), the examination room
years of age, previous CAG and/or PCI by TRA or TUA, temperature being recorded before each DUS.
severe heart failure, hemodynamic impairment, uncon- Statistical analysis
trolled hypertension, acute or chronic renal failure or An independent Student’s t-test was used to compare
hepatic diseases, chronic pulmonary disease, bleeding data for continuous variables which were expressed as
diathesis, severe skeletal forearm deformities, peripheral mean and standard deviation [age, blood pressure, blood
artery disease (i.e. Raynaud’s disease), or previous coro- biochemistry measurements (fasting glucose, LDL cho-
nary artery bypass surgery using an RA graft. lesterol, and triglycerides)]. The diameter and Doppler
Sakarya University Medical Faculty Ethics Commit- flow of the RA and UA before and after the procedure
tee approved the study protocol and each patient signed a were compared using a paired t-test. Categorical varia-
written informed consent form. 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, zation procedures for CAG and PCI were included in this
dyslipidaemia, and chronic kidney disease) were com- study. The demographic, clinical, laboratory and proce-
pared by Pearson’s chi-squared test or Fisher’s exact test, dural time data of these groups are detailed in Table I.
when appropriate. Ultrasonographic measurements of radial and ulnar
All analyses were performed using the SPSS Statis- arteries before and after catheterization are specified in
tics software package for Windows, version 22.0 (SPSS table II.
Inc., Chicago, IL, USA). A value of P <0.05 was consid- Comparing the right and left RA and UA preprocedur-
ered statistically significant. al ultrasound parameters, no significant differences were
Considering the RA and UA catheterizations as two found concerning the diameters (left RA and 2.64±0.74
different access methods (from 2020 Medcalc statisti- mm, left UA 2.47±0.98 mm) or Doppler flow measure-
cal software, Ostend, Belgium), the measurements were ments (all p>0.05). The RA was the dominant forearm
plotted using the Bland-Altman method to determine artery in most cases (198 patients, 62.2%).
whether the differences showed a systematic distribution Due to the detection of significant changes in the
around zero and its prevalence. diameter and PSV of both arteries after catheterization,
the compliance levels examined using the Bland-Altman
Results method showed no difference. The plots showed a sys-
tematic distribution of the differences around zero from
A total of 328 consecutive adult patients undergoing the distribution graphs and a clear relationship between
selective transradial (171) or transulnar (157) catheteri- the differences and the means (fig 1).
Discussion
RA diameter was greater than that of the UA may have intervention; however, they reported that this decrease
affected these results. The fact that the basal UA flow did not reach statistical significance [8]. Madssen et al
velocity was higher than the basal RA flow velocity (as found a significant reduction in the right RA diameter
in the left arm) should not be ignored since the results compared to left RA diameter at a control USG examina-
obtained in our study may be related to the diameter and tion 12 months after CAG via the right RA [9].
flow characteristics found during basal measurements (in Our study has some limitations. This was a single-
other words, already existing). centre study with a nonrandomized design. Larger, rand-
Considering the data obtained, our study showed a omized, multi-centre and prospective studies are needed
significant increase in blood flow rates in the UA in cases to identify changes in diameter and blood flow parameters
with RA catheterization, indicating a functional continu- using DUS in patients undergoing coronary angiography/
ity between the radial and ulnar circulation in both hands. PCI to confirm our results. Since all DUS examinations
Changes in the inner diameter of an artery can affect arte- were realized by the same radiologist, and the examina-
rial blood flow rates by causing changes in arterial wall tions were carried out in a short period before and after
properties. This equation shows a strong and inverse re- the procedure, the intraobserver variation analysis was
lationship between the blood flow velocity of one artery not performed in this study. In addition, in our study, only
and the internal diameter of the other artery in the same the early effects of catheterization in patients who un-
arm. Therefore, the decrease in vessel lumen diameter derwent PCI (patients are generally discharged within 48
of the catheterized artery may be associated with the in- hours after the procedure and it is thought that the effects
crease in the flow velocity of the other artery in the same of catheterization in the long period may be reversible)
arm. With RA catheterization, the distal perfusion pres- were evaluated.
sure of the capillary bed decreases, causing an increase
in the pressure gradient between the UA and the capillary Conclusion
bed, resulting in increase of UA blood flow [17,18].
Arterial cannulation/compression or catheterization The data obtained from our study suggest that cath-
results in a decrease in blood flow and sometimes oc- eterization (starting with cannulation, ending with the re-
clusion of the catheterized artery in the early or late pe- moval of the compression bandage after 4 hours) causes
riods. If the radial/ulnar artery is narrowed or occluded, vascular wall changes in the catheterized artery. In our
distal perfusion of the capillary bed will be decreased and opinion, while the procedure causes a decrease in diame-
blood flow in the ipsilateral artery will be increased. In- ter and blood flow velocity through vascular wall chang-
creases in the diameter and flow rates of the other arter- es and arterial compression in the catheterized artery, it
ies in patients developing stenosis or RA occlusion and protects the arm against ischemia by causing an increase
stenosis or UA occlusion ensure sufficient blood supply in the compensatory diameter and blood flow in the other
to the hand and protect the hand against ischemic events. artery of the same arm. In addition, the UA can be safely
Significant increases in the diameter, velocity, and vol- used as an alternative to the RA catheterization.
ume flow of the arteries during reciprocal compression/
cannulation or catheterization of the two arteries in dif- Conflict of interest: none
ferent studies, including our study, support this assump-
tion [19,20]. References
It has been demonstrated that RA cannulation would 1. Kar S. Transulnar cardiac catheterization and percutaneous
decrease RA blood flow immediately after catheter place- coronary intervention: techniques, transradial comparisons,
ment and the compensatory increase in UA blood flow anatomical considerations, and comprehensive literature
might quickly commence. Significant changes continued review. Catheter Cardiovasc Interv 2017;90:1126-1134.
after 24 hours from the beginning of monitoring [18,21]. 2. Kaplanoglu H, Dinc E. Diameters and Flow Characteristics
Zhenxian et al investigated the impact of transradial of Forearm Arteries in the Preoperative Period Using Dop-
coronary procedures on the RA by ultrasonography. The pler Ultrasonography in Healthy Individuals. Iran J Radiol
PSV and diameter of the right RA were measured before, 2018;15:e12904.
3. Marchese RM, Geiger Z. Anatomy, Shoulder and Upper
the first day and the first month after transradial coro-
Limb, Forearm Radial Artery. StatPearls Publishing, 2021.
nary interventions. They found that the diameter of the
4. Pancholy SB, Heck LA, Patel T. Forearm arterial anatomy
RA decreased on the first day after the procedure and and flow characteristics: a prospective observational study.
this decrease continued in the control performed 1 month J Invasive Cardiol 2015;27:218-221.
later [22]. Abe et al found that a slight decrease in RA 5. Kim SY, Lee JS, Kim WO, Sun JM, Kwon MK, Kil HK.
diameter continued even 3 months after the transradial Evaluation of radial and ulnar blood flow after radial artery
Med Ultrason 2022; 24(1): 52-57 57
cannulation with 20- and 22-gauge cannulae using duplex 15. Doscher W, Viswanathan B, Stein T, Margolis IB. Hemo-
Doppler ultrasound. Anaesthesia 2012;67:1138-1145. dynamic assessment of the circulation in 200 normal hands.
6. Tonks AM, Lawrence J, Lovie MJ. Comparison of ulnar Ann Surg 1983;198:776–779.
and radial arterial blood-flow at the wrist. J Hand Surg Br 16. Yilmaztepe MA, Yilmaz E. Effect of transient ulnar ar-
1995;20:240–242. tery compression on radial artery diameter. Exp Ther Med
7. Ruengsakulrach P, Brooks M, Hare DL, Gordon I, Buxton 2018;16:3735-3739.
BF. Preoperative assessment of hand circulation by means 17. Beniwal S, Bhargava K, Kausik SK. Size of distal radial
of Doppler ultrasonography and the modified Allen test. J and distal ulnar arteries in adults of southern Rajasthan and
Thorac Cardiovasc Surg 2001;121:526–531. their implications for percutaneous coronary interventions.
8. Abe S, Meguro T, Naganuma T, Kikuchi Y. Change in the Indian Heart J 2014;66:506-509.
diameter of the radial artery transradial intervention using 18. Kim EJ, Soh S, Kim SY, et al. Impact of Diabetes Mellitus
a 6 French system in Japanese patients. J Invasive Cardiol on Radial and Ulnar Arterial Vasoreactivity after Radial Ar-
2001;13:573-575. tery Cannulation: A Randomized Controlled trial. Int J Med
9. Madssen E, Haere P, Wiseth R. Radial artery diameter and Sci 2016;13:701-707.
vasodilatory properties after transradial coronary angiogra- 19. Saito S, Ikei H, Hosokawa G, Tanaka S. Influence of the
phy. Ann Thorac Surg 2006;82:1698–1702. ratio between radial artery inner diameter and sheath outer
10. Peruga JP, Peruga JZ, Kasprzak JD, et al. Ultrasound evalu- diameter on radial artery flow after transradial coronary in-
ation of forearm arteries in patients undergoing percutane- tervention. Catheter Cardiovasc Interv 1999;46:173-178.
ous coronary intervention via radial artery access: results of 20. Hahalis G, Aznaouridis K, Tsigkas G, et al. Radial Artery
one-year follow-up. Kardiol Pol 2015;73:502-510. and Ulnar Artery Occlusions Following Coronary Pro-
11. Mason PJ, Shah B, Tamis-Holland JE, et al; American Heart cedures and the Impact of Anticoagulation: ARTEMIS
Association Interventional Cardiovascular Care Committee (Radial and Ulnar ARTEry Occlusion Meta-Analys IS)
of the Council on Clinical Cardiology; Council on Cardio- Systematic Review and Meta-Analysis. J Am Heart Assoc
vascular and Stroke Nursing; Council on Peripheral Vascu- 2017;6:e005430.
lar Disease; Council on Genomic and Precision Medicine. 21. Brodman RF, Hirsh LE, Frame R. Effect of radial artery
An Update on Radial Artery Access and Best Practices for harvest on collateral forearm blood flow and digital perfu-
Transradial Coronary Angiography and Intervention in sion. J Thorac Cardiovasc Surg 2002;123:512-516.
Acute Coronary Syndrome: A Scientific Statement From 22. Yan Z, Zhou Y, Zhao Y, Zhou Z, Yang S, Wang Z. Impact of
the American Heart Association. Circ Cardiovasc Interv transradial coronary procedures on radial artery. Angiology
2018;11:e000035. 2010;61:8-13.
12. He GW, Yang CQ. Characteristics of adrenoceptors in the 23. Huzjan R, Brkljacić B, Delić-Brkljacić D, Biocina B, Sutlić
human radial artery: clinical implications. J Thorac Cardio- Z. B-mode and color Doppler ultrasound of the forearm ar-
vasc Surg 1998;115:1136-1141. teries in the preoperative screening prior to coronary artery
13. Brzezinski M, Luisetti T, London MJ. Radial artery bypass grafting. Coll Antropol 2004;28 Suppl 2:235-241.
cannulation:a comprehensive review of recent anatomic and 24. Ashraf T, Panhwar Z, Habib S, Memon MA, Shamsi F, Arif
physio-logic investigations. Anesth Analg 2009;109:1763– J. Size of radial and ulnar artery in local population. J Pak
1781. Med Assoc 2010;60:817-819.
14. Zabad MN, Janzer S, George JC. Radial and Ulnar Artery 25. Yan ZX, Zhou YJ, Zhao YX, Zhou ZM, Yang SW, Wang
Occlusion Following Transradial and Transulnar Coronary ZJ. Anatomical study of forearm arteries with ultrasound
Intervention Requiring Endovascular Revascularization. for percutaneous coronary procedures. Circ J 2010;74:686-
Cath Lab Digest 2019;27:6. 692.
Original papers Med Ultrason 2022, Vol. 24, no. 1, 58-64
DOI: 10.11152/mu-3088
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
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 cardium became thinner, systolic thickening rate became
of PTT based on different calculation methods in differ- greater and LVEF increased significantly. With Iso injec-
entiating Aden or/and Iso group from control subjects, tion, the heart rate increased more significantly (nearly
determine the optimal cut-off points and validity param- 500 bpm). Compared to the control group, the systolic
eters. Bland & Altman was used to measure the inter/ cardiac chambers became even smaller, the myocardium
intra observer variability. A value of p<0.05 was consid- thickened at the end of systole and the LVEF increased
ered statistically significant. All statistical analysis was more significantly. But, the RVd and the difference be-
performed with SPSS version 16 software for Windows tween RVd and RVs became smaller. In addition, systolic
(SPSS Inc, Chicago, IL). thickening rate in the Iso group was slightly lower than in
the Ade group. With a combination injection of Ade and
Results Iso, there was no change in heart rate. Compared to the
control group, the cardiac size, myocardium thickness,
Tolerability and contrast echocardiography quality systolic thickening rate, LVEF and RV function did not
All rats could tolerate the dosage of Ade or/and Iso change significantly, too.
injected through the tail vein. No adverse reactions were PTT values
observed. Using the high frequency transducer and the As shown in figure 2, whether it was PTT1,PTT2 or
dosage of Sonovue mentioned above, good image quality PTT3, with the Ade injection, they all significantly de-
was achieved in all rats. clined (PTT1: from 0.99±0.39 s to 0.72±0.26 s, p<0.05;
Echocardiographic and hemodynamic PTT2: from 4.77±2.39 s to 2.44±0.95 s, p<0.01; PTT3:
characteristics from 32.71±9.31 s to 16.59±5.23 s, p<0.001). With the
As shown in Table I, with Ade injection, heart rate Iso injection, however, they all significantly length-
increased significantly. Compared to the control group, ened (PTT1: from 0.99±0.39 s to 2.87±0.59 s, p<0.001;
the systolic cardiac chambers became smaller, the myo- 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- were maximal in PTT3, i.e, Ade is most likely to cause
ingly, with a combination injection of Ade and Iso, all changes in PTT3. In the Iso group, the AUC, specific-
the PTT values were the same as the control group’s ity and Youden’s index were all were maximal in PTT1:
(PTT1:1.02±0.43 s vs.0.99±0.39 s, p>0.05; PTT2: therefore, Iso is the most likely to cause changes in PTT1.
4.73±2.51s vs. 4.77±2.39 s, p>0.05; PTT3: 32.94±9.35 s In the Ade+ Iso group, each AUC was less than 0.5, i.e,
vs. 32.71±9.31 s, p>0.05). the combined administration of Ade and Iso could not
ROC curve analysis cause changes in any PTT.
Comparison of the performance of PTT based on dif- Reproducibility
ferent calculation methods in differentiating the Ade or/ Bland-Altman analysis showed the interobserver and
and Iso group from the control subjects, in other words, intraobserver agreement for the measurement of PTT
which one was more susceptible to Ade or/and Iso among values were very good: intraclass correlation coefficients
PTT1, PTT2 and PTT3, was shown in Table II. In the were all >0.97 for interobserver and intraobserver vari-
Ade group, the AUC, sensitivity and Youden’s index 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 nary vascular status, and then result in the PTT changes.
In this study, the rats were divided into 4 groups of con-
This is the first report on the effects of Ade, Iso and trol, Ade, Iso, and Aden+Iso just to explore their effect on
their combinations on PTT in rats using contrast echocar- PTT in detail. The results showed that a shortened PTT
diography. The results presented here indicate that Ade was found in the Ade group, a lengthened PTT in the Iso
or/and Iso exerted distinct effects on PTT: Ade shortened group and an unchanged PTT in the Ade+Iso group. Its
PTT, Iso lengthened PTT and the combinations of Ade underlying mechanisms are not fully known and the possi-
and Iso had no significant effect on PTT. Moreover, Ade ble mechanisms involved require to be further discussed.
had a greater effect on PTT3 while Iso had a greater ef- Firstly, Ade and Iso, alone or simultaneously, have
fect on PTT1. These findings suggest that we need to different effects on cardiac function. Through 2D echo
consider the effects of medicine (especially cardiopul- examination, it was found that the ability of Ade to en-
monary vasoactive drugs) on the PTT values during their hance RV contraction (represented by RVFWSTR and
usage. It also provides the basis for the translation into the difference of RVd and RVs) was greater than that of
clinical practice of consecutive Iso/Ade treatment from Iso, with a normal LV diastolic volume (represented by
the perspective of pulmonary circulation. LVd) and RV diastolic volume (i.e., the volume of sys-
Although PTT is defined as the time required for a temic circulation back to the heart; represented by RVd).
volume of contrast to travel from RV to LA or LV. The Enhanced right ventricular systolic function, normal RV
concept is still not very clear now, because it does not diastolic volume and normal LV diastolic volume (equiv-
specify when (initial or peak opacification?) starts in alent to normal left atrial pressure) would result in a
order to time when the contrast agent reaches the heart shortened PTT. The ability of Iso to enhance LV contrac-
cavity (RV, LA or LV). Most studies take the time re- tion (represented by LVEF and the difference of LVd and
quired for a volume of contrast to travel from RV peak LVs) was greater than that of Ade, with a normal LV di-
opacification to LA (or LV) peak opacification as PTT, astolic volume (represented by LVd), but with a reduced
i.e., PeakRV-PeakLV PTT (PTT2) in our study. There are RV contractility (represented by the difference of RVd
also some studies that use the time required for a volume and RVs), a decrease in RV diastolic volume (represented
of contrast to travel from RV initial appearance to LA (or by RVd) and with a thicker myocardium and impaired
LV) the initial appearance as PTT, i.e., OnsetRV-OnsetLV systolic thickening rate. Reduced right ventricular systol-
PTT (PTT1) in our study. There is another calculation of ic function, decreased RV diastolic volume (equivalent to
PTT: the time required for a volume of contrast to travel RV preload) and normal LV diastolic volume would re-
from RV initial appearance to LA (or LV) peak opacifica- sult in a lengthened PTT. Compared to the control group,
tion, i.e. OnsetRV-PeakLV PTT (PTT3) in our study. PTT1 the combination of Ade and Iso had no significant effect
is believed to mainly reflect the pulmonary vascular sta- on cardiac size, cardiac function and so on. Therefore, no
tus, such as intrapulmonary vasodilatation, pulmonary changes were seen in PPT.
arteriovenous fistula and abnormal angiogenesis, directly Secondly, Ade and Iso, alone or simultaneously, have
connected arteries-veins, and so on. Our previous study different effects on pulmonary vascular status. In our
confirmed PTT1 is a reliable marker for the differential study, Ade and Iso were all administered intravenously
diagnosis of pulmonary nodules [9]. PTT3 is believed to through the tail vein. Ade acted as a potent selective pul-
a very important potential parameter for the calculation monary vasodilator by increasing intracellular cAMP via
of pulmonary vascular volume although some studies A2 receptor agonism [14] because of its rapid endothelial
have used PTT2 rather than PTT3 as a reference to calcu- metabolism [15]. Pulmonary angiectasia would result in
late it [7,12]. Therefore, in this study, three kinds of PTT a decreased pulmonary vascular volume and a shortened
were calculated for comparison. PTT. Iso, a non-selective β-adrenergic agonist, acting on
Ade and Iso, in addition to routine clinical use, also the β2 receptor of the bronchial smooth muscle, caused
used to mimic the potent cardioprotection of temperature a strong relaxation effect on the bronchial smooth mus-
preconditioning by consecutive activation of protein ki- cle, and relived bronchial spasm. However, the bronchial
nases A and protein kinases C [10,11]. Moreover, Iso and smooth muscle relaxation made airway resistance re-
Ade, alone or simultaneously, protected isolated rat hearts duced, the latter further made ventilation perfusion ratio
but the consecutive treatment gave the highest protection abnormal and hypoxemia aggravated, and finally caused
[13]. Such a strategy seems to be useful in cardiac sur- a hypoxic pulmonary vasoconstriction. In this setting,
gery involving ischaemic cardioplegic arrest and cardio- PTT was significantly prolonged. As for the effects of
pulmonary bypass. However, in vivo, it may also bring Ade+Iso on pulmonary vascular status, they might cancel
about impacts on cardiopulmonary function and pulmo- each other out, so PTT did not change.
Med Ultrason 2022; 24(1): 58-64 63
In our previous experimental studies (unpublished References
data), the dose of Ade and Iso was explored, and the dose
1. Brittain EL, Doss LN, Saliba L, Irani W, Byrd BF 3rd, Mo-
gradient (Ade: 0.10, 0.23, 0.46 and 0.92 uM/mL; Iso: 1.2,
nahan K. Feasibility and diagnostic potential of pulmonary
2.4, 4.8, 9.6 uM/mL) had been done, and each dose of
transit time measurement by contrast echocardiography: a
Ade and Iso had similar effects on PTT. In this study, the pilot study. Echocardiography 2015;32:1564-1571.
optimal concentration (Ade: 0.46 uM/mL; Iso: 2.4 uM/ 2. Herold IHF, Saporito S, Bouwman RA, et al. Reliability, re-
mL) was adopted. Our previous studies had explored the peatability, and reproducibility of pulmonary transit time
modalities of administration of Ade and Iso, too. Ade ad- assessment by contrast enhanced echocardiography. Car-
ministered via the jugular vein was complicated and had diovasc Ultrasound 2016;14:1.
a low success rate. Although intraperitoneal injection of 3. Monahan K, Coffin S, Lawson M, Saliba L, Rutherford R,
Iso had a high success rate; it couldn’t be administered Brittain E. Pulmonary transit time from contrast echocardi-
simultaneously with Ade. Therefore, simultaneous ad- ography and cardiac magnetic resonance imaging: Compar-
ministration of Ade and Iso through tail vein was the best ison between modalities and the impact of region of interest
characteristics. Echocardiography 2019;36:119-124.
option. Due to the small size and fast circulation of rats,
4. Zhao H, Tsauo J, Zhang X, et al. Pulmonary transit time
Ade may be metabolized less than we think. derived from pulmonary angiography for the diagnosis of
Even so, our study had several limitations that should hepatopulmonary syndrome. Liver Int 2018;38:1974-1981.
be considered. First of all, doses of Ade and Iso used in 5. Crosara S, Ljungvall I, Margiocco ML, Häggström J, Tar-
this study were selected according to our previous stud- ducci A, Borgarelli M. Use of contrast echocardiography
ies, without reference to those in vitro studies (sequen- for quantitative and qualitative evaluation of myocardial
tially with 5 nM Iso and 30 μM Ade) [10,11]. Further perfusion and pulmonary transit time in healthy dogs. Am J
systematic exploration is required. Next, Ade and Iso in Vet Res 2012;73:194-201.
this study were all administered through tail-intravenous 6. Streitberger A, Hocke V, Modler P. Measurement of pul-
injection rather than subcutaneous injection, intraperito- monary transit time in healthy cats by use of ultrasound
contrast media “Sonovue”: Feasibility, reproducibility, and
neal injection, heart cavity injection, pulmonary vascular
values in 42 cats. J Vet Cardiol 2013;15:181-187.
injection and bronchial injection, etc. The conclusions
7. Ali LA, Aquaro GD, Peritore G, et al. Cardiac magnetic
about the effects of Ade and Iso, alone or simultaneously resonance evaluation of pulmonary transit time and blood
on PTT may be one-sided. Finally, we still need more volume in adult congenital heart disease. J Magn Reson Im-
systematic researchs, including pathology at least, to sup- aging 2019;50:779-786.
port our idea. 8. Colin GC, Pouleur AC, Gerber BL, et al. Pulmonary hy-
pertension detection by computed tomography pulmonary
Conclusion transit time in heart failure with reduced ejection fraction.
Eur Heart J Cardiovasc Imaging 2020;21:1291-1298.
In this study, the effects of Ade, Iso and their combina- 9. Wang B, Sun F, Zheng XZ, Sun CY. A novel application of pul-
tions on PTT in a rat model using contrast echocardiogra- monary transit time to differentiate between benign and ma-
lignant pulmonary nodules using myocardial contrast echo-
phy was compared. The findings demonstrated that Ade
cardiography. Int J Cardiovasc Imaging 2021;37:1215-1223.
or/and Iso exerted distinct effects on PTT: Ade shortened 10. Lewis M, Szobi A, Balaska D, et al. Consecutive Isoproter-
PTT, Iso lengthened PTT, and the combinations of Ade enol and Adenosine Treatment Confers Marked Protection
and Iso had no significant effect on PTT. Although it had against Reperfusion Injury in Adult but Not in Immature
some limitations, this study still reminds us of the need Heart: A Role for Glycogen. Int J Mol Sci 2018;19:494.
to consider the effects of medicine (especially cardiopul- 11. Khaliulin I, Halestrap AP, Bryant SM, Dudley DJ, James
monary vasoactive drugs) on the PTT values. At the same AF, Suleiman MS. Clinically-relevant consecutive treat-
time, it provides the basis for the clinical transformation ment with isoproterenol and adenosine protects the fail-
of consecutive Iso/Ade treatment from the perspective of ing heart against ischaemia and reperfusion. J Transl Med
pulmonary circulation. 2014;12:139.
12. Galanti G, Jayaweera AR, Villanueva FS, Glasheen WP,
Ismail S, Kaul S. Transpulmonary transit of microbubbles
Acknowledgements: The authors gratefully ac-
during contrast echocardiography: implications for estimat-
knowledge the assistance of Xiancheng Xia from the ing cardiac output and pulmonary blood volume. J Am Soc
Department of Ultrasound, The Yancheng Clinical Col- Echocardiogr 1993;6:272-278.
lege of Xuzhou Medical University, Yancheng, Jiangsu 13. Khaliulin I, Parker JE, Halestrap AP. Consecutive pharma-
Province, P.R. China. cological activation of PKA and PKC mimics the potent
cardioprotection of temperature preconditioning. Cardio-
Conflict of interest: none 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
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” lung parenchymal diseases with their subpleural consoli-
topic the technical requirements were explained and the dations were reflected [2].
use of ultrasound (US) in the lung and pleura in paedi- Herewith, in the third part the use of US for chest wall,
atric patients were discussed [1]. In the second part, the mediastinum, diaphragmatic diseases, trachea, interven-
tions and artifacts in paediatric patients are summarized.
Received 05.03.2021 Accepted 21.05.2021
Med Ultrason
2022, Vol. 24, No 1, 65-76 CHEST WALL
Corresponding author: Prof. Dr. med. Christoph F. Dietrich, MBA
Department of Internal Medicine (DAIM) Examination technique
Kliniken Hirslanden Bern, Beau Site,
Salem and Permanence
The superficially located structures of the thoracic
Schänzlihalde 11, 3031 Bern, Switzerland wall close to the US transducer can be optimally exam-
E-mail: c.f.dietrich@googlemail.com 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)
Thymus
oscopy in assessing diaphragmatic movement disorders • The anterior central view from the subxiphoid
[18]. The diaphragm has a crucial role in breathing. Mul- window is recommended for the left and right
tiple methods for evaluation were developed including hemidiaphragm. This view allows the comparison
fluoroscopic sniff test, electromyography and nerve con- between the two hemidiaphragms and their move-
duction, which are used in detection of diaphragmatic ments, using the sectorial or a micro convex probe,
dysfunction. However, these are time-consuming and in B mode, transversely positioned transabdomi-
procedure-related complications might occur [19]. US nal, cranially upright oriented [19].
evaluation of the diaphragm has proved to be useful in • The lateral views, in the so-called apposition area,
patients with suspicion of diaphragm paralysis for dec- between the axillar anterior line and mid-clavicu-
ades. From diaphragm paralysis [20], to the evaluation lar line, with the linear high frequency transducer,
of respiratory function [21], diaphragm US has been con- positioned longitudinally, perpendicular to the
firmed to be a valuable technique, useful in the diagnosis lower ribs;
of different signs and symptoms like paradoxical respi- • The right hemidiaphragm is easily seen through
ration, dyspnoea or asymmetric radiographic findings. the liver window, while the left diaphragm is not
Other conditions that can be diagnosed by US include easy through the spleen window, because of stom-
diaphragmatic hernias, eventrations, peridiaphragmatic ach [23] or other abdominal content [20].
abscess, thoracic tumours, pleural collections and lesions A good image may be obtained between the midclav-
causing diaphragm paralysis [17]. icular and anterior axillary lines in the subcostal area and
Examination technique by directing the probe medially and cranially to achieve
Diaphragmatic US is performed ideally with the pa- the best view of the right hemidiaphragm at the apposi-
tient in the supine position or in a sitting position if the tion zone [24]. For the posterior view, in the sitting posi-
former is not possible, e.g., in children with a neuromus- tion, using a convex probe, the renal parenchyma may
cular disorder. conduct the US to enable the diaphragm to be assessed in
Lateral parts of the diaphragm can be imaged in an children, but the window might be smaller and not pos-
axillary longitudinal section and appear similar to mus- sible in every child as it is thoracic wall size dependent
cles in being generally hypoechoic. Above the spleen and [17].
the liver, the diaphragm appears as a thin hypoechoic The intercostal approach can be used but the inter-
three-layered muscular structure when using high fre- costal space is different in children, compared to adults.
quency transducers. The interface between air-filled tis- However, the preferred intercostal spaces in adults be-
sue (lung) and solid organs (liver, spleen) is generated tween 7th-8th rib or 8-9th rib even 9-10th are potentially
by changes in the acoustic impedance and appears as a applicable in supine children [17]. The estimation of the
bright echo line using low frequency transducers. This lateral diaphragm parts is important, because of their
so-called bright diaphragm line does not represent the main role in breathing compared to the anterior central
diaphragm itself but an interface. The position of the zone which has less effect on inspiration [25]. Diaphrag-
diaphragm and respiratory diaphragmatic movement are matic US has a few limitations: firstly, standard reference
the focus of the evaluation. The best examination posi- parameters such as thickness are only available in some
tion is the sub-diaphragmatic longitudinal view along the populations; secondly, it is difficulty to visualize the left
anterior axillary line. Comparison between the left and diaphragm in overweight children. Lastly, assessment of
right side can be performed by tilting the transducer to the diaphragm motion can be affected by inspiratory ef-
the cranial sub sternal cross-section. In this case, the right fort and interposition of abdominal organs [17].
side of the diaphragm can be very well seen while show- Diaphragm parameters
ing the left side can be challenging due to the air-filled The measurable diaphragm parameters are: diaphrag-
stomach. Because of the necessary depth of penetration, matic thickness excursion and thickening fraction. The
low-frequency convex transducers (3- 9 MHz) should be first is a static parameter while the others are dynamic
used. By using M-mode, the mobility of the diaphragm parameters.
can be shown and measured. The diaphragm thickness (DT) is increased in inspi-
Deep breathing might be difficult in non-compliant ration and decreases in expiration. DT is well correlated
children [17]. The linear, high frequency transducers (7- with body weight in children and adults. Other param-
17 MHz) are primarily used for the diaphragmatic muscle eters like trans diaphragmatic pressure is higher in chil-
assessment and the convex probe (micro convex for chil- dren compared to adults [26]. Normal DT value scales
dren and thin patients) or a sectorial for smaller children are available in healthy adults [27] and children [26] but
[22]: only in limited populations. Atrophy of the diaphragm
Med Ultrason 2022; 24(1): 65-76 71
occurs in ventilated children with acute respiratory fail- US and it has been shown to be superior to fluoroscopy
ure, especially in those receiving neuromuscular blockers [39].
[28]. DT is a very important parameter for detection of Specific pathological findings
diaphragmatic atrophy of mechanically ventilated chil- Diaphragmatic conditions range from congenital mal-
dren [29] and is a reliable prognostic indicator for venti- formations such as diaphragmatic hernia to diaphragm
lator-induced diaphragmatic atrophy [29] and for the risk paralysis secondary to phrenic nerve injury in cardiac
for extubation failure in children [30]. US can be used surgery or muscular dystrophies and neuromuscular dis-
for diagnosing atrophy secondary to other diseases such orders [40]. Detecting traumatic diaphragm rupture using
as myodystrophy, malnutrition, trauma, compression US has been suggested as an additional examination in
and associated breathing disorders with restrictive dys- “Focused Abdominal Sonography for Trauma” (FAST)
function, malnutrition and electrolyte disturbances (hy- examination for adults and children [41,42].
pophosphatemia, hypokalaemia, hypocalcaemia) [31]. In diaphragmatic impairment, lung function assessed
Diaphragmatic excursion (DE) is assessed by M- by spirometry reveals a restrictive pattern, but spirom-
mode US, measuring the distance of diaphragm move- etry is challenging to perform in children, particularly in
ment between end inspiration and end expiration [32] and those with neuromuscular diseases. It has been found that
normal reference values exist in small adult populations a significant correlation exists between lung volumes and
[33] and also a few in children [20]. Decreased DE can be diaphragm function in adults [43] and a linear correla-
suggestive of a number of diaphragm disorders, such as tion between diaphragmatic thickening fraction and lung
neuromuscular diseases (including hypocalcaemia, hy- volumes was demonstrated in children [27]. However,
pokalaemia, hypophosphatemia [31]), neuropraxia sec- this correlation was not confirmed in a study evaluating
ondary to induced hypothermia, ventilated patients and the relationship between body plethysmography and dia-
pathology in the vicinity, from lung and pleura (pneumo- phragm respiratory course in adults [44].
nia, pleural effusions) or abdomen (peritonitis). Congenital diaphragmatic disorders include anatomi-
Diaphragm thickening fraction (DTF) is defined cal defects such as Bochdalek and Morgagni congenital
as the fraction of the difference between the thickness- hernia, eventration, diaphragmatic agenesis and func-
at-end inspiration and thickness-at-end expiration / tional disorders including diaphragm paralysis, atrophy,
thickness-at-end-expiration [19]. The increased DTF is posttraumatic and iatrogenic injury during surgery and
strongly correlated with spontaneous breathing fraction neuromuscular diseases, all have important effects in
and can predict weaning success, correlated with an in- childhood [27].
creased percentage of successful extubation [34], even in Congenital diaphragmatic hernia (CDH) is one of
post transplanted patients [35]. the most frequent major congenital anomalies, with a
Diaphragm motility significant postnatal mortality rate due to acute compli-
The evaluation of diaphragm motility is essential cations or development of pulmonary hypertension [45].
when there is a suspicion of diaphragmatic paralysis. This Diaphragmatic hernia can be easily diagnosed with US.
could be performed by chest X-ray, which can demon- Chest X-ray is the first imaging modality postnatally. In
strate elevation of a hemidiaphragm, but does not provide fact, most cases are diagnosed prenatally by foetal US
information on diaphragm motion. CT would be more combined with foetal MRI for risk evaluation (e.g., lung
specific in detecting the structural changes but provide volumetry). CDH can be diagnosed on chest X-ray when
less information on motility and expose the patients to gas containing bowel or stomach are projected over the
increased ionizing radiations. Diaphragm motility might lung fields; if the solid organs such as liver are involved.
be evaluated by MRI which does not involve ionizing US is the method of choice for diagnosis and can differ-
radiation, but it is more time consuming and expensive entiate CDH from pleural effusion in post-operative pa-
and sedation might be required [19,36,37]. tients [22]. US before corrective surgery can accurately
It has been shown that US is a “highly sensitive meth- evaluate the size of the defect in CDH and can provide
od of demonstrating generalised or localised abnormali- anatomical information, which can consequently influ-
ties of diaphragmatic motion” [18] and new data suggest ence the surgical approach [46].
that diaphragm motility is a predictor for effective ex-
tubation in children [38], confirming the importance of US GUIDED/ASSISTED INTERVENTIONS
US for diaphragm evaluation. One of the most impor-
tant indications for diaphragm US is after cardiac sur- As discussed previously, US has an important role in
gery where there is a suspicion of diaphragm paralysis. assessing all thoracic compartments: chest wall, pleura,
The diaphragm paralysis can be promptly evaluated by 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- and microbiologic testing. If there is suspicion that the
ten best and most safely performed using US guidance pleural effusion is not secondary to infection, cytological
or assistance. US guided drainage of chest wall or lung analysis is mandatory.
abscesses aids the microbiological diagnosis, while of- When therapeutic thoracocentesis is performed, the
fering a minimally invasive therapeutic opportunity [3]. drainage tube is placed to one-way suction, ranging from
Accessible mediastinum lymph nodes and solid tumours -5 to -20 cm of water [55].
of the chest wall or lung may be biopsied via US guid- Currently, there is no consensus regarding pleural
ance [3,4,6,7]. US guidance is preferred for peripheral drainage volumes in children. Volumes equal to 10 ml/
lung biopsy for both children and adults [38,47-49]. The kg with a maximum of 1.5 l are generally considered
utility and low radiation burden of US guided inser- safe. Re-expansion pulmonary oedema (RPE) is the most
tion of arterial and venous catheters is well established feared complication of large and rapid pleural effusion
[5,50,51]. In addition, venous malformations of the chest drainage. Still, RPE is rare in children.
wall may benefit from US guided sclerotherapy [6,52]. Intrapleural fibrinolytic agents are used successfully
Diagnostic and therapeutic US guided drainage of pleural for pleural drainage. There are several protocols used
effusion and empyema is well described in the literature for pleural sclerotherapy using either urokinase or tissue
[53,54]. As per recent guidelines, fibrinolytic treatment plasminogen activator [61,62].
is initiated via transthoracic catheter [55]. The use of US Pleuritic pain must be managed with analgesics in or-
guided insertion of small-bore pig-tail catheters appears der to ensure appropriate expansion of the lung after the
safe and efficient both for symptomatic pleural effusion procedure. When the output of the chest tube decreases to
[56] and empyema [57]. Although, US is not usually em- < 1 ml/kg/24 hours, it may be removed [61].
ployed for identifying pneumothorax in some cases it
may be a more accessible and rapid method of diagnosis ARTEFACTS
and treatment [58].
US guided minimally invasive therapeutic approach- Recognition of artefacts is crucial in order to avoid
es are employed for congenital pulmonary airway mal- misinterpretation of the findings and false diagnosis.
formation (CPAM) with sequestration, both in the foetal Physicians should thus familiarise themselves with the
and neonatal period [6,7,59]. In a case series, the authors terminology and nature of these technical phenomena.
describe the occlusion of the feeding vessel in utero via The term “beam width artefact” describes the visu-
interstitial laser, thrombogenic coil embolization and ra- alisation of echoes generated from a point lying outside
dio-frequency ablation [59]. the US beam, inside the imaging plane and represents an
Interventions in the pleural space expression of US lateral resolution. To understand this
US is the most specific and sensitive imaging study artefact, it is important to remember that the US beam is
used to confirm the presence and type of pleural effusion, bow-tie shaped, being narrowed at the level of the focus
as well as the optimal site for tube placement [55]. It can point and then widens reaching a width larger than the
also signal complications such as the presence of septa- probe width. As a result, a reflection originating from a
tion or debris within the pleural space [55]. The size of point lying outside the width of the transducer but within
the pleural effusion, cultures and respiratory compromise the widened width of the beam will be erroneously dis-
guide the decision for drainage of the pleural effusion played inside the field-of-view. In practice, this explains
[60] (Table II). why bright echoes can be falsely displayed within an-
In children, 10F and 12F drainage tubes are consid- echoic structures such as the gallbladder or an anechoic
ered equally effective. When choosing the site of the pleural effusion. The solution to this problem is always to
thoracocentesis, whether diagnostic or therapeutic, one keep the focal point at the level of structure under exami-
should always have in mind the “safe triangle”. Sam- nation, using multiple focal zones or placing the point of
ples of the pleural effusion must be sent for biochemical interest at the centre of transducer. In that way, the beam
Non-linear artefacts or incomplete suppression may be 7. Supakul N, Karmazyn B. Ultrasound of the pediatric chest–
encountered with highly echogenic interfaces or struc- the ins and outs.Semin Ultrasound CT MR 2013;34:274-
tures, which are projected on the contrast-specific image. 285.8.
8. Newman B. Thoracic neoplasms in children. Radiol Clin
These signals should not be perceived as true enhance-
North Am 2011;49:633-664.
ment and can be differentiated from microbubbles signal
9. Milos RI, Moritz T, Bernathova M, et al. Superficial
by comparing the contrast-specific and low-mechanical desmoid tumors: MRI and ultrasound imaging characteris-
index B-mode image prior to contrast administration, as tics. Eur J Radiol 2015;84:2194-2201.
they will appear identical in both images. In the case of 10. Yildiz AE, Elhan AH, Fitoz S. Prevalence and sonographic
lung US, gas bubbles from abscesses can be visualised features of ectopic thyroidal thymus in children: A retro-
as echogenic areas projected on contrast-specific images. spective analysis. J Clin Ultrasound 2018;46:375-379.
When assessing a lung consolidation, it should be kept 11. Yildiz AE, Ceyhan K, Siklar Z, et al. Intrathyroidal Ectopic
in mind that near-field signal loss can be caused by the Thymus in Children: Retrospective Analysis of Grayscale
disruption of microbubbles incurred by the higher acous- and Doppler Sonographic Features. J Ultrasound Med
tic pressure in this part of the imaging plane. Imaging 2015;34:1651-1656.
12. Bang MH, Shin J, Lee KS, Kang MJ. Intrathyroidal ectopic
plane signal loss can be seen when scanning on the same
thymus in children: A benign lesion. Medicine (Baltimore)
plane for a long period of time. The continuous acoustic 2018;97:e0282.
pressure disrupts microbubbles. This mainly interferes 13. Balakrishnan K, Fonacier F, Sood S, Bamji N, Bostwick H,
with liver examinations for the characterization of focal Stringel G. Foregut Duplication Cysts in Children. JSLS
lesions and is less common in the lung where the trans- 2017;21:e2017.00017.
ducer typically sweeps through areas of consolidation or 14. Wanner MR, Marine MB, Dahl JP. Ultrasound diagnosis of
pleural pathology to detect necrosis. High doses of mi- tracheal cartilaginous sleeve in a patient with Pfeiffer syn-
crobubbles cause shadowing similar to ribs, due to the at- drome. Pediatr Radiol 2018;48:1814-1816.
tenuation of beam. This can be observed in intracavitary 15. Diwakar A, Adam RJ, Michalski AS, et al. Sonographic
applications of CEUS, where the solution to be adminis- evidence of abnormal tracheal cartilage ring structure in
cystic fibrosis. Laryngoscope 2015;125:2398-2404.
tered needs to be carefully reconstituted, including only
16. Gollu G, Onat Bermede A, Khanmammadov F, et al. Use of
a drop (0.1 ml) of microbubbles [70,71].
ultrasonography as a noninvasive decisive tool to determine
Conclusion the accurate endotracheal tube size in anesthetized children.
Arch Argent Pediatr 2018;116:172-178.
The series of reviews is intended to summarize and il- 17. Sarwal A, Walker FO, Cartwright MS. Neuromuscular ul-
lustrate current knowledge on thoracic ultrasound in pae- trasound for evaluation of the diaphragm. Muscle Nerve
diatric patients. The reader is warmly welcome to stimu- 2013;47:319-329.
late a discussion and to illustrate the initiated WFUMB 18. Haber K, Asher M, Freimanis AK. Echographic evaluation
Atlas. of diaphragmatic motion in intra-abdominal diseases. Radi-
ology 1975;114:141-144.
Conflict of interest: none 19. Fayssoil A, Behin A, Ogna A, et al. Diaphragm: Pathophys-
iology and Ultrasound Imaging in Neuromuscular Disor-
References ders. J Neuromuscul Dis 2018;5:1-10.
20. Riccabona M, Sorantin E, Ring E. Application of M-mode
1. Jaworska J, Buda N, Ciuca IM, et al. Ultrasound of the sonography to functional evaluation in pediatric patients.
pleura in children, WFUMB review paper. Med Ultrason Eur Radiol 1998;8:1457-1461.
2021. doi: 10.11152/mu-3058. 21. Riccabona M. Ultrasound of the chest in children (mediasti-
2. Dietrich CF, Buda N, Ciuca IM, et al. Lung ultrasound in num excluded). Eur Radiol 2008;18:390-399.
children, WFUMB review paper (part 2). Med Ultrason 22. Trinavarat P, Riccabona M. Potential of ultrasound in the
2021. doi: 10.11152/mu-3059. pediatric chest. Eur J Radiol 2014;83:1507-1518.
3. Mong A, Epelman M, Darge K. Ultrasound of the pediatric 23. Cohen E, Mier A, Heywood P, Murphy K, Boultbee J, Guz
chest. Pediatr Radiology 2012;42:1287-1297. A. Excursion-volume relation of the right hemidiaphragm
4. Joshi P, Vasishta A, Gupta M. Ultrasound of the pediatric measured by ultrasonography and respiratory airflow meas-
chest. Br J Radiol 2019;92:20190058. urements. Thorax 1994;49:885-889.
5. Cox M, Soudack M, Podberesky DJ, Epelman M. Pediat- 24. El-Halaby H, Abdel-Hady H, Alsawah G, Abdelrahman A,
ric chest ultrasound: a practical approach. Pediatr Radiol El-Tahan H. Sonographic Evaluation of Diaphragmatic Ex-
2017;47:1058-1068. cursion and Thickness in Healthy Infants and Children. J
6. Coley BD. Chest sonography in children: current indica- Ultrasound Med 2016;35:167-175.
tions, techniques, and imaging findings. Radiol Clin North 25. Houston JG, Morris AD, Howie CA, Reid JL, McMillan N.
Am 2011;49:825-846. Technical report: quantitative assessment of diaphragmatic
Med Ultrason 2022; 24(1): 65-76 75
movement--a reproducible method using ultrasound. Clin 42. Gangahar R, Doshi D. FAST scan in the diagnosis of acute
Radiol 1992;46:405-407. diaphragmatic rupture. Am J Emerg Med 2010;28:387.e1-3.
26. Rehan VK, McCool FD. Diaphragm dimensions of the 43. Ayoub J, Metge L, Dauzat M, et al. Diaphragm kinet-
healthy term infant. Acta Paediatr 2003;92:1062-1067. ics coupled with spirometry. M-mode ultrasonographic
27. McCool FD, Tzelepis GE. Dysfunction of the diaphragm. N and fluoroscopic study; preliminary results. J Radiol
Engl J Med 2012;366:932-942. 1997;78:563-568.
28. Goligher EC, Laghi F, Detsky ME, et al. Measuring dia- 44. Scott S, Fuld JP, Carter R, McEntegart M, MacFarlane NG.
phragm thickness with ultrasound in mechanically venti- Diaphragm ultrasonography as an alternative to whole-
lated patients: feasibility, reproducibility and validity. In- body plethysmography in pulmonary function testing. J
tensive Care Med 2015;41:642-649. Ultrasound Med 2006;25:225-232.
29. Johnson RW, Ng KWP, Dietz AR, et al. Muscle atrophy in 45. Brownlee EM, Howatson AG, Davis CF, Sabharwal AJ.
mechanically-ventilated critically ill children. PLoS One The hidden mortality of congenital diaphragmatic hernia: a
2018;13:e0207720. 20-year review. J Pediatr Surg 2009;44:317-320.
30. Terhart MN, Hanekom S, Lupton-Smith A, Morrow B. Re- 46. Hattori K, Takamizawa S, Miyake Y, et al. Preoperative so-
liability of ultrasonic diaphragm thickness measurement in nographic evaluation of the defect size and the diaphragm
mechanically ventilated infants and children: A pilot study. rim in congenital diaphragmatic hernia - preliminary expe-
South Afr J Crit Care (Online) 2018;34:22-27. rience. Pediatr Radiol 2018;48:1550-1555.
31. McParland C, Resch EF, Krishnan B, Wang Y, Cujec B, Gal- 47. Jarmakani M, Duguay S, Rust K, Conner K, Wagner JM.
lagher CG. Inspiratory muscle weakness in chronic heart Ultrasound Versus Computed Tomographic Guidance for
failure: role of nutrition and electrolyte status and systemic Percutaneous Biopsy of Chest Lesions. J Ultrasound Med
myopathy. Am J Respir Crit Care Med 1995;151:1101- 2016;35:1865-1872.
1107. 48. Safai Zadeh E, Keber CU, Dietrich CF, et al. Perfusion pat-
32. Turton P, ALAidarous S, Welters I. A narrative review of terns of peripheral pulmonary granulomatous lesions using
diaphragm ultrasound to predict weaning from mechanical contrast-enhanced ultrasound (CEUS) and their correlation
ventilation: where are we and where are we heading? Ultra- with immunohistochemically detected vascularization pat-
sound J 2019;11:2. terns. JUltrasound Med 2021. doi: 10.1002/jum.1573049.
33. Boussuges A, Gole Y, Blanc P. Diaphragmatic motion stud- 49. Safai Zadeh E, Beutel B, Dietrich CF, et al. Perfusion
ied by m-mode ultrasonography: methods, reproducibility, Patterns of Peripheral Pulmonary Lesions in COVID-19
and normal values. Chest 2009;135:391-400. Patients Using Contrast-Enhanced Ultrasound (CEUS):
34. Glau CL, Conlon TW, Himebauch AS, et al. Progressive A Case Series. J Ultrasound Med 2021. doi:10.1002/
Diaphragm Atrophy in Pediatric Acute Respiratory Failure. jum.15624.
Pediatr Crit Care Med 2018;19:406-411. 50. Bajaj M, Wells J, Liyanage A, Evans S, Hamill J. Radiation
35. Sharma A, Karna ST, Tandon M, et al. Use of ultrasound- burden of pediatric ultrasound-guided percutaneous central
guided preoperative diaphragmatic thickness as a predictor venous access devices: A prospective cohort study. J Pedi-
of postoperative weaning failure in recipients and donors atr Surg 2018;53:802-807.
scheduled for living donor liver transplant surgery. Saudi J 51. Jenssen C, Brkljacic B, Hocke M, et al. EFSUMB Guide-
Anaesth 2018;12:406-411. lines on Interventional Ultrasound (INVUS), Part VI - Ul-
36. Caraiani C, Dong Y, Rudd AG, Dietrich CF. Reasons for in- trasound-Guided Vascular Interventions. Ultraschall Med
adequate or incomplete imaging techniques. Med Ultrason 2016;37:473-476.
2018;20:498-507. 52. Kumar S, Bhavana K, Kumar S, Kumar P. Ultrasound-
37. Caraiani C, Petresc B, Dong Y, Dietrich CF. Contraindica- guided polidocanol foam sclerotherapy for treating venous
tions and adverse effects in abdominal imaging. Med Ultra- malformations. J Clin Ultrasound 2018;46:23-31.
son 2019;21:456-463. 53. Corcoran JP, Tazi-Mezalek R, Maldonado F, et al. State of
38. Lee EP, Hsia SH, Hsiao HF, et al. Evaluation of diaphrag- the art thoracic ultrasound: intervention and therapeutics.
matic function in mechanically ventilated children: An ul- Thorax 2017;72:840-849.
trasound study. PLoS One 2017;12:e0183560. 54. Safai Zadeh E, Gorg C, Dietrich CF, Gorlach J, Alhyari A,
39. Sanchez de Toledo J, Munoz R, Landsittel D, et al. Diag- Trenker C. Contrast-Enhanced Ultrasound for Evaluation
nosis of abnormal diaphragm motion after cardiothoracic of Pleural Effusion: A Pictorial Essay. J Ultrasound Med
surgery: ultrasound performed by a cardiac intensivist vs. 2021. doi:10.1002/jum.15705.
fluoroscopy. Congenit Heart Dis 2010;5:565-572. 55. Feola GP, Hogan MJ, Baskin KM, et al. Quality Improve-
40. Dube BP, Dres M. Diaphragm Dysfunction: Diagnos- ment Standards for the Treatment of Pediatric Empyema. J
tic Approaches and Management Strategies. J Clin Med Vasc Interv Radiol 2018;29:1415-1422.
2016;5:113. 56. Miraglia R, Maruzzelli L, Piazza M, et al. Real-time ultra-
41. Blaivas M, Brannam L, Hawkins M, Lyon M, Sriram K. sound-guided placement of a pigtail catheter in supine posi-
Bedside emergency ultrasonographic diagnosis of dia- tion for draining pleural effusion in pediatric patients who
phragmatic rupture in blunt abdominal trauma. Am J Emerg have undergone liver transplantation. J Clin Ultrasound
Med 2004;22:601-604. 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 63. Feldman MK, Katyal S, Blackwood MS. US artifacts. Ra-
ultrasound-guided pigtail catheter chest drain for compli- diographics 2009;29:1179-1189.
cated parapneumonic effusion and empyema in children - 64. Baad M, Lu ZF, Reiser I, Paushter D. Clinical Significance
16-year, single-centre experience of radiologically placed of US Artifacts. Radiographics 2017;37:1408-1423.
drains. Pediatr Radiol 2018;48:1410-1416. 65. Lichtenstein D, Meziere G, Biderman P, Gepner A, Barre
58. Migliaro F, Sodano A, Capasso L, Raimondi F. Lung O. The comet-tail artifact. An ultrasound sign of alve-
ultrasound-guided emergency pneumothorax nee- olar-interstitial syndrome. Am J Respir Crit Care Med
dle aspiration in a very preterm infant. BMJ Case Rep 1997;156:1640-1646.
2014;2014:bcr2014206803. 66. Stefanidis K, Dimopoulos S, Kolofousi C, et al. Sonograph-
59. Baud D, Windrim R, Kachura JR, et al. Minimally inva- ic lobe localization of alveolar-interstitial syndrome in the
sive fetal therapy for hydropic lung masses: three different critically ill. Crit Care Res Pract 2012;2012:179719.
approaches and review of the literature. Ultrasound Obstet 67. Dietrich CF, Mathis G, Blaivas M, et al. Lung artefacts and
Gynecol 2013;42:440-448. their use. Med Ultrason 2016;18:488-499.
60. Bradley JS, Byington CL, Shah SS, et al. Executive sum- 68. Dietrich CF, Mathis G, Blaivas M, et al. Lung B-line arte-
mary: the management of community-acquired pneumonia facts and their use. J Thorac Dis 2016;8:1356-1365.
in infants and children older than 3 months of age: clinical 69. Sidhu PS, Cantisani V, Dietrich CF, et al. The EFSUMB
practice guidelines by the Pediatric Infectious Diseases So- Guidelines and Recommendations for the Clinical Practice
ciety and the Infectious Diseases Society of America. Clin of Contrast-Enhanced Ultrasound (CEUS) in Non-Hepatic
Infect Dis 2011;53:617-630. Applications: Update 2017 (Short Version). Ultraschall
61. Sonnappa S, Cohen G, Owens CM, et al. Comparison of Med 2018;39:154-180.
urokinase and video-assisted thoracoscopic surgery for 70. Fetzer DT, Rafailidis V, Peterson C, Grant EG, Sidhu P,
treatment of childhood empyema. Am J Respir Crit Care Barr RG. Artifacts in contrast-enhanced ultrasound: a pic-
Med 2006;174:221-227. torial essay. Abdom Radiol (NY) 2018;43:977-997.
62. Hawkins JA, Scaife ES, Hillman ND, Feola GP. Current 71. Dietrich CF, Ignee A, Hocke M, Schreiber-Dietrich D,
treatment of pediatric empyema. Semin Thorac Cardiovasc Greis C. Pitfalls and artefacts using contrast enhanced ul-
Surg 2004;16:196-200. trasound. Z Gastroenterol 2011;49:350-356.
Review Med Ultrason 2022, Vol. 24, no. 1, 77-84
DOI: 10.11152/mu-2961
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
in clinical stage I endometrial carcinoma to aid the deci- mensional” were used as a search terms. No other meth-
sion of whether lymphadenectomy should be performed odological filters in the database were included to avoid
or not. Frozen section has shown to be highly accurate possible omission of relevant studies, according to the
(94%) for determining myometrial invasion [8]. Howev- recommendation of Leeflang et al [16].
er, it may be time consuming and it is not performed in all Study selection and data collection
hospitals. Therefore, a method that could reliably assess Two authors (TC and RB) screened the titles iden-
myometrial invasion preoperatively would be helpful to tified by the searches to exclude irrelevant articles, not
provide individual tailoring of the surgical approach [9]. strictly related to the topic. Abstracts of these articles
There is no consensus about the optimal imaging tech- were revised and some of these articles were excluded
nique for evaluation of myometrial and cervical invasion. due to no relevant content (not relevant topic) or non-
Two-dimensional transvaginal ultrasound (2D-TVS) and English languages. Full text publications were indepen-
magnetic resonance imaging (MRI) have shown accura- dently analysed by two authors (TC and RB) who read
cies of 69-74% and 66-90% respectively [10-13]. Three- and collected data of all the publications.
dimensional transvaginal ultrasonography (3D-TVS) has Inclusion criteria were: 1) Use of 3D-TVS as index
several advantages when evaluating myometrial infiltra- test to assessment myometrial invasion preoperatively
tion [14]. However, although several studies have shown with subjective impression estimation of myometrial in-
the utility of this technique in myometrial evaluation of vasion (less than 50% of myometrium or more) by the
EC, the routine implementation in clinical practice has investigator/s; 2) Reference standard was surgical patho-
not been stablished so far. Theoretically, from an efficient logical data; 3) Presence of results sufficient to construct
point of view, the assessment of pre-surgical staging with the 2x2 table of diagnostic performance as minimum
3D-TVS reveals a notorious benefit in terms of economic data requirement. To avoid inclusion of duplicate cohorts
approach as compared to MRI. Since the advent of 3D- from at least two pair of studies reported by the same
TVS, several studies have been published aiming at eval- authors, the study period of each study was examined. If
uating the role of this technique for detecting the depth of dates overlapped, the latest study published or the study
myometrial infiltration in endometrial cancer. that could provide us a 2x2 table was chosen. Disagree-
On the other hand, more hospitals, especially in de- ments arising during the process of study selection and
veloped countries, tend to implement 3D-TVS in their data collection were resolved by consensus among three
daily practices, revealing an increasing need of training authors (JLA, TC and RB)
and investigation of the actual performance of this tech- The PICOS (Patients, Intervention, Comparator,
nique. To the best of our knowledge, not a single meta- Outcome, Study design) criteria used for inclusion and
analysis analysed the diagnostic performance of 3D-TVS exclusion of studies are shown in Table I. Diagnostic
for detecting myometrial invasion in women with EC. accuracy and additional useful information on patients
Thus, the purpose of this systematic review and and procedures were retrieved from a selected primary
meta-analysis is to evaluate the diagnostic accuracy of study written independently by the same authors (JLA,
3D-TVS in the preoperative detection of deep myome- TC and RB).
trial invasion in patients with endometrial cancer, using Risk of bias in individual studies
definitive frozen section diagnosis after surgery as the Quality assessment was conducted using the tool pro-
reference standard. vided by QUADAS (Quality Assessment of Diagnostic
Accuracy Studies –2). The QUADAS-2 format includes
Material and methods four domains: 1) patient selection, 2) index test, 3) refer-
ence standard, 4) flow and timing. For each domain, the
Protocol and registration risk of bias and concerns regarding applicability was ana-
This systematic review and meta-analysis has been lysed and rated as low, high and unclear risk. The qual-
made according to the PRISMA Statement [15]. All ity assessment was used to provide an evaluation of the
methods for inclusion and exclusion criteria, data extrac- overall quality of the studies and to investigate potential
tion and quality assessment were specified in advance. sources of heterogeneity.
The protocol was not registered. Three authors (JLA, TC and RB) evaluated the meth-
Data sources and searches odological quality independently. Disagreements were
Two electronic databases, PubMed/Medline and Web solved by discussion among these authors. The assess-
of Science were screened in November 2020 by three of ment of quality was based on whether the study described
the authors (TC, RB and JLA) to identify eligible stud- the study´s design as well as inclusion and exclusion cri-
ies. “Endometrial cancer”, “ultrasound” and “three-di- teria for the patient selection domain. The evaluation of
Med Ultrason 2022; 24(1): 77-84 79
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- Fig 1. Flow chart showing studies selection process.
plored, drawing forest plots of sensitivity and specificity.
We then formally assessed the presence of heterogeneity papers, 12 were excluded after reading abstract (not rel-
for sensitivity and specificity using Cochran´s Q test and evant to the review) and one due to other language than
the I2 Index. A test for heterogeneity examines the null English (Chinese n=1). The full text of the remaining
hypothesis that all studies are evaluating the same effect twelve articles were examined. One of them [18] was
(Higgins et al [17]). excluded due to the impossibility of obtain 2x2 tables.
Cochran´s Q statistic is computed by summing the Alcázar et al [19] and Mascilini et al [20] analysed myo-
squared deviations of each study´s estimate from the metrial invasion with TDS and tumour/uterine volume
overall meta-analytic estimate, weighting each study´s 3D ratio without subjective impression as a method. For
contribution in the same manner as in the meta-analysis. that reason, they were also discarded.
A p-value <0.1 indicates heterogeneity. The I2 index de- Nine studies were finally taken into account
scribes the percentage of total variation across studies [9,14,18,21-26] (Table I). All studies analysed myome-
that is due to heterogeneity rather than chance. Accord- trial invasion with “subjective impression method”. No
ing to Higgins et al values of 25%, 50%, and 75% would additional relevant studies were found from references
be considered to indicate low, moderate and high hetero- cited in the papers included in the review. A flowchart
geneity, respectively [17]. summarizing literature identification and selection is
Summary receiver-operating characteristics (sROC) given in Figure 1.
curves for each approach were plotted to illustrate the Characteristics of included studies
relationship between sensitivity and specificity. The nine studies [9,14,18,21-26] published between
All analyses were performed using Meta- analytical 2009 and 2019, reported data from 581 patients, 231
integration of Diagnostic Accuracy Studies (MIDAS) women having myometrial invasion >50% in the defini-
and METANDI commands in STATA version 12 for Win- tive pathological samples. This was considered as the
dows (Stata Corporation, College Station, TX, USA). p- prevalence and, therefore, pre-test probability.
value <0.05 was considered statistically significant. The mean patient age was reported in 8 out of the 9
studies, in Trujillo et al no reference to this item being
Results made [22]. Patients’ mean age was 61.4 years, ponderated
by the number of patients in each study. The prevalence
Search results of ≥ 50% myometrial invasion was 39.8%. Hormonal
The electronic search provided 108 citations. A flow- status differentiating between pre- and postmenopausal
chart summarizing literature identification and selection women was given in 3 out of 9 studies [14,21,26]. The
is given in Figure 1. Eighty-three of them were excluded differentiation between histological grades was referred
due to no relevant topic (i.e. 3D MRI) or being review 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 in spite that this technique is relevant especially for early
being considered high-risk profile. All studies described stages since true tumour stage can only be determinated
data regarding tumour histological type (endometroid after surgery. One study excluded patients with leiomio-
and non-endometroid) except for Christensen et al [24]. mas greater than 3 cm and submucous myomas [21]. This
As mentioned before, in all studies selected for quan- study was considered as a high risk for bias in patient se-
titative analysis, method to estimate myometrial invasion lection domain since selecting exclusively ideal patients
was the subjective impression on 3D-TVS by examiner/s. for the validation of the 3D-TVS, can suppose a bias
Most of the publications studied in this meta-analysis in the validity of the test. Two studies [9,21] excluded
make a comparison between subjective impression with patients whose videos are incomplete and one study ex-
3D-TVS and MRI [18,22,24-26]. Two papers also com- cludes patients with suboptimal image resolution (classi-
pared subjective impression with 3D-TVS and objective fied by score 3 of 4 of quality) [14]. Incomplete videos is
methods with 3D-TVS [9,21]. Another paper compares understandably criteria for exclusion, but if only optimal
2D-TVS and 3D-TVS [23]. Only one publication uses the image resolution is selected, the scenario left is made
subjective impression of 3D-TVS as the only method as- up of exclusively ideal situations and this can lead to an
sessed [14]. Pathological evaluation of the removed uter- overestimation of the accuracy of the test.
us was considered as a reference standard in all papers. Regarding the study design, prospective design with
Methodological quality of included studies US prior to surgery is carried out in 5 of the 9 studies
A graphical display of the risk evaluation of bias and [9,14,18,21,26]. The rest of papers describe retrospec-
concerns regarding applicability of the selected studies, tive methodology. Christensen et al [24] evaluated the
according to predefined criteria, is shown in figure 2. volume of the uterus 6 months after surgery, the US ex-
Regarding risk of bias and the domain patient selec- aminers being blinded to the pathological result. In 2 of
tion, all of them explain specifically patterns of inclusion. the retrospective studies the examiner was blinded to the
Three studies do not include advanced stage patients pathology [23,24] and in 2 studies no explanation was
[14,22,24]. Christensen et al [24] included endometrial provided [22,25].
hyperplasia with atypia in preoperative histology into the Concerning the domain index test, all studies describe
group of patients studied. This last case can increase the clearly the index test as well as how it was performed
risk of overestimating the accuracy of 3D-TVS for the and interpreted. However, Christensen et al [24], describ-
estimation of <50% of MI. In contrast, the exclusion of ing a 4-step subjective methodology (1- initial subjec-
advanced stage cases was not considered a risk of bias, 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
Discussion
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
adverse events and is much more cost-efficient [5-8]. Inclusion and exclusion criteria
The evaluation of the LNs during endobronchial ultra- Two reviewers (JW and YS) screened the titles and
sound (EBUS) according to conventional B-mode fea- the abstracts of retrieved studies independently. Before
tures such as size, shape, echogenicity, distinct border, identifying the literature, they established the inclusion
central hilar structure and coagulation necrosis sign, has and exclusion criteria together to increase validity and
been found helpful in the detection of LNs metastasis [9]. reproducibility. Inconsistencies between the reviewers
Furthermore, studies demonstrated that vascular pattern were resolved through discussion.
on power Doppler mode during EBUS can also predict The inclusion criteria were as follows: (1) diagnostic
nodal metastasis in patients with lung cancer. However, it studies were included; (2) studies evaluating the diag-
is still difficult for certain cases to predict LNs metastasis nostic accuracy of EBUS elastography in distinguishing
only by conventional B-mode features [10,11]. malignant and benign hilar and mediastinal LNs were in-
Ultrasound elastography exhibits potential applica- cluded; (3) a reference standard was adopted to confirm
tion to the differential diagnosis of benign and malignant malignant intrathoracic LNs, such as cytology obtained
LNs [12,13]. In recent years, EBUS elastography has by EBUS-TBNA or other method, histology of surgical
been introduced as a novel modality in the evaluation of resection, or more than 6 months of follow-up; (4) if sev-
hilar and mediastinal LNs [14-17]. However, the diag- eral diagnostic methods were used in a study, only the
nostic performance of EBUS elastography for the differ- best result was chosen.
entiation of benign and malignant hilar and mediastinal The exclusion criteria were as follows: (1) case re-
LNs is variable across previous studies, with the sensitiv- ports, letters, consensus statements, and unpublished
ity ranging from 64% to 100% and the specificity ranging articles; (2) studies without sufficient data to construct
from 65% to 92% [14-16]. diagnostic 2x2 tables; (3) studies that contained an over-
Whether EBUS elastography can be considered as lapped population.
a valuable scanning modality is still a controversial is- Data extraction
sue. Therefore, we performed a meta-analysis to assess Two reviewers (JW and YS) independently extracted
the diagnostic performance of EBUS elastography in the the relevant data from the eligible studies using a pre-
noninvasive discrimination between benign and malig- designed data collection form. Any discrepancies were
nant hilar and mediastinal LNs. resolved by discussion with the senior author. For eligi-
ble studies, the following items were extracted: last name
Material and methods of the first author, year of publication, country, study
type, sample method, malignancy prevalence, diagnostic
Meta-analysis principles method, blinding method, cut off, ultrasound equipment,
The present meta-analysis was carried out on the basis probe frequency, sample size, lymph node, short-axis di-
of the Preferred Reporting Items for Systematic Reviews ameter, mean age, gender, standard reference, time be-
and Meta-Analyses (PRISMA) guidelines, which include tween EBUS elastography and the standard reference.
27 items and provide specific guidance for the reporting Study quality assessment
of systematic reviews [18]. We registered our protocol The Quality Assessment of Diagnostic Accuracy
with the International Platform of Registered Systematic Studies-2 (QUADAS-2) tool [19] was used to assess the
Review and Meta-analysis Protocols (INPLAY) (regis- risk of bias and methodological quality. The quality of
tration number: INPLASY2020110117). each eligible study was assessed by an appraisal of the
Search strategy risk of bias of four domains and clinical applicability of
The Pubmed, EMBASE and Cochrane Library were three domains of the study characteristics. Four domains
systematically searched from inception to November included patient selection, index test, reference standard
2020, to identify English-language studies on EBUS elas- and flow and timing. Every domain was assessed for risk
tography for differentiating benign and malignant LNs. of bias, and the first three domains were assessed for ap-
The search terms were as follows: “elasticity”, “elasto- plicability. The quality assessment was carried out using
grams”, “elastography”, “elastosonography”, “elasticity RevMan 5.3 software (Nordic Cochrane Centre, Copen-
imaging”, “endobronchial ultrasound”, “EBUS”, “EBUS hagen, Denmark).
elastography”, “lymph node”, “lymph nodes”, “lymphad- Statistical analysis
enopathy”, and “lymphaden”. Detailed search terms are This meta-analysis was performed by StataSE 15
provided in supplementary file 1. Reference lists of the (Stata Corporation, College Station, Texas). All statisti-
included studies and relevant reviews were also screened cal analyses were performed by one author (JW), who
to find additional relevant studies. 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
erence standard domain, 4 studies [16,26,30,33] were dom effects method on the basis of significant statistical
considered as “unknown” because the blinded status of heterogeneity (I2=89.56% for sensitivity; I2=69.31% for
index test was not definitely reported. With regard to specificity). Overall, the pooled sensitivity and specific-
the flow and timing domain, 6 studies were considered ity of EBUS elastography were 0.90 (95% CI, 0.84-0.94)
as having high bias because they did not enroll all the and 0.78 (95% CI, 0.74-0.81; fig 3). The pooled PLR,
patients for analysis [14,15,28,30,32,36]. Regarding ap- NLR, and DOR of EBUS elastography were 4.1 (95%
plicability, for patient selection, index test and reference CI, 3.4-4.9), 0.12 (95% CI, 0.07-0.21), and 33 (95% CI,
standard domains, all studies were considered to have 17-64), respectively and the post-test probability was
low concerns. 50% and 3% (fig 4). The AUC under the SROC curve for
Data synthesis and publication bias the value of EBUS elastography in the diagnosis of LNs
Summaries of diagnostic sensitivity and specificity of was 0.86 (95% CI, 0.82-0.88; fig 5).
EBUS elastography for differentiating malignant and be- The Deeks’ funnel plot was carried out to evaluate
nign hilar and mediastinal LNs were analysed by the ran- 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 sampling (consecutive versus others). Among the various
existed (p=0.00). This indicated publication bias might potential covariates, study design and year of publication
be part of source of heterogeneity. were associated with the heterogeneity of the sensitivity,
Meta-regression and subgroup analyses while study design, year of publication, reference stand-
Due to the significant heterogeneity among studies, ard, diagnostic method, and sampling were associated
a meta-regression analysis was performed to explore with the heterogeneity of the specificity (fig 7).
other potential sources of heterogeneity. The covariables The subgroups with respect to study design, prospec-
included study design (prospective versus others), year tive studies had a lower diagnostic performance to others
of publication (2014-2017 versus 2018-2020), reference (sensitivity: 0.89 and 0.92, p=0.04; specificity: 0.73 and
standard (pathology versus pathology or follow-up), di- 0.83, p=0.00). The subgroups with regard to year of pub-
agnostic method (quantitative versus qualitative), and 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
EBUS elastography used qualitative diagnostic meth- ity of the stiffer area comprised more than 31% of the
ods or quantitative diagnostic methods for diagnosing entire lymph node area for diagnosing malignancy were
hilar and mediastinal LNs in eligible studies. Qualita- 72.1% and 84%, and by adding sonographic features to
tive diagnostic methods in enrolled studies included the elastographic findings, the sensitivity and specificity
the 3-, 4-, or 5-image pattern classifications and quan- improved to 93.7% and 89.4%. A retrospective study of
titative diagnostic methods included the stiffness area 149 mediastinal and hilar LNs by Uchimura et al [35]
ratio, the strain histogram method, and the strain ratio. reported that by adding B-mode sonography to EBUS
Generally, unlike qualitative diagnostic methods, quan- elastography, the sensitivity increased to 98.3%. There-
titative diagnostic methods could avoid the disadvan- fore, the combination of EBUS elastography and B-mode
tage relative to subjective evaluations by using specific EBUS may lead to a higher diagnostic accuracy than
numerical values [16,33]. In the subgroup analysis, 9 either technology alone for differentiating benign and
studies [16,23-26,29,33-35] using quantitative diagnos- malignant LNs. In clinical practice, the procedure of B-
tic methods showed pooled sensitivity of 0.89 (95% CI, mode EBUS is always before EBUS elastography, so the
0.81-0.96) and specificity of 0.76 (95% CI, 0.71-0.81); combination of EBUS elastography and B-mode EBUS
in comparison, the pooled sensitivity of the 8 stud- for assessing the LNs is more feasible.
ies [14,15,27,28,30-32,36] using qualitative diagnostic As the significant heterogeneity in the present study
methods was 0.92 (95% CI, 0.85-0.98) and the specific- (sensitivity: I2=89.56%, p=0.00; specificity: I2=69.31%,
ity was 0.80 (95% CI, 0.75-0.86). A higher specificity of p=0.00), meta-regression analyses were carried out to ex-
qualitative diagnostic methods was observed (p=0.00); plore the sources of heterogeneity. We found that study
in contrast, the sensitivities were comparative (p=0.05). design and year of publication accounted for part of the
However, in three studies using qualitative diagnostic significant sources of heterogeneity in terms of sensitiv-
methods [14,15,28,30,32,36], the comparative analysis ity and study design, year of publication, reference stand-
was performed between LNs with colour type 1 (not pre- ard, diagnostic method and sampling in terms of specific-
dominantly blue) and 3 (predominantly blue) and LNs ity. Furthermore, the indicated publication bias might be
with a mixed colour (type 2) were excluded. This might another source of heterogeneity. On the other hand, there
have resulted in patient selection bias. On the other hand, were other factors such as specialties of EBUS perform-
qualitative methods are operator dependent and may ers, different levels of experience and different equip-
not provide an objective diagnosis. Therefore, the result ment, which might also have played a role in significant
should be interpreted with caution and large prospective heterogeneity among studies. Further meta-regression
studies are still required to verify the present result. analyses was not carried out to explore the sources of
Ultrasound elastography is mainly used to detect the heterogeneity on the basis of the factors referred above
tissue stiffness [38,39]. Several previous studies found because of the insufficient information supplied in the
out that some malignant LNs were very soft, while some eligible studies.
benign LNs appeared quite hard [26,32,40,41]. Possible EBUS elastography cannot currently replace the more
explanations for the findings may lie in the high vascular accurate EBUS-TBNA for differentiating hilar and medi-
invasion or necrotic structure of malignant LNs that ap- astinal LNs, which has been considered as the first-line
pear soft under elastographic evaluation. By comparison, technology for hilar and mediastinal lymph node staging
nonspecific inflammation, having areas of hard fibrotic or of lung cancer in many clinic guidelines [42,43]. How-
anthracotic tissues frequently occur in benign LNs and, ever, EBUS elastography as a supplemental modality
therefore, the strain within the node may increase as the could be helpful for operators to perform EBUS-TBNA
“stiffer” components of the node are assessed, ultimately efficiently. It could help in selecting the most suspicious
influencing the application of elastography techniques. lymph node for biopsy in stations where multiple LNs
This confirms again that EBUS elastography reveals tis- are found and help operators to choose harder areas of
sue stiffness and not malignancy or benignity. LNs for fine needle aspiration (FNA) to avoid necrosis
Studies that compared conventional EBUS imaging or blood vessels, then improving the quality of the speci-
to EBUS elastography have shown that EBUS elastogra- mens and decreasing the number of punctures [14,23,24].
phy is superior to imaging by EBUS alone [14,23,25,30]. It is important to consider several limitations regard-
However, a prospective study by Gu et al [15] showed ing this study. First, only studies written in English were
that EBUS elastography in combination with conven- enrolled, which may result in potential selection bias.
tional EBUS B-mode features could improve the speci- Second, only several included studies [16,26,27,32,36]
ficity from 65% to 72.7%. The study by Fujiwara et al evaluated the intra- or inter-observer variability. There-
[17] also demonstrated that the sensitivity and specific- fore, further studies are needed to evaluate the variabil-
Med Ultrason 2022; 24(1): 85-94 93
ity. Third, the diagnostic accuracy of EBUS elastogra- chial needle aspiration compared with mediastinoscopy for
phy was acquired from enrolled studies which provided mediastinal lymph node staging of lung cancer. J Thorac
various elastic parameters with diverse threshold values, Cardiovasc Surg 2011;142:1393-400.e1.
8. Verdial FC, Berfield KS, Wood DE, et al. Safety and costs
which may result in heterogeneity. Therefore, the clas-
of endobronchial ultrasound-guided nodal aspiration and
sification of LNs by EBUS elastography still needs to be
mediastinoscopy. Chest 2020;157:686-693.
further explored, developed and improved because of the 9. Hylton DA, Turner J, Shargall Y, et al. Ultrasonographic
lack of uniform standards. Finally, most of the enrolled characteristics of lymph nodes as predictors of malignancy
studies had methodological disadvantages, especially in during endobronchial ultrasound (EBUS): a systematic re-
domains such as patient selection, the index test, refer- view. Lung Cancer 2018;126:97-105.
ence standard and flow and timing, so improvements in 10. Darwiche K, Özkan F, Wolters C, Eisenmann S. Endobron-
the future study design are needed to address the issue. chial ultrasound (EBUS) - update 2017. Ultraschall Med
2018;39:14-38.
Conclusion 11. Agrawal S, Goel AD, Gupta N, Lohiya A, Gonuguntla HK.
Diagnostic utility of endobronchial ultrasound (EBUS)
features in differentiating malignant and benign lymph
In summary, EBUS elastography is a valuable tech-
nodes - a systematic review and meta-analysis. Respir Med
nology in the differentiation of benign and malignant hilar 2020;171:106097.
and mediastinal LNs with high sensitivity and moderate 12. Ophir J, Alam SK, Garra B, et al. Elastography: ultrason-
specificity, which may provide supplementary diagnostic ic estimation and imaging of the elastic properties of tis-
information, increase the diagnostic yield, and reduce the sues. Proc Inst Mech Eng H 1999;213:203-233.
number of unnecessary biopsies during EBUS-TBNA. 13. Carlsen J, Ewertsen C, Sletting S, et al. Ultrasound elas-
Furthermore, the combination of EBUS elastography and tography in breast cancer diagnosis. Ultraschall Med
B-mode EBUS could improve the diagnostic accuracy 2015;36:550-562.
for hilar and mediastinal LNs. However, the conclusion 14. Izumo T, Sasada S, Chavez C, Matsumoto Y, Tsuchida T.
of this study should be interpreted with caution because Endobronchial ultrasound elastography in the diagnosis
of mediastinal and hilar lymph nodes. Jpn J Clin Oncol
of the significant heterogeneity and the publication bias.
2014;44:956-962.
Large prospective international multicentre studies are
15. Gu Y, Shi H, Su C, et al. The role of endobronchial ultra-
still required to support the present conclusion. sound elastography in the diagnosis of mediastinal and hilar
lymph nodes. Oncotarget 2017;8:89194-89202.
Conflict of interest: none 16. Verhoeven RLJ, Trisolini R, Leoncini F, et al. Predictive
value of endobronchial ultrasound strain elastography in
References mediastinal lymph node staging: the e-predict multicenter
study results. Respiration 2020;99:484-492.
1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA 17. Fujiwara T, Nakajima T, Inage T, et al. The combination of
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 ing endobronchial ultrasound-guided transbronchial needle
tumor-infiltrating lymphocytes in development, progres- aspiration for predicting nodal metastasis. Thorac Cancer
sion, and prognosis of non-small cell lung cancer. J Thorac 2019;10:2000-2005.
Oncol 2016;11:789-800. 18. McInnes MDF, Moher D, Thombs BD, et al. Preferred re-
3. Woodard GA, Jones KD, Jablons DM. Lung cancer staging porting items for a systematic review and meta-analysis of
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. ment. JAMA 2018;319:388-396.
Endobronchial ultrasound-transbronchial needle aspiration 19. Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2:
(EBUS/TBNA): a diagnostic challenge for mediastinal le- a revised tool for the quality assessment of diagnostic ac-
sions. Ann Transl Med 2018;6:92. curacy studies. Ann Intern Med 2011;155:529-536.
5. Naur TMH, Nilsson PM, Pietersen PI, Clementsen PF, 20. Hanley JA, McNeil BJ. The meaning and use of the area
Konge L. Simulation-based training in flexible bronchos- under a receiver operating characteristic (ROC) curve. Ra-
copy and endobronchial ultrasound-guided transbron- diology 1982;143:29-36.
chial needle aspiration (EBUS-TBNA): a systematic re- 21. Deeks JJ. Systematic reviews in health care: 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- 2001;323:157-162.
nal staging procedures in lung cancer: EBUS, TBNA and 22. Deeks JJ, Macaskill P, Irwig L. The performance of tests
mediastinoscopy. Curr Opin Pulm Med 2009;15:334-342. of publication bias and other sample size effects in system-
7. Yasufuku K, Pierre A, Darling G, et al. A prospective con- atic reviews of diagnostic test accuracy was assessed. J Clin
trolled trial of endobronchial ultrasound-guided transbron- 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 assessment strategy: an explorative study. Respiration
ultrasound elastography for diagnosing mediastinal and hi- 2019;97:337-347.
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- bronchial ultrasonography elastography for predicting
ing and localizing nodal metastases during endobronchial malignancy. Turk Gogus Kalp Damar Cerrahisi Derg
ultrasound. Respiration 2015;90:499-506. 2020;28:158-165.
25. Rozman A, Malovrh MM, Adamic K, Subic T, Kovac V, 35. Uchimura K, Yamasaki K, Sasada S, et al. Quantitative
Flezar M. Endobronchial ultrasound elastography strain analysis of endobronchial ultrasound elastography in com-
ratio for mediastinal lymph node diagnosis. Radiol Oncol puted tomography-negative mediastinal and hilar lymph
2015;49:334-340. nodes. Thorac Cancer 2020;11:2590-2599.
26. Korrungruang P, Boonsarngsuk V. Diagnostic value of 36. Gompelmann D, Kontogianni K, Sarmand N, et al. Endo-
endobronchial ultrasound elastography for the differ- bronchial ultrasound elastography for differentiating benign
entiation of benign and malignant intrathoracic lymph and malignant lymph nodes. Respiration 2020;99:779-783.
nodes. Respirology 2017;22:972-977. 37. Chen YF, Mao XW, Zhang YJ, et al. Endobronchial ul-
27. Sun J, Zheng X, Mao X, et al. Endobronchial ultrasound trasound elastography differentiates intrathoracic lymph
elastography for evaluation of intrathoracic lymph nodes: a nodes: a meta-analysis. Ann Thorac Surg 2018;106:1251-
pilot study. Respiration 2017;93:327-338. 1257.
28. Huang H, Huang Z, Wang Q, et al. Effectiveness of the 38. Krouskop TA, Wheeler TM, Kallel F, Garra BS, Hall T.
benign and malignant diagnosis of mediastinal and hilar Elastic moduli of breast and prostate tissues under com-
lymph nodes by endobronchial ultrasound elastography. J pression. Ultrason Imaging 1998;20:260-274.
Cancer 2017;8:1843-1848. 39. Lyshchik A, Higashi T, Asato R, et al. Elastic moduli of
29. Ma H, An Z, Xia P, et al. Semi-quantitative analysis of thyroid tissues under compression. Ultrason Imaging
EBUS elastography as a feasible approach in diagnos- 2005;27:101-110.
ing mediastinal and hilar lymph nodes of lung cancer pa- 40. Seo M, Sohn YM. Differentiation of benign and metastatic
tients. Sci Rep 2018;8:3571. axillary lymph nodes in breast cancer: additive value of
30. Lin CK, Yu KL, Chang LY, et al. Differentiating malignant shear wave elastography to B-mode ultrasound. Clin Imag-
and benign lymph nodes using endobronchial ultrasound ing 2018;50:258-263.
elastography. J Formos Med Assoc 2019;118:436-443. 41. Bae SJ, Park JT, Park AY, et al. Ex vivo shear-wave elas-
31. Hernández Roca M, Pérez Pallarés J, Prieto Merino D, et al. tography of axillary lymph nodes to predict nodal metasta-
Diagnostic value of elastography and endobronchial ultra- sis in patients with primary breast cancer. J Breast Cancer
sound in the study of hilar and mediastinal lymph nodes. J 2018;21:190-196.
Bronchology Interv Pulmonol 2019;26:184-192. 42. VanderLaan PA, Wang HH, Majid A, Folch E. Endo-
32. Fournier C, Dhalluin X, Wallyn F, et al. Performance of bronchial ultrasound-guided transbronchial needle as-
endobronchial ultrasound elastography in the differentia- piration (EBUS-TBNA): an overview and update for
tion of malignant and benign mediastinal lymph nodes: re- the cytopathologist. Cancer Cytopathol 2014;122:561-
sults in real-life practice. J Bronchology Interv Pulmonol 576.
2019;26:193-198. 43. Herth FJ, Krasnik M, Vilmann P. EBUS-TBNA for the
33. Verhoeven RLJ, de Korte CL, van der Heijden EHFM. diagnosis and staging of lung cancer. Endoscopy 2006;38
Optimal endobronchial ultrasound strain elastography Suppl 1:S101-S105.
Review Med Ultrason 2022, Vol. 24, no. 1, 95-106
DOI: 10.11152/mu-3217
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
Shear wave based elastography – how does it work? kPa or can they change to m/s?
SWE techniques include vibration controlled tran- The propagating speed [12] of the generated shear
sient elastography (VCTE) and ARFI based techniques. wave is reported in meters per second (m/s) but can also
The shear waves are generated by a body-surface vibra- be converted to Young’s modulus values in kilopascals
tion, as in VCTE, or by the push-pulse of a focused US (kPa) by applying the formula E=3ρVs2, where E is tis-
beam, as in ARFI techniques. In VCTE, a body-surface sue elasticity, Vs is the shear wave speed, and ρ is the
vibration creates a shear wave, which then travels to the density of tissue in kg/m3, and making some assumptions
organ of interest. The frequency of the vibration is con- [1-3,6]. One main reason why it is preferable to report
trolled (50 Hz), as are its shape and amplitude. VCTE results in units of ms-1 rather than kPa is the fact that the
is implemented on the Fibroscan®, which is a dedicated SWS is measured by the scanner in ms-1. However, main-
device that does not display an anatomical image. In ly for liver application, the units of the Young’s modu-
ARFI-based techniques, the shear waves are generated lus are largely used, as many clinicians are familiar with
directly in the tissue. A convex or linear transducer trans- them.
mits focused US pulses (also known as a push pulses or Angle of insonation
ARFI) that generate shear waves. The pulses are repeated The angle of insonation has a significant influence on
several times over a short period of time, and the shear the measurement, which is of importance when a curved
waves generated travel at a much slower rate than US. transducer is used [15]. The shear waves are generated
B-mode tracking pulses are used to detect the propaga- perpendicular to the ARFI push pulse therefore the B-
tion velocity of the shear wave [12] by measuring the mode tracking must be in the same angle to accurately
difference in arrival time (time lag) between two points estimate the SWS.
at known distances apart from each other [1,6,13]. Such Region of interest
push pulses generate much slower shear waves off-axis The ROI should be positioned so that the push pulse
[14]. ARFI-based techniques include point shear-wave is generated perpendicular to the center of the transducer
elastography (pSWE) and multidimensional SWE (2D- surface. For more information about the technology we
SWE, 3D-SWE). pSWE measures the stiffness at the also refer to the recently published guidelines on elastog-
focal (~1cm3) point in the tissue whereas with 2D-SWE raphy [1,7,15,16].
the stiffness is measured over a much larger area and a Does the size and/or the shape of the ROI influences
color-coded image of the qualitative elastic properties is measurements?
displayed on the monitor of the US system [3,4]. The size of the ROI depends on the tissue being
Shear wave speed evaluated. Even though a larger ROI would give SWE
The shear wave speed (SWS) is almost one thousand information over a larger amount of tissue, it risks the in-
times lower than the velocity of US in soft tissues, the clusion of artifacts particularly in heterogeneous masses.
shear waves attenuate very rapidly and some do not prop- By using two different 2D-SWE US systems, it has been
agate in the simple fluids [14]. The shear wave propagates shown that, for the assessment of breast lesions, a small
faster in stiffer tissue than in softer tissue. The expected round ROI (approximately 2 mm in diameter) placed
SWS in the liver in normal and pathologic states is typi- over the stiffest area of the lesion was more accurate than
cally in the range 0.7 to 5.0 m/s (1.5 kPa to 75 kPa). For a larger ROI manually drawn along the margin of the le-
breast cancers it can be up to 10 m/s (300 kPa) and even sion [17]. In another study that assessed the influence on
higher for normal tendons. Pathology in any tissue often the accuracy of 2D-SWE in evaluating breast lesions by
creates changes in tissue stiffness making elastography a using three different ROI size (1, 2 and 3 mm), the diag-
method to characterize pathological changes. SWS val- nostic accuracy was not affected by changing the ROI
ues may vary depending on the vendor; therefore, vendor size [18]. In general, malignant lesions are heterogene-
specific cut-off values may be necessary. ous in stiffness and using the area of highest stiffness is
Differences of equipment more accurate in characterizing the lesion. However, for
Different equipment may give different values of homogenous tissue like liver, a larger ROI can average
stiffness within the same tissue in the same patient. This the stiffness over a larger area of tissue.
is because the measured values of SWS will vary with In ex vivo study involving porcine muscle, a sig-
a number of system factors, in particular shear-wave nificant increase of SWS (p<0.001) was observed for
vibration mean frequency and bandwidth. In addition, larger ROI widths. In this animal model, the SWS was
measurement bias may occur due to the algorithm em- also influenced by several other factors, including probe
ployed to calculate relative shear wave arrival time and frequency, applied pressure, muscle orientation, different
speed. machine settings, and placement depth [19].
Med Ultrason 2022; 24(1): 95-106 97
Artefacts Liver
SWE images are reconstructed using time-of-flight
based images. In heterogeneous tissues these algorithms Main objective, clinical value
might introduce a variety of artifacts. One of them is SWE The liver is an important target organ for the use of
under- and overestimation from reflections at stiffness in- elastography; stiffness correlates with the degree of fi-
terfaces. Reflected waves violate the assumption of a sin- brosis and indirectly with portal hypertension and the
gle direction of propagation, leading to artifacts in SWE risk of developing hepatocellular carcinoma. Due to the
images [20]. To avoid this, directional filters had been large overlap between stiffness values, guidelines do not
applied [20,21]. By separating the forward and backward recommend the use of SWE to differentiate benign and
components, it is possible to almost entirely remove the malignant focal liver lesions [15,25-28].
reflected wave [21]. It is highly recommended in transient The most important clinical management may be
shear wave applications to avoid reflection artifacts. For summarized as follows:
liver assessment, common artifacts include reverberation 1. SWE values within the normal range can rule out
from the liver capsule, respiratory/cardiac motion and compensated advanced chronic liver disease (cA-
vessel pulsation/loss of the SWE signal (fig 1). The pen- CLD) when in agreement with the clinical and labo-
etration of the US beam can also generate artifacts since ratory data.
consistent elasticity estimates cannot be obtained in the 2. SWE technologies perform best to rule out cirrhosis.
far field due to attenuation of the ARFI pulse. The most 3. SWE technologies can be used as first line assessment
consistent estimates are generally obtained near the focus for the severity of liver fibrosis but are much less reli-
zone of the ARFI pulse, where the largest displacements able in differentiating intermediate stages of fibrosis.
is generated by the push pulses [22]. A detailed analysis 4. An interquartile range/median (IQR/M) ≤30% with
of all the artifacts is out of the scope of this review article measurements taken in kPa or <15% when taken in
and can be found elsewhere [22-24]. 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
Fig 1. Example of the reverberation artifact from the liver be used can be given.
capsule in SWE. The red and teal areas are the artifacts; the ARFI-based techniques: transducer (frequency) selection
blue areas are the accurate stiffness measurements. Note that In adults, the convex transducer is used for perform-
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- ing the elastography studies, whereas in children the
SWE the artifact can be identified on the color map and be choice of the probe, either the linear or the curvilinear
avoided. one, depends on the body habitus and age. Generally, the
98 Giovanna Ferraioli et al How to perform shear wave elastography. Part I
same rule used for the choice between the two probes ment reported in m/s the IQR/M should be ≤15% because
for the B-mode image of the liver applies also to the as- the conversion between the two is not linear [29].
sessment of liver stiffness (LS) in children. However, it Pre-compression
should be kept in mind that the difference in frequency Pre-compression should be avoided.
between the two probes gives different readings in the How many measurements?
same subject. In phantom studies, it has been shown that Based on literature data, for the pSWE technique the
the readings with the higher frequency of the linear trans- EFSUMB and WFUMB guidelines have recommended
ducer are higher than those obtained with the convex to use the median value of 10 acquisitions [3,4,9]. How-
transducer. Moreover, in children the acquisition could ever, some studies have shown that the accuracy does not
be more challenging due to the lack of cooperation and decrease when fewer acquisitions (up to five) are obtained
this could affect the feasibility of the technique [30,31]. [38-40,42]. For 2D-SWE, the EFSUMB updated guide-
Description of (other) parameters lines have recommended to obtain at least three acquisi-
The strength of the push-pulse is higher in the center tions [3,4]. The updated WFUMB and SRU guidelines
of the transducer, thus the sampling should be done in are more cautious and have suggested five acquisitions
the central area of the image, whereas the sampling at the when a quality factor is available [9,43]. The higher num-
edge should be avoided. ber of acquisitions suggested by the WFUMB updated
The influence of depth on the estimation of the elas- guidelines may give a better estimation of the variabil-
tic properties is not negligible [32]. The acoustic push ity assessed through the calculation of the IQR/M ratio.
pulse is progressively attenuated as it traverses the tissue. Reproducibility
The results with the lowest variability are obtained at a The intra-observer reproducibility of VCTE [44-46],
depth of 4-5 cm from the skin surface [33]. The attenua- pSWE [34,36,47-49] and 2D-SWE [50-52] for LS as-
tion is higher in stiffer liver, thus in cirrhotic or steatotic sessment is excellent with ICC above 0.90.
patients, measurements are more variable [15]. The ROI How to use shear wave elastography
box should be perpendicular to the transducer. The transducer should be positioned in an intercostal
Region of interest (ROI) size, shape, others space; perpendicular to the liver in both superior/inferior
The region of interest should be in between 2-6 cm and right/left planes, avoiding the ribs or the lung arti-
below the liver capsule. facts. As the SWS is calculated based on B-mode, the
In pSWE the size of the region of interest (ROI) is quality of the B-mode US image affects the quality of
small and fixed by the manufacturer because the tech- the SWE acquisitions. The most common limitations en-
nique assesses the stiffness at a single location by using a countered with US, i.e. poor acoustic window, limited
sequence of push-pulses, generally up to five. penetration, and rib or lung shadowing, may influence
In 2D-SWE the size of the ROI is user-adjustable and both the feasibility and the performance of the SWE
can theoretically be as large as the ARFI FOV image. techniques. Some of these limitations can be avoided,
However, the larger the ROI the higher the risk of includ- thus the operator should obtain an optimal scan of the
ing artifacts. Thus, generally the ROI’s size in SWE tech- liver before launching the acquisition. The perpendicular
nique may influence the quality of the elastogram. Fol- position of the transducer can be assessed by looking at
lowing EFSUMB guidelines and recommendations, we the liver capsule that appears as a sharp white line, paral-
suggest using an ROI of 2.5x2.5 cm in size [3,4]. Many lel to the transducer’s line (fig 2). Motion of the probe
vendors have quality or confidence maps, which help to or of the patient affects the quality of the measurement
identify and avoid artifacts [15]. as well. The patient should breathe normally while the
Position of the transducer operator is searching for the best acoustic window and
The measurements should be performed through for the best area of liver parenchyma where the sample
the intercostal space rather than the subcostal approach box will be positioned. This area should be homogene-
yielding the highest intra- and interobserver agreement ous, i.e., free of vessels or ligaments. Before launching
[15,34-36]. the acquisition, the operator asks the patient to hold the
Description of quality parameters breath in a neutral position without performing a Vals-
The most important criterion for a measurement of alva’s maneuver for the few seconds needed for the ac-
good quality seems an IQR/M ≤30% when the results are quisition [15]. Special applications in pediatric patients
reported in Young’s modulus [29]. This ratio, in fact, is a are discussed elsewhere [30,31,53].
measure of the variability between consecutive acquisi- Tips and tricks
tions, and studies have reported a decrease in accuracy Depth as assessed by the skin-to-liver capsule dis-
when this criterion is not fulfilled [37-41]. For measure- tance may influence the SWS values assessed by all
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
Fig 2. Figure demonstrating the positioning of the liver capsule has been reported that portal vein thrombosis is also a
and FOV box in liver stiffness assessment. The transducer, liver
capsule and top of the FOV box should be parallel lines. The confounder [67]. On the other hand, SWE can play a role
liver capsule should be a sharp echogenic line. in cases of liver congestion due to right-sided heart fail-
ure, congenital heart diseases or valvular diseases as well
SWE-techniques. Due to the attenuation of the US beam, as in the hepatic sinusoidal obstruction syndrome or in
the depth for reliable measurements is up to 7 cm in most the Budd-Chiari syndrome [31,68].
systems; measurements performed deeper have a lower Specific artifacts
signal/noise ratio. Using a deep abdominal probe may Measurements should be performed at least 1-2 cm
allow for measurements at a greater depth in high BMI below the liver capsule to avoid reverberation artifacts.
patients. In ARFI techniques, the US beam that generates However, when using 2D-SWE with a quality map the
the shear waves is also attenuated by the interaction with measurement can be taken closer to the liver capsule as
the tissue that it traverses thus, its strength is inversely the artifact can be visualized and avoided. This is helpful
related to the depth; this attenuation is higher in cases in high BMI or steatotic patients since the reverberation
of liver steatosis or severe fibrosis. Due to these factors, artifact in these patients can be as small as 5mm and visu-
measurements in patients with significant liver steatosis alizing the artifact on 2D-SWE may help with placing the
or severe fibrosis could have a higher rate of unreliable ROI closer to the liver surface and still avoid the rever-
results or failures. This is also true in obese patients with beration artifact.
thick subcutaneous tissue due to higher attenuation of
the US beam in the near field. In staging liver fibrosis, Breast
Metavir-derived cutoff values are system-specific and
could not be applied interchangeably across different Main objective, clinical value
US systems. A recent study has shown that the agree- Various studies have shown that malignant and be-
ment between LS measurements obtained with different nign breast tumours differ significantly in their elastic-
US systems is good to excellent; however, the difference ity [7,69-75]. Benign alterations tend to be softer than
between values was higher than two kPa, assigning the malignant lesions. This fact forms the basis for the use
patient to different stages of liver fibrosis [54]. Because of elastography to differentiate between different breast
the overlap of LS values between METAVIR-derived tumors. SWE is a new method introduced in 2009 and,
scores is as large if not larger than the difference between unlike strain elastography, allows quantitative measure-
vendors, the updated SRU consensus advises that sepa- ment of tissue stiffness. SWE is not only capable of the
rate cut-off values for each vendor are not required when differentiation between benign and malignant tumours,
determining the likelihood for cACLD [29]. The SWE but can also be used for therapy monitoring under neoad-
values might be overestimated in certain diseases, e.g. juvant chemotherapy [76-79]. Recently, the fifth edition
sinusoidal obstruction syndrome due to congestion [55]. of the ACR BI-RADS Atlas 2013 incorporated elasticity
100 Giovanna Ferraioli et al How to perform shear wave elastography. Part I
Salivary glands
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)
Fig 7. a) Bilateral big and echogenic kidneys in a 6-year-old boy. b) Intense twinkling artifacts are seen.
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).
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).
Med Ultrason 2022; 24(1): 107-113 113
nary Radiology: Kidney, Bladder and Urethra. Springer- 6. Back SJ, Andronikou S, Kilborn T, Kaplan BS, Darge
Verlag London, 2013:95-119. K. Imaging features of tuberous sclerosis complex with
2. Bergmann C. Genetics of Autosomal Recessive Polycystic autosomal-dominant polycystic kidney disease: a con-
Kidney Disease and Its Differential Diagnoses. Front Pedi- tiguous gene syndrome. Pediatr Radiol 2015;45:386-
atr 2018;5:221. 395.
3. Gimpel C, Avni EF, Breysem L, et al. Imaging of Kidney 7. Katabathina VS, Kota G, Dasyam AK, Shanbhogue AK,
Cysts and Cystic Kidney Diseases in Children: An Inter- Prasad SR. Adult renal cystic disease: a genetic, biological,
national Working Group Consensus Statement. Radiology and developmental primer. Radiographics 2010;30:1509-
2019;290:769-782. 1523.
4. Müller RU, Benzing T. Cystic Kidney Diseases From the 8. Barisic I, Boban L, Loane M, et al. Meckel-Gruber Syn-
Adult Nephrologist’s Point of View. Front Pediatr 2018;6:65. drome: a population-based study on prevalence, prenatal
5. Randle SC. Tuberous Sclerosis Complex: A Review. Pedi- diagnosis, clinical features, and survival in Europe. Eur J
atr Ann 2017;46:e166-e171. Hum Genet 2015;23:746-752.
Case report Med Ultrason 2022, Vol. 24, no. 1, 114-116
DOI: 10.11152/mu-3325
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
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 (fig 4). A lytic lesion to the L1 vertebral body was also
neurologic deficits. revealed on CT-scan. Due to accessibility, we performed
Blood tests were suggestive for liver insufficiency a CT-guided biopsy of the liver lesions. Histopathologi-
(hypoalbuminemia, hypocholesterolemia) as well as cal analysis showed a mesenchymal proliferation of fusi-
signs of kidney injury (creatinine 2.02 mg/dL), increased form and epithelioid cells. Immune-histochemical test
levels of uric acid (11.7 mg/dL), severe anemia (6.6 g/ were negative for CD117, CD34, WT1, DOG1 (exclud-
dL) and thrombocytopenia. Serum markers for chronic ing gastro-intestinal stromal tumor), and positive for Vi-
viral hepatitis as well as for autoimmune disorders of the mentin (confirming mesenchymal origin) and SMA (spe-
liver were negative and the patient did not have proteinu- cific for myofibroblasts).
ria. The patient was negative for HIV. We concluded that the diagnosis was primary leio-
Abdominal ultrasonography revealed moderate as- myosarcoma of the spleen associated with liver, bones
cites, enlarged lymph nodes in the upper abdomen, en- and lymph nodes metastases. Due to the poor clinical sta-
larged liver with multiple hypoechoic tumors, measuring tus of the patient, as well as the extension of disease, the
up to 5 cm, important splenomegaly, with a long axis of 21 patient was referred for palliative treatment.
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 Fig 3. CEUS examination, late phase. Non-enhancing celiac
nodes and multiple hypodense liver and splenic lesions, lymph node, suggesting necrosis (asterisc). Arrow indicates
the largest located at the inferior splenic pole (93 mm) a liver metastasis.
116 Elena Simona Ioanițescu et al Primary splenic leiomyosarcoma – case report and literature review
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.
Fig 4. Abdominal CT scan (parenchymal phase) revealing mul- 4. Ioanitescu ES, Copaci I, Mindrut E, et al. Various aspects
tiple liver and splenic lesions, predominantly in the spleen. of Contrast-enhanced Ultrasonography in splenic lesions - a
pictorial essay. Med Ultrason 2020;22:2521.
Discussion 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
Splenic sarcomas are the rarest primary tumors aris- Applications: Update 2017 (Long Version). Ultraschall
ing in the spleen with only three cases of leiomyosarcoma Med 2018;39:e2–e44.
reported so far [8–10]. Two of these were in young wom- 6. Trenker C, Görg C, Freeman S, et al. WFUMB Position
en (49 and 54 years old) while the third was described in Paper—Incidental Findings, How to Manage: Spleen. Ul-
an 87 year old male patient. Previous reports have based trasound Med Biol 2021;47:2017-2032.
their diagnosis on histological evaluation after splenec- 7. Omar A, Freeman S. Contrast-enhanced ultrasound of the
tomy, but in our case the intervention was considered to spleen. Ultrasound 2016 Feb;24:41–49.
have an increased risk and biopsy from metastases was 8. Farah BL, Chee Y, Ching S, Tan C. Human Pathology: Case
elected. Two of the previously reported cases described Reports. Primary splenic leiomyosarcoma as an exception-
ally rare cause of ruptured splenomegaly – A case report
hemorrhage secondary to the splenic lesions; in our case,
and review of primary splenic sarcomas. Hum Pathol Case
the patient’s anemia was corrected by blood transfusions
Reports 2020;22:200452.
and ultrasonography and CT evaluation did not reveal 9. Piovanello P, Viola V, Costa G, et al. Locally advanced leio-
signs of bleeding. Another important aspect in our case myosarcoma of the spleen. A case report and review of the
is the extent of the disease; the previously reported cases literature. World J Surg Oncol 2007;5:135.
had only splenic involvement, while our patient presented 10. Daudia AT, Walker S, Morgan B, Lloyd DM. Images in surgery
liver, bones and lymph nodes metastases. 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
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 and coughing. The patient reported that the pain had ap-
peared suddenly six days earlier when standing against
Thoracic cage injuries present a wide range of potential a hard surface at the toilet while vomiting. She also re-
diagnoses. Costal cartilage (CC) injuries pose a diagnos- ported that she had several episodes of vomiting in the
tic challenge, often being missed in the initial assessment last week, due to an episode of gastroenteritis.
involving a chest radiograph. Costal cartilage fractures Clinical examination revealed no significant abnor-
should be considered, especially in cases of blunt trauma mality. In palpation, there was an area of significant
involving contact sports athletes or high-energy trauma, tenderness over the left costal margin near the left ster-
such as motor vehicle accidents [1]. These fractures are nocostal synchondrosis. An ultrasonography (US) of the
often underdiagnosed, while atraumatic cases have rarely area of pain was ordered, which revealed the findings de-
been reported in the literature [2]. In this article, a case of scribed in figure 1.
vomiting-induced costal cartilage fracture is presented. A CT scan was ordered to rule out a rib fracture. CT
images of the thorax identified no rib fracture. At axial
Case report CT images there was a tiny low-density area at the sev-
enth costal cartilage and swelling of the adjacent soft tis-
A 53-year-old woman presented at a primary care ser- sue. As the CT was non-diagnostic, further investigation
vice institution due to significant pain at the left hemitho- with MRI was performed (fig 2). Interestingly, the MRI
rax. The pain was more intense during deep breathing revealed a hyperintense linear area at the superior part
of the seventh costal cartilage that corresponded to the
Received 17.06.2020 Accepted 25.09.2020 tiny low density area identified in the CT. The diagno-
Med Ultrason
2022, Vol. 24, No 1, 117-119
sis of a vertical incomplete fracture of the seventh costal
Corresponding author: Emmanouil K Symvoulakis cartilage with associated soft tissue edema secondary to
Assistant Professor of Primary Health Care injury or significant vomiting was made.
Clinic of Social and Family Medicine, Analgesics and rest were recommended. At follow up
School of Medicine, University of Crete, Greece
Voutes, 71003, Heraklion, Greece
four weeks later, U/S showed that the soft tissue edema/
Phone: +302810394621 hematoma had partially resolved and the patient was
Email: symvouman@yahoo.com feeling significantly better.
118 Eleni Drakonaki et al Vomiting-induced costal cartilage fracture: a case report
Discussion
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,
Received 27.01.2022 Accepted 07.02.2022 Fig 1. Axial views of ESA (arrows) of left (asymptomatic) and
Med Ultrason right (symptomatic) sides. Focal hypoechoic and swollen ESA
2022, Vol. 24, No 1, 120-121, DOI: 10.11152/mu-3596, (arrowheads) at the right enthesis (A). A small linear calcifi-
Corresponding author: Ming-Yen Hsiao, MD, PhD cation (crosses, B) and regional hetero-echogenicity were also
Department of Physical Medicine and noted (void arrows, C). The corresponding probe positions at
Rehabilitation, College of Medicine, different levels of posterior medial iliac crest (D). US-guided
National Taiwan University, Taipei, Taiwan
injection of the ESA enthesis (E).
7, Zhongshan S. Rd., Zhongzheng Dist.,
Taipei City 100, Taiwan
Email: myhsiao@ntu.edu.tw arranged and the patient reported complete resolution of
Phone: +886-23123456 ex 67316 symptoms subsequently.
Med Ultrason 2022; 24(1): 120-128 121
ESA overlies the ESM dorsally in the lumbar region References
and fuses with the thoracolumbar fascia caudally, attach- 1. Daggfeldt K, Huang QM, Thorstensson A. The visible hu-
ing to the iliac crest and sacrum [1]. Although ESA has man anatomy of the lumbar erector spinae. Spine (Phila Pa
been proposed as a pain generator of lower back [2], im- 1976) 2000;25:2719-2725.
age-based pathological findings are rarely reported. 2. Creze M, Soubeyrand M, Nyangoh Timoh K, Gagey O. Or-
However, US-guided injection of the ESA enthesis ganization of the fascia and aponeurosis in the lumbar par-
aspinal compartment. Surg Radiol Anat 2018;40:1231-1242.
has been proposed as an effective method in treating iliac
3. Ricci V, Ozcakar L. Ultrasound-guided injection of the
crest pain syndrome [3]. On US examination, the ESA
erector spinae enthesis for iliac crest pain syndrome. J Res
can be visualized between the thoracolumbar fascia su- Med Sci 2019;24:69.
perficially and ESM deeply, as a band-like hyperechoic 4. Todorov P, Nestorova R, Batalov A. The sonoanatomy of
structure [4]. Our case demonstrated typical sonograph- lumbar erector spinae and its iliac attachment - the poten-
ic findings of ESA enthesopathy and its potential role in tial substrate of the iliac crest pain syndrome, an ultrasound
non-specific low back pain. study in healthy subjects. J Ultrason 2018;18:16-21.
Department of Physical Medicine and Rehabilitation, National Taiwan University Hospital, Taipei, Taiwan
Dear Editor, compound motor action potential and nearly absent sen-
sory nerve action potential in the left median nerve. EMG
A 70-year-old man received a total endovascular re- needle was inserted into pronator teres (PT) using surface
pair of thoracic aortic aneurysm with left axillary arte- anatomy for localization but failed to precisely locate
rial catheterization. Inability to flex left thumb, index flexor pollicis longus (FPL) as the expected increased
and middle fingers was noted immediately after surgery. spontaneous activities were not observed. Ultrasound
Additionally, he experienced numbness over left index (US)-guided muscle samplings were applied and there
and middle fingers and visited the rehabilitation clinic. were marked spontaneous activities in FPL and flexor
Physical examination revealed poor left thumb and in- digitorum superficialis (FDS) (fig 1a,b). Denervation
dex finger flexion with atrophied left thenar and forearm signs, including hyperechoic texture and decreased vol-
muscles. Tinel sign was elicited at the left axillary region. ume, were observed in multiple median-innervated fore-
Three months after surgery, nerve conduction studies arm muscles (fig 1c,d). US nerve tracking revealed focal
revealed prolonged distal motor latency with decreased 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
Received 27.11.2021 Accepted 23.01.2022 at axillary level.
Med Ultrason US guidance has been significantly important for pre-
2022, Vol. 24, No 1, 121-122, DOI: 10.11152/mu-3548,
Corresponding author: Kuo-Chang Wei, MD.
cise EMG samplings of atrophied muscles because it not
Department of Physical Medicine and only provide precise localization but also prevents un-
Rehabilitation, National Taiwan University necessary neurovascular injuries. Here, accidental radial
Hospital, Taipei, Taiwan. artery puncture might have occurred without US guid-
46, Gongyuan Rd., Zhongzheng Dist.,
Taipei City 100, Taiwan (R.O.C.).
ance during FPL sampling, considering their adjacency.
E-mail: gordon80446@gmail.com US is useful in detecting muscles’ denervation changes,
Phone: 886-2-23123456-66635 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 EMG which could make electrodiagnostic studies more
our case, atrophied left forearm flexors appeared hyper- precise and safer.
echoic under US, indicating median nerve injury level
References
was at or proximal to elbow level. Since all median-in-
1. Gunreben G, Bogdahn U. Real-time sonography of acute
nervated muscles lie within the forearm and hand, injury
and chronic muscle denervation. Muscle Nerve 1991;14:
level cannot be determined only by EMG when the me- 654-664.
dian nerve is injured above the elbow. US nerve tracking 2. Domkundwar S, Autkar G, Khadilkar SV, Virarkar M. Ul-
provides additional morphological information on the trasound and EMG-NCV study (electromyography and
injured nerve when incorporated into electrodiagnostic nerve conduction velocity) correlation in diagnosis of nerve
studies [2]. In conclusion, we suggest combining US and pathologies. J Ultrasound 2017;20:111-122.
1Department of Imaging Medicine, 2Department of Ultrasound, Shanghai University of Medicine and Health Sciences
Affiliated Zhoupu Hospital, Shanghai, P.R. China
1Department of Imaging Medicine, 2Department of Ultrasound, Shanghai University of Medicine and Health Sciences
Affiliated Zhoupu Hospital, Shanghai, P.R. China
To the Editor,
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
1Private Academic Consultant, Bangkok Thailand, 2Dr DY Patil University, Pune, India
1Department of Respiratory Disease, AORN Sant’Anna e San Sebastiano di Caserta, Caserta, Italy, 2Intensive Care
Unit, Hospital Morales Meseguer, Murcia, Spain
Author’s Reply
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, ings with respiratory failure, and in other studies it has
been shown that these patients with mild disease have
We read with interest the comments sent by Di Cos- little sonographic abnormalities [3].
tanzo et al. The authors agree on the potential role of lung The possibility of other processes was also studied,
ultrasound (LUS) and the degree of respiratory failure since an echocardiogram was performed along with
and prognosis in patients with COVID-19 [1] but raise LUS, with no evidence suggesting heart failure, bacterial
some methodological questions. superinfection and none of our patients had a history of
First, we would like to highlight that the study was pulmonary fibrosis at the time of inclusion in the study.
carried out in a field hospital, during the first wave of the The criteria for starting NIV were severe dysp-
pandemic in Spain, to decongest hospitals. That is why nea, tachypnoea over 30 bpm, PaO2/FiO2 <200 (or the
we did not have the facilities used in conventional hos- need for FiO2 greater than 0.4 to achieve an SpO2 of at
pitals. Patients were admitted to a conventional Internal least 92%) or acute ventilatory failure (pH <7.35 with
Medicine ward. Only 5 of the 107 patients had received PaCO2 >45 mm Hg).
positive ventilatory support [2] before LUS (2 of them We appreciate the comments and agree that it is nec-
in the first 10 hours), this is unlikely to have an impact. essary new well-designed studies to provide new insights
None had High-Flow Nasal Cannula Oxygen Therapy on the correlation between LUS and prognosis of COV-
(HFNCOT). The timing from admission to LUS evalu- ID-19.
ation was 7.64±4.13 days and we wanted to emphasize
the number of days from symptom onset as an important References
marker to develop adult respiratory distress syndrome 1. Hernández-Píriz A, Tung-Chen Y, Jiménez-Virumbrales D,
(ARDS). et al. Usefulness of lung ultrasound in the early identifica-
Regarding diaphragmatic ultrasound we agree it is an tion of severe COVID-19: results from a prospective study.
interesting parameter worth to study, however, our study Med Ultrason 2021 Doi: 10.11152/mu-3263.
focused on alterations of the lung parenchyma and the se- 2. Mateos-Rodríguez A, Ortega-Anselmi J, Candel-González
verity of ARDS and, at that moment, was not considered. FJ, et al. Alternative CPAP methods for the treatment of
secondary serious respiratory failure due to pneumonia by
We agree that the use of partial pressure of oxygen is
COVID-19, Med Clin (Barc) 2021;156:55-60.
more reliable than oxygen saturation by pulse oximetry, 3. Tung-Chen Y, Martí de Gracia M, Díez-Tascón A, et al. Cor-
however, we did not have access to a 24-hour laboratory relation between Chest Computed Tomography and Lung
to perform it. Ultrasonography in Patients with Coronavirus Disease
Although many patients with mild disease were not 2019 (COVID-19). Ultrasound Med Biol 2020;46:2918-
admitted, our aim was to correlate the ultrasound find- 2926.
In memoriam
Abstract (on a separate page) preceding the text body. Tables should be added at the end of the main document,
In the case of original papers, abstracts should not exceed in the same word document.
250 words and should have the following structure: 1) aims;
2) material and methods; 3) results; 4) conclusions. Abstracts 4. Editorial policy
for literature reviews and educational papers should not
exceed 200 words. For case reports, the abstract must not Medical Ultrasonography promotes evaluation of all the
exceed 120 words and must underline the following: 1) pur- scientific papers by independent reviewers. The editor-in-
pose of the presentation; 2) peculiarities of the case; ranking chief or one of the editors evaluates each manuscript and, in
of the issues approached within the general knowledge of the 1-3 weeks from reception, decides upon their priority level
respective condition. (sent to review, rejected without being sent for review or re-
Three to five keywords must be selected for every paper turned to authors with suggestions for improvement before
from the Index Medicus (http://www.ncbi.nlm.nih.gov/sites/ submitting to review). The editors reserve the right to request
entrez?db=mesh); the key words should be inserted after the any changes they may consider appropriate, in the title, struc-
abstract and separated by semi-colon (term1; term2; term3). ture or body of the paper.
The papers are submitted to two blind reviewers with
The main document has to be structured as follows: expertise in ultrasonography. Based on the reviewers’ recom-
Introduction – should define the topic of the paper and mendations, the editors decide whether a paper is published
present the status of current knowledge in the field. or not. In case of marked discrepancy between the decisions
Material and methods – should describe the equipment of the two reviewers, the editor may send the manuscript to
employed, the group of patients studied and the methodology. another arbitrator for additional comments and a recommend-
We recommend specification of the type of ultrasound ed decision. The full decisional process may last 6-8 weeks.
equipment employed. The statistical analysis methodology Failure of the authors to comply with the editorial revision
used must also be described. requests may induce publication rejection.
Results – should present the obtained data, in a concise A 150 euro processing fee (100 euro for the first author if
manner, preferably in tables and diagrams. SRUMB member) is charged for each accepted paper starting
Discussions – should present the interpretation of the re- from 1st March 2021. The fee must be paid ahead of publication,
searcher’s results from the perspective of relevant literature at the time when the authors give the authorization for publi-
data. cation of the final proof. Instructions for online payment are
Conclusions of the paper must be clearly stated in the end. available on the journal site. No free reprints of the published
Acknowledgements – should be made only to those who paper are supplied. No additional reprints may be ordered.
have made a substantial contribution to the study
References must include only papers that are quoted in the 5. Submitting manuscripts for publication
text and that have been published. References must be num-
bered in Arabic numerals in the order in which they appear To submit a manuscript for publication, one author
in the text (where they should be inserted between square should register on our website, and follow the five steps of the
brackets [ ]) and listed in numerical order. Titles of medical submission process. The journal does not accept email or
journals must be abbreviated according to the Index Medicus. CD submissions of articles. The editorial office will send a
All authors must be quoted for an article, if they are up to confirmation e-mail to the correspondence address.
six. Over seven authors, only the first three will be quoted, Remark: Medical Ultrasonography cannot be held respon-
followed by the “et al” indication. References should be listed sible for losing or damaging the files delivered through the
according to the following format (examples): Internet.