See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/287977265

Lung Ultrasound for Early Diagnosis of Ventilator-Associated Pneumonia

Article  in  Chest · December 2015

DOI: 10.1016/j.chest.2015.12.012

CITATIONS READS
78 1,837

8 authors, including:

Silvia Mongodi Gabriele Via

Policlinico San Matteo Pavia Fondazione IRCCS Cardiocentro Ticino
95 PUBLICATIONS   1,239 CITATIONS    106 PUBLICATIONS   3,370 CITATIONS   

SEE PROFILE SEE PROFILE

Martin Girard Benoit Misset

Centre hospitalier de l'Université de Montréal (CHUM) Centre Hospitalier Universitaire Rouen
25 PUBLICATIONS   548 CITATIONS    159 PUBLICATIONS   7,184 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Nosocomial Infection View project

Postoperative surgical outcomes View project

All content following this page was uploaded by Antonio Braschi on 22 February 2016.

The user has requested enhancement of the downloaded file.

Accepted Manuscript

Lung Ultrasound for Early Diagnosis of Ventilator-Associated Pneumonia

Silvia Mongodi, Gabriele Via, Martin Girard, Isabelle Rouquette, Benoit Misset,
Antonio Braschi, Francesco Mojoli, Bélaïd Bouhemad

PII: S0012-3692(15)00340-2
DOI: 10.1016/j.chest.2015.12.012
Reference: CHEST 208

To appear in: CHEST

Received Date: 5 October 2015

Revised Date: 3 November 2015
Accepted Date: 1 December 2015

Please cite this article as: Mongodi S, Via G, Girard M, Rouquette I, Misset B, Braschi A, Mojoli F,
Bouhemad B, Lung Ultrasound for Early Diagnosis of Ventilator-Associated Pneumonia, CHEST (2016),
doi: 10.1016/j.chest.2015.12.012.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Lung Ultrasound for Early Diagnosis of Ventilator-Associated

Pneumonia

Silvia Mongodi1,2,3; Gabriele Via2; Martin Girard4; Isabelle Rouquette1;

PT
Benoit Misset5, Antonio Braschi2,3, Francesco Mojoli2,3 and Bélaïd

RI
Bouhemad1

SC
1
Réanimation Chirurgicale, Department of Anesthesiology and Critical
Care, Groupe Hospitalier Saint Joseph, 185 rue Raymond-Losserand,
75674 Paris Cedex 14 (France)
U
AN
2
First Department of Anesthesia and Intensive Care, Fondazione IRCCS
Policlinico San Matteo, P.zzale Golgi 2, 27100 Pavia (Italy)
M

3
, Department of Clinical, Surgical, Diagnostic and Pediatric Sciences,
Section of Surgery and Anesthesiology, Unit of Anesthesia, Intensive Care
D

and Pain Therapy, University of Pavia (Italy)

4
Réanimation, Centre Hospitalier de l'Université de Montréal, Pavillon
TE

Notre-Dame, Montréal (Canada)

5
Réanimation Polyvalente, Groupe Hospitalier Saint Joseph, 185 rue
EP

Raymond-Losserand, 75674 Paris Cedex 14 (France)

C
AC

Corresponding author:

Dr Bélaïd Bouhemad, Service Anesthésie Réanimation CHU Dijon, 14 rue Gaffarel

21079 Dijon Cedex, France

e-mail: belaid_bouhemad@hotmail.com
 2
ACCEPTED MANUSCRIPT
Authors’ Contributions section:

- Silvia Mongodi: involvement in the conception, hypothesis, outline and

design of the study; data acquisition; drafting the article and substantial
involvement in its revision prior to submission
- Francesco Mojoli: data analysis and interpretation; drafting the article and
substantial involvement in its revision prior to submission

PT
- Gabriele Via: involvement in the conception, hypothesis, outline and
design of the study; data acquisition; substantial involvement in revision of

RI
the article prior to submission
- Martin Girard: involvement in data acquisition;

SC
- Isabelle Rouquette: substantial involvement in revision of the article prior
to submission

U
-
- Benoit Misset: substantial involvement in revision of the article prior to
AN
submission
- Antonio Braschi: substantial involvement in revision of the article prior to
M

submission
- Bélaïd Bouhemad: involvement in the conception, hypothesis, outline and
D

design of the study; data acquisition and analysis ; drafting the article and
substantial involvement in its revision prior to submission.
TE

No sources of support in the form of grants, gifts and/or equipment.

C EP

Word count: 2681 (exclude abstract, references, legends)

AC

Abstract word count: 250

This work was presented in part at the SMART Congress, Milan, May 2013, SMART

Congress, Milan, May 2014, and SMART Milan, May 2015.

Number pages from abstract: 14

3
ACCEPTED MANUSCRIPT
Abstract (250 words)
Background: Lung ultrasound (LUS) has been successfully applied for monitoring
aeration in ventilator-associated pneumonia (VAP) and to diagnose and monitor
community-acquired pneumonia. However, there is as yet no scientific evidence
about whether LUS reliably improves VAP diagnosis.
Methods: In a multicenter prospective study of 99 patients with suspected VAP, we

PT
investigated the diagnostic performance of LUS findings of infection, subpleural
consolidation, lobar consolidation, dynamic arborescent/linear air-bronchogram. We

RI
also evaluated the combination of LUS with direct microbiological examination of
endotracheal aspirations (EA). Scores for LUS findings and EA were analyzed:

SC
-Clinical-LUS score (VPLUS Ventilator-associated Pneumonia Lung Ultrasound
Score): ≥2 areas with subpleural consolidations 1 point; ≥1 area with dynamic
arborescent/linear air-bronchogram 2 points; purulent EA 1 point.

U
- VPLUS-EAgram: ≥2 areas with subpleural consolidations 1 points; ≥1 area with
AN
dynamic arborescent/linear air-bronchogram 2 points; purulent EA 1 point; positive
direct gram stain EA examination 2 points.
M

Results: For the diagnosis of VAP, subpleural consolidation and dynamic

arborescent/linear air-bronchogram had a positive predictive value of 86% with a
positive likelihood ratio of 2.8. Two dynamic linear/arborescent air-bronchograms
D

gave a positive predictive value of 94% with a positive likelihood ratio of 7.1. The
TE

area under the curve for VPLUS-EAgram and VPLUS were respectively 0.832 and
0.743. The VPLUS-EAgram ≥3 had of 77% (58-90) specificity and 78 (65-88)
EP

sensitivity; VPLUS ≥2 had 69% (50-84) specificity and 71% (58-81) sensitivity.
Conclusions: By detecting ultrasound features of infection, LUS is a reliable tool for
early VAP diagnosis at the bedside.
C

Clinical trial registration: ClinicalTrials.gov NCT02244723

AC
 4
ACCEPTED MANUSCRIPT
Introduction

Ventilator-associated pneumonia (VAP) is the most frequent nosocomial infection

among critical patients and is associated with increased mortality, morbidity and

duration of mechanical ventilation (MV) and ICU stay, with an obvious impact on

patient management costs.1-3 There is still no definitive validated diagnostic gold-

PT
standard procedure and early diagnosis of VAP remains a challenge to the

RI
intensivist. 4

Treatment involves identifying the causal germ(s) and active antibiotic therapy.

SC
Any delay in starting antibiotics in severe sepsis increases mortality.5,6 There is

therefore a pressing need for reliable diagnostic tools to diagnose VAP early in order

U
to start antibiotics promptly, avoiding two extreme approaches. On the one hand
AN
waiting for positive broncho-alveolar lavage (BAL) or protected distal sampling delays
M

treatment and increases mortality. On the other hand, giving antibiotics to all patients

suspected of VAP, and de-escalating or suspending treatment in case of negative

D

4
microbiologic samples results in inappropriate, massive use of antibiotics and
TE

fosters multi-resistant bacteria.7-9

At the bedside diagnostic and monitoring imaging of pulmonary infections

EP

mainly relies on chest X-ray (CXR), but this is unreliable in daily practice in the
C

critically ill.10-14 Lung ultrasound (LUS) is a simple, non-irradiating, non-invasive, cost-

AC

effective and bedside technique. Many studies have already showed that it is an

important means of distinguishing respiratory distress etiologies.13 It has also been

successfully employed to diagnose and monitor community-acquired pneumonia.15-17

In the critically ill, a Lung Ultrasound Score (LUSS) was reliable for quantifying lung

aeration and monitoring antibiotic efficacy in VAP.18 LUS also easily identified
18
subpleural foci of infection (small, rounded hypoechoic images) or dynamic air-
5
ACCEPTED MANUSCRIPT
bronchogram whitish consolidations.19 In a retrospective study, a score combining

procalcitonine (PCT) and LUS consolidation performed better than the Clinical

Pulmonary Infection Score (CPIS).20 Thus it seems that LUS could be valuable for

the early diagnosis of VAP.

We conducted a prospective pragmatic study to determine whether LUS could

PT
improve early VAP diagnosis. We evaluated the diagnostic performance of LUS

RI
alone and in combination with clinical and microbiological data.

U SC
AN
M
D
TE
C EP
AC
 6
ACCEPTED MANUSCRIPT
Materials and Methods

Patient selection and study design

The study was registered at ClinicalTrials.gov (NCT02244723) and approved by the

institutional review board (IRB) of each institution. Written consent was obtained from

each patient or next of kin. We prospectively included 103 patients with suspected

PT
VAP in ICUs at the Saint Joseph Hospital (Paris, France), Fondazione IRCCS

RI
Policlinico S. Matteo (Pavia, Italy) and Centre Hospitalier de l'Université de Montréal

(Canada). Four patients with missing data were excluded.

SC
Clinical suspicion of VAP was based on the classical criteria: MV ≥48 hours,

U
newly appeared/evolving CXR infiltrate and two or more of the following clinical

criteria: temperature ≥38.5°C or hypothermia (<36.5°C), leukocytosis >10 4/mL or

AN
leukopenia <4.103/mL; purulent tracheal secretions; PaO2/FiO2 <300. Exclusion
M

criteria were ongoing pneumonia and contraindication to fiber-bronchoscopy.

D

Data management
TE

Patients were included at the time VAP was suspected. At inclusion we calculated
21
EP

CPIS , recorded PCT ,did fiberoptic broncho-alveolar lavage (BAL), and sampled

endotracheal aspirate (EA) for direct gram stain examination (EAgram) and culture
C

(EAquant); LUS. BAL, and EA were all done within 8 hours.

AC

VAP diagnosis was confirmed by positive BAL (≥1 micro-organism with a

concentration ≥104 CFU/mL) or simultaneous presence of all clinical criteria with

negative BAL if antibiotics had been modified/introduced in the previous 48 hours.

EAgram was considered positive if any bacteria could be visualized after gram stain

test on tracheal secretions.

7
ACCEPTED MANUSCRIPT

Lung ultrasound

A comprehensive scan was taken in six areas for each lung (superior and inferior

areas in anterior, lateral, posterior fields using anterior and posterior axillary lines as
22
landmarks) and the following ultrasound findings were collected: 1) small

PT
18
subpleural consolidations: echo-poor regions >0.5 cm in diameter (Figure 1); 2)

RI
lobar/hemilobar consolidation defined by a tissue-like pattern (Figure 2); 3) dynamic

linear/arborescent air-bronchogram within lobar/hemilobar consolidations:

SC
hyperechoic images moving synchronously with inspiration 19,23 (Figures 2 and 3).

Statistical analysis
U
AN
For this pilot study we could not establish the number of patients needed or do a
M

24
power analysis. Consecutive patients were included as recommended . Statistical

uncertainty was indicated by reporting 95% confidence intervals.

D

Results are expressed as mean ± SD or median (inter-quartile range - IQR). Two

TE

groups were compared, patients with and without VAP, using the unpaired t-test or

Mann-Whitney test for numerical data and Fisher’s exact test for categorical data.
EP

Sensitivity, specificity, positive/negative predictive values, and positive/negative

C

likelihood were calculated for LUS signs (lobar/hemilobar consolidation, dynamic

AC

linear/arborescent air-bronchograms, subpleural consolidation) and for clinical

(purulent secretions), microbiological (EA) and laboratory (PCT) parameters.

Two post-hoc LUS-based scores were computed (Table 1):

1) Clinical-LUS score (VPLUS – Ventilator-associated pneumonia lung ultrasound

score): ≥2 areas with subpleural consolidations 1 point; ≥1 areas with dynamic

arborescent/linear air-bronchogram 2 points; purulent EA 1 point

 8
ACCEPTED MANUSCRIPT
2) Clinical-microbiological-LUS score (VPLUS-EAgram/VPLUS-EAquant) : ≥2 areas

subpleural consolidations 1 points; ≥1 areas with dynamic arborescent/linear air-

bronchogram 2 points; purulent EA 1 point; positive EAgram/EAquant 2 points.

ROC curves were plotted for VPLUS, VPLUS-EAgram, VPLUS-EAquant, CPIS,

PT
CPIS-EAgram, CPIS-EAquant. VPLUS-EAgram and CPIS-EAgram were calculated

RI
on the basis of EAgram; VPLUS-EAquant and CPIS-EAquant were computed on the

basis of EAquant (Table 1).

U SC
AN
M
D
TE
C EP
AC
9
ACCEPTED MANUSCRIPT
Results

The main clinical characteristics of patients with suspected VAP are listed in Table 2.

The prevalence of VAP was 64%. The only significant differences between the two

groups were in three parameters: purulent secretions were more frequent in patients

with VAP (63 vs. 31%; p=0.0054); PCT (1.0 vs. 5.0 ng/ml; p=0.0030) and SAPSII

PT
(46±18 vs. 59±15; p=0.0018) were lower in VAP patients. Gram-negative bacteria

RI
(Enterobacteriaceae and P. aeruginosa) were the most frequent microorganisms

(Table 3).

SC
The diagnostic performances of clinical, microbiological, laboratory and ultrasound

findings are depicted in Table 4. Among ultrasound findings, ≥1 area with

U
lobar/hemilobar consolidations was not specific for VAP, since it was seen in almost
AN
all our patients. Subpleural consolidations and dynamic linear/arborescent air-
M

bronchograms were the ultrasonographic signs of VAP. The presence of both signs

(28 patients) was highly specific (88%) with high PPV (86%) for VAP diagnosis and a
D

positive likelihood ratio (2.9).

TE

The diagnostic performances of the combined ultrasonographic and EA

microbiological findings are depicted in Table 5. For VAP diagnosis, the combination
EP

of dynamic linear/arborescent air-bronchogram and subpleural consolidation with

C

positive EAgram or EAquant had respectively 97% (84-100) and 97% (85-100)
AC

specificity, and positive likelihood ratios of 6.6 (0.9-48.9) and 7.1 (1.0-52.9).

The diagnostic performances of clinical and ultrasound scores are reported in Table

6. CPIS and CPIS-EAgram ≥ 6 yielded sensitivity/specificity of respectively 69/50%

and 84/47%. VPLUS ≥2 and VPLUS-EAgram ≥ 3 had sensitivity/specificity 72/69%

and 78/77%. The simultaneous presence of all VPLUS criteria (VPLUS 4) gave 100%

(89-100) specificity for diagnosing VAP.

 10
ACCEPTED MANUSCRIPT
ROC curves for CPIS, CPIS-EAgram, VPLUS and VPLUS-EAgram are shown in

Figure 4; areas under the curve (AUC) were respectively 0.576 (0.473-0.675), 0.693

(0.585-0.787), 0.743 (0.645-0.825) and 0.832 (0.737-0.903). CPIS-EAquant and

VPLUS-EAquant AUC, not shown in the figure, were 0.745 (0.636-0.835) and 0.874

(0.782-0.938).

PT
VPLUS AUC was not different from the CPIS-EAgram AUC (p=0.33); the VPLUS-

RI
EAgram AUC was higher than the CPIS-EAgram AUC (p <0.02) and equivalent to

the CPIS-EAquant AUC (p=0.38).

U SC
AN
M
D
TE
C EP
AC
11
ACCEPTED MANUSCRIPT
Discussion

The combination of purulent secretions, direct microbiological examination of

EA and specific LUS signs allows a reliable early diagnosis of VAP.

Diagnosis of VAP

PT
Diagnosis and treatment of VAP remain problematic. VAP is suspected when

RI
a new radiographic infiltrate develops in a patient with fever/hypothermia,

leukocytosis/leukopenia, purulent tracheal secretions and impaired oxygenation.

SC
However, many non-infectious processes can cause fever and pulmonary infiltrate in

U
ventilated patients making the diagnosis of VAP difficult. Thus clinical signs are non-

specific.25 The “clinical approach” suggests broad-spectrum antibiotics for all patients
AN
clinically suspected of VAP. This implies widespread use of antibiotics, which only
M

increases the prevalence of multi-drug resistant bacteria.7-9

A positive quantitative culture of BAL (≥104 CFU/mL) is considered the gold

D

standard for a diagnosis of VAP but it takes 24-48 hours to get definitive results. This
TE

“biological approach” aims at treating only microbiologically confirmed VAP; it is of

course very specific but may delay the introduction of antibiotics, resulting in
EP

increased mortality.5,6 A recent study reported that EA and BAL are associated with
C

similar clinical outcomes and overall use of antibiotics.26 Reliable diagnostic tools for
AC

early diagnosis of VAP are therefore essential in order to start antibiotic treatment as

soon as possible, avoiding the negative effects of these two extreme approaches.

Our clinical findings confirm previous reports. A combination of clinical criteria

(CPIS) has already been proposed for VAP diagnosis.21,27 However, our results

indicate poor accuracy with CPIS, as also seen in previous studies.28 Combining
 12
ACCEPTED MANUSCRIPT
clinical parameters and microbiological results (EA-quant or BAL gram-staining)29

gives better diagnostic performance, as confirmed by our results (CPIS-EAquant);

nevertheless, any quantitative culture requires 24 to 48 hours so antibiotic treatment

may be delayed.

Among laboratory findings PCT was lower in patients with VAP, suggesting its

PT
20 29
low impact on early diagnosis. Both retrospective and prospective studies found

RI
similar results.

Under mechanical ventilation, bacteria can colonize the patient’s oropharynx

SC
and enter the lower respiratory tract around the endotracheal tube cuff or through the

lumen. This may result in VAP. Clinical signs are not useful to distinguish

U
colonization and infection and VAP diagnosis requires a new persistent infiltrate
AN
CXR, which may be hard to assess in critically ill patients. Lung imaging in VAP
M

diagnosis still relies on CXR and CT-scan. CXR is the most routinely used tool, but

its quality and reliability in ICU patients are very poor.10-14 Many critically ill patients
D

have preexisting lung infiltrates often related to non-infectious causes, preventing

TE

accurate detection of new infiltrates due to VAP. It has in fact been reported that 38%

of patients under MV had abnormal CXR findings at ICU admission.30 Moreover,

EP

portable chest radiographs for ICU patients have only limited diagnostic

performance13; this makes it difficult to identify VAP and the reported specificity of a
C

pulmonary opacity consistent with pneumonia is only 27-35%.31,32 Recommendations

AC

have taken account of this limited accuracy and restrict the utility of CXR to two well-

defined situations: in mechanically-ventilated patients with suspected pneumonia,

normal CXR rule out VAP; in patients with confirmed VAP, CXR defines the severity

of the pneumonia (multilobar or not) and indicates complications for diagnosis (e.g.

cavitations).
13
ACCEPTED MANUSCRIPT
Interpretation of chest infiltrates in critically ill patients could be improved with

the use of CT-scan, but it is less easily available, irradiating, and patients have to be

transferred, which may be hazardous for many critically ill individuals. Besides, no

specific CT-findings of pneumonia were reported in 40 ARDS ventilated patients and

the CT diagnostic performance for VAP mainly relies on identifying patients without

PT
pneumonia.33

RI
CT-scan at admission (i.e. a “baseline” CT-scan) should still be mandatory to
30
ensure that pulmonary infiltrates are new but in fact it cannot be reasonably done,

SC
and it would be easier using LUS.18

Lung ultrasound for diagnosis of VAP

U
AN
15,16 34
Several studies and a meta-analysis support that community-acquired
M

pneumonia in ED can be diagnosed from consolidations at LUS and monitored in

function of their number and size.15,16 In ICU ventilated patients lower lobe
D

consolidations are frequently detected by LUS with good sensitivity and specificity;
TE

however, numerous causes other than pneumonia can explain asymmetric

consolidations. As clearly shown by our results, lung consolidation in a ventilated

EP

patient is not specific for VAP.

C

VAP is, however, marked by several histopathological features. Firstly itis a

AC

multifocal process involving both lungs, particularly the lower lobes. This is probably

due to dissemination of germs from the colonized tracheobronchial tree by positive-

pressure MV. Secondly VAP involves the simultaneous presence of different

histological grades of bronchopneumonia. As the foci of bronchopneumonia spread

to the periphery, with loss of aeration, the different grades can be detected by LUS:

irregularly spaced B-line foci very likely correspond to the early stages of infection
 14
ACCEPTED MANUSCRIPT
with interstitial inflammation,35 while focal and confluent bronchopneumonia are

detected as small subpleural consolidations or lobar/hemilobar consolidations.

Bronchi filled with air and secretions appear as hyperechoic linear/arborescent

bronchograms within lobar/hemilobar consolidations, moving synchronously with

respiratory acts, and can be easily detected by LUS.19 The specific histopathology of

PT
VAP makes LUS particularly effective for early diagnosis.

RI
Interestingly no relationship was found between bacterial burden and the

stage or histologic grade of lung infection.36 However, we previously showed that the

SC
intensity of ultrasound findings is proportional to the spread of infectious foci. In fact,

we showed using CT-scan, that antibiotics effectiveness in VAP patients can be

reliably monitored using LUS.18

U
AN
All told, therefore, LUS could be valuable in the diagnosis of VAP. Using
M

antibiotics in a patient with LUS consolidation may lead to wrongly treating cases

without pulmonary infections. Consolidations with a dynamic linear/arborescent air-

D

bronchogram or subpleural consolidations are less common, therefore more useful

TE

for the diagnosis of VAP. This is in line with CXR findings that air-bronchograms were

the only sign closely correlated with pneumonia.32

EP

A combined approach
C
AC

EA are easily sampled by direct suctioning through the endotracheal tube; however,

microscopic examination and EA culture are frequently inconclusive in patients

suspected of VAP, because the upper respiratory tract of most ICU patients is

colonized anyway by potential pulmonary pathogens whether or not a deep

pulmonary infection is present.37,38 Consequently culture analysis of EA specimens

15
ACCEPTED MANUSCRIPT
from patients under MV shows moderate-to-high sensitivity but generally low

specificity.3

Our observational study assessed the value of a combination of clinical

criteria, LUS and EA gram stain direct examination for early diagnosis of VAP. The

combination of LUS with a simple clinical criterion such as purulent secretions (i.e.

PT
VPLUS) had balanced specificity and sensitivity around 70%, very similar to CPIS-

RI
EAquant. Addition of the microbiological contribution (VPLUS-EAgram and VPLUS-

EAquant) brings both up to 80%.

SC
VPLUS and VPLUS-EAgram scores are easy to calculate and can be used in

patients suspected of VAP to identify high-risk cases and start antibiotics (Tables 1

U
and 6): if all VPLUS criteria are met, the diagnosis of VAP is sure (specificity 100%,
AN
PPV 100%).
M

The results were best with the combination of dynamic linear/arborescent

bronchogram or ≥ 2 areas with subpleural consolidations and a positive gram stain

D

examination. The positive likelihood ratio for the combination is 14.5 and indicates a
TE

conclusive increase in the likelihood of disease.

The score-based approach gives better results than the usual scores like CPIS
EP

and CPIS associated with EA (Figure 4). A retrospective study showed that a

Clinical-LUS score based on the presence of lobar consolidation or small antero-

C

lateral subpleural consolidations performed better than CPIS.20 We based our score
AC

on the more specific sign of lung infections (Table 3), i.e. dynamic linear/arborescent

air-bronchograms and subpleural consolidations. The score is very accurate, with an

AUC around 0.8, higher than any previous score.

Limitations
 16
ACCEPTED MANUSCRIPT
VAP was only diagnosed when signs and symptoms were present. Value of

LUS in monitoring development of pulmonary infection was not tested, which could

be tested by daily search of ultrasonographic signs of VAP. We did not test this

hypothesis as we conducted a pragmatic study, in accordance with a practical routine

approach.

PT
Diagnosis of VAP is an important issue for which there is still no validated definitive

RI
diagnostic gold-standard procedure. Even bronchoscopic procedures can be

questioned.39 However, these techniques are used in VAP diagnosis at the bedside

SC
when physicians have to decide whether or not an ICU patient should receive

antibiotics and in studies on VAP.40

U
There was a high prevalence of VAP in our series (68%), which might explain the
AN
good PPV and low NPV. Specificity and the positive likelihood ratio are independent
M

of the prevalence of the disease and remain high.

The study was conducted on 99 patients so it clearly needs to be confirmed with a

D

larger population.
TE

LUS is operator-dependent and requires a trained physician. Some patients may be

difficult to examine by LUS, as obese individuals and patients having subcutaneous

EP

emphysema or large thoracic dressings.

C
AC
17
ACCEPTED MANUSCRIPT

Conclusions

We studied a novel approach for rapid, cheap, safe bedside diagnosis of VAP

based on the detection of specific ultrasound features. The specificity of the

examination for VAP diagnosis could be increased by daily LUS monitoring of ICU

PT
patients.

RI
Clinical trials are now needed to assess whether this new approach combining

LUS, purulent secretions and direct examination of EA leads to earlier, appropriate

SC
prescription of antibiotics to patients who do develop VAP, and prevent overuse of

antibiotics in patients who do not develop VAP.

U
AN
Acknowledgments:
M

Dr Ilaria Godi and Dr Guido Tavazzi for data acquisition.

D

IRB Committee name and project approval number :

TE

- Fondazione IRCCS Policlinico S. Matteo – Pavia: Committee name:

COMITATO ETICO REFERENTE PER L’AREA DI PAVIA; Protocol number:

EP

P-20120030900
C

- Comité de Protection des Personnes Ile de France II. IRB registration #

AC

00001072Ref. 2013-07-01

- Comité d'éthique de la recherche du CHUM 13.223 - CÉR - approbation

FINALE - 2014-01-10
 18
ACCEPTED MANUSCRIPT

References

1. Bergmans DC, Bonten MJ, Gaillard CA, et al. Indications for antibiotic use in ICU
patients: a one-year prospective surveillance. J Antimicrob Chemother.
1997;39(4):527-535.

PT
2. Bregeon F, Ciais V, Carret V, et al. Is ventilator-associated pneumonia an independent
risk factor for death? Anesthesiology. 2001;94(4):554-560.
3. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med.

RI
2002;165(7):867-903.
4. Guidelines for the management of adults with hospital-acquired, ventilator-associated,
and healthcare-associated pneumonia. Am J Respir Crit Care Med. 2005;171(4):388-

SC
416.
5. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Campaign: international
guidelines for management of severe sepsis and septic shock: 2008. Intensive Care

U
Med. 2008;34(1):17-60.
6. Ibrahim EH, Sherman G, Ward S, Fraser VJ, Kollef MH. The influence of inadequate
AN
antimicrobial treatment of bloodstream infections on patient outcomes in the ICU
setting. Chest. 2000;118(1):146-155.
7. Rello J, Ausina V, Ricart M, et al. Risk factors for infection by Pseudomonas
aeruginosa in patients with ventilator-associated pneumonia. Intensive Care Med.
M

1994;20(3):193-198.
8. Trouillet JL, Chastre J, Vuagnat A, et al. Ventilator-associated pneumonia caused by
potentially drug-resistant bacteria. Am J Respir Crit Care Med. 1998;157(2):531-539.
D

9. Yu VL, Singh N. Excessive antimicrobial usage causes measurable harm to patients

with suspected ventilator-associated pneumonia. Intensive Care Med. 2004;30(5):735-
TE

738.
10. Graat ME, Choi G, Wolthuis EK, et al. The clinical value of daily routine chest
radiographs in a mixed medical-surgical intensive care unit is low. Crit Care.
EP

2006;10(1):R11.
11. Graat ME, Kroner A, Spronk PE, et al. Elimination of daily routine chest radiographs
in a mixed medical-surgical intensive care unit. Intensive Care Med. 2007;33(4):639-
644.
C

12. Kroner A, Binnekade JM, Graat ME, et al. On-demand rather than daily-routine chest
radiography prescription may change neither the number nor the impact of chest
AC

computed tomography and ultrasound studies in a multidisciplinary intensive care

unit. Anesthesiology. 2008;108(1):40-45.
13. Lichtenstein D, Goldstein I, Mourgeon E, Cluzel P, Grenier P, Rouby JJ. Comparative
diagnostic performances of auscultation, chest radiography, and lung ultrasonography
in acute respiratory distress syndrome. Anesthesiology. 2004;100(1):9-15.
14. Rouby JJ, Puybasset L, Cluzel P, Richecoeur J, Lu Q, Grenier P. Regional distribution
of gas and tissue in acute respiratory distress syndrome. II. Physiological correlations
and definition of an ARDS Severity Score. CT Scan ARDS Study Group. Intensive
Care Med. 2000;26(8):1046-1056.
15. Lichtenstein DA, Meziere GA. Relevance of lung ultrasound in the diagnosis of acute
respiratory failure: the BLUE protocol. Chest. 2008;134(1):117-125.
19
ACCEPTED MANUSCRIPT
16. Reissig A, Kroegel C. Sonographic diagnosis and follow-up of pneumonia: a
prospective study. Respiration. 2007;74(5):537-547.
17. Sperandeo M, Carnevale V, Muscarella S, et al. Clinical application of transthoracic
ultrasonography in inpatients with pneumonia. Eur J Clin Invest. 2010;41(1):1-7.
18. Bouhemad B, Liu ZH, Arbelot C, et al. Ultrasound assessment of antibiotic-induced
pulmonary reaeration in ventilator-associated pneumonia. Crit Care Med.
2010;38(1):84-92.
19. Lichtenstein D, Meziere G, Seitz J. The dynamic air bronchogram. A lung ultrasound
sign of alveolar consolidation ruling out atelectasis. Chest. 2009;135(6):1421-1425.

PT
20. Zagli G, Cozzolino M, Terreni A, Biagioli T, Caldini AL, Peris A. Diagnosis of
ventilator-associated pneumonia: a pilot, exploratory analysis of a new score based on
procalcitonin and chest echography. Chest. 2014;146(6):1578-1585.

RI
21. Luna CM, Blanzaco D, Niederman MS, et al. Resolution of ventilator-associated
pneumonia: prospective evaluation of the clinical pulmonary infection score as an
early clinical predictor of outcome. Crit Care Med. 2003;31(3):676-682.

SC
22. Bouhemad B, Brisson H, Le-Guen M, Arbelot C, Lu Q, Rouby JJ. Bedside ultrasound
assessment of positive end-expiratory pressure-induced lung recruitment. Am J Respir
Crit Care Med. 2011;183(3):341-347.
23. Volpicelli G, Elbarbary M, Blaivas M, et al. International evidence-based

U
recommendations for point-of-care lung ultrasound. Intensive Care Med.
2012;38(4):577-591.
AN
24. Bossuyt PM, Reitsma JB, Bruns DE, et al. The STARD statement for reporting studies
of diagnostic accuracy: explanation and elaboration. Ann Intern Med.
2003;138(1):W1-12.
M

25. Fabregas N, Ewig S, Torres A, et al. Clinical diagnosis of ventilator associated

pneumonia revisited: comparative validation using immediate post-mortem lung
biopsies. Thorax. 1999;54(10):867-873.
26. Group TCCCT. A randomized trial of diagnostic techniques for ventilator-associated
D

pneumonia. N Engl J Med. 2006;355(25):2619-2630.

27. Pugin J, Auckenthaler R, Mili N, Janssens JP, Lew PD, Suter PM. Diagnosis of
TE

ventilator-associated pneumonia by bacteriologic analysis of bronchoscopic and

nonbronchoscopic "blind" bronchoalveolar lavage fluid. Am Rev Respir Dis.
1991;143(5 Pt 1):1121-1129.
EP

28. Fartoukh M, Maitre B, Honore S, Cerf C, Zahar JR, Brun-Buisson C. Diagnosing

pneumonia during mechanical ventilation: the clinical pulmonary infection score
revisited. Am J Respir Crit Care Med. 2003;168(2):173-179.
29. Jung B, Embriaco N, Roux F, et al. Microbiogical data, but not procalcitonin improve
C

the accuracy of the clinical pulmonary infection score. Intensive Care Med.
AC

2010;36(5):790-798.
30. Nseir S, Favory R, Jozefowicz E, et al. Antimicrobial treatment for ventilator-
associated tracheobronchitis: a randomized, controlled, multicenter study. Crit Care.
2008;12(3):R62.
31. Lefcoe MS, Fox GA, Leasa DJ, Sparrow RK, McCormack DG. Accuracy of portable
chest radiography in the critical care setting. Diagnosis of pneumonia based on
quantitative cultures obtained from protected brush catheter. Chest. 1994;105(3):885-
887.
32. Wunderink RG, Woldenberg LS, Zeiss J, Day CM, Ciemins J, Lacher DA. The
radiologic diagnosis of autopsy-proven ventilator-associated pneumonia. Chest.
1992;101(2):458-463.
 20
ACCEPTED MANUSCRIPT
33. Winer-Muram HT, Steiner RM, Gurney JW, et al. Ventilator-associated pneumonia in
patients with adult respiratory distress syndrome: CT evaluation. Radiology.
1998;208(1):193-199.
34. Hu QJ, Shen YC, Jia LQ, et al. Diagnostic performance of lung ultrasound in the
diagnosis of pneumonia: a bivariate meta-analysis. Int J Clin Exp Med. 2014;7(1):115-
121.
35. Goldstein I, Bughalo MT, Marquette CH, Lenaour G, Lu Q, Rouby JJ. Mechanical
ventilation-induced air-space enlargement during experimental pneumonia in piglets.
Am J Respir Crit Care Med. 2001;163(4):958-964.

PT
36. Fabregas N, Torres A, El-Ebiary M, et al. Histopathologic and microbiologic aspects
of ventilator-associated pneumonia. Anesthesiology. 1996;84(4):760-771.
37. Baselski V. Microbiologic diagnosis of ventilator-associated pneumonia. Infect Dis

RI
Clin North Am. 1993;7(2):331-357.
38. Johanson WG, Jr., Pierce AK, Sanford JP, Thomas GD. Nosocomial respiratory
infections with gram-negative bacilli. The significance of colonization of the

SC
respiratory tract. Ann Intern Med. 1972;77(5):701-706.
39. Chastre J, Trouillet JL, Combes A, Luyt CE. Diagnostic techniques and procedures for
establishing the microbial etiology of ventilator-associated pneumonia for clinical
trials: the pros for quantitative cultures. Clin Infect Dis. 2010;51 Suppl 1:S88-92.

U
40. Chastre J. Ventilator-associated pneumonia and ventilator-associated conditions:
apples are not oranges (mix only in a salade de fruits!). Crit Care Med.
AN
2015;43(1):227-229.
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
Table 1. Parameters and relative points for ultrasound VAP score (VPLUS –
Ventilator associated Pneumonia Lung Ultrasound Score) and CPIS (Clinical
Pulmonary Infection Score)

VPLUS Points

≥ 2 Areas with subpleural consolidations 1

≥ 1 Areas with dynamic linear air bronchogram 2

PT
Purulent secretions 1
Positive EAgram* / EAquant§ 2

RI
CPIS Points
Temperature (°C)
≥ 36.5 and ≤ 38.4 0

SC
≥ 38.5 and ≤ 38.9 1
≥ 39.0 or ≤ 36.0 2
Leukocytes /mL

U
≥ 4000 and ≤ 11000 0
< 4000 or >11000 1
AN
Tracheal secretions
Few 0
Moderate 1
Large 2
M

Purulent +1
Oxygenation
PaO2/FiO2 > 240 or presence of ARDS 0
D

PaO2/FiO2 ≤ 240 and absence of ARDS 1

Chest X-rays
TE

No infiltrates 0
Patchy or diffuse infiltrate 1
Localized infiltrate 2
Positive EAgram* / EAquant§
EP

VPLUS, Ventilator-associated Pneumonia Lung Ultrasound Score. CPIS, Clinical

C

Pulmonary Infection Score. EA, Endotracheal aspirate. Positive EAgram, positive

direct gram stain examination of endotracheal aspirate. Positive EAquant, positive
AC

culture of endotracheal aspirate. * EAgram was used to calculate CPIS-EAgram and

VPLUS-EAgram. § EAquant was used to calculate CPIS-EAquant and VPLUS-
EAquant.
 ACCEPTED MANUSCRIPT
Table 2. Patients’ main details

Patients with Patients p

All Patients
VAP without VAP
(99)
(63) (36)

Age (years) 66 ± 12 65 ± 12 67 ± 12 0.55

Male no. %) 78 (78) 51 (75) 27 (84) 0.44
BMI (kg/m2) 27 [24-30] 26 [23-31] 27 [24-29] 0.99
Previous/ongoing ATB (no. %) 87 (87) 57 (84) 30 (94) 0.22

PT
Under ATB at inclusion (no. %) 63 (63) 38 (56) 25 (78) 0.05
SAPS II 50 ± 18 46 ± 18 59 ± 15 0.0018
MV duration at enrollment 8 [5-12] 9 [5-13] 7 [4-11] 0.39

RI
(days)
Procalcitonine (ng/ml) 2 [1-5.5] 1.0 [0.4-3.6] 5.0 [1.5-15.3] 0.0030

SC
TRANSFERRED FROM:
Medical (no. %) 64 (64) 45 (66) 19 (59) 0.51
Surgical (no. %) 36 (36) 23 (34) 13 (41) 0.51

U
CLINICAL CRITERIA
Purulent secretions (no. %) 52 (52) 42 (62) 10 (31) 0.0054
AN
Body temperature (°C) 38.0 ± 1.0 38.0 ± 1.0 37.9 ± 1.2 0.53
Leukocytes (cell*103/mL) 13.3 [10.5- 12.7 [10.2- 14.6 [11.5- 0.08
17.7] 16.8] 19.6]
M

PaO2/FiO2 (mmHg) ù 188 [142-255] 197 [130-255] 177 [143-264] 0.86

CPIS 6.0 ± 1.4 6.0 ± 1.4 5.8 ± 1.6 0.45

D

PROGNOSIS
ICU Mortality (no. %) 33 (33) 24 (35) 9 (28) 0.51
TE

MV total duration 17 [9-31] 18 [10-32] 15 [8-22] 0.26

(days)
ICU stay (days) 25 [16-38] 24 [16-40] 26 [16-36] 0.98
EP

Definition of abbreviations: VA, Ventilator-associated pneumonia; ATB, Antibiotics; CPIS,

Clinical Pulmonary Infection Score. Data are mean ± standard deviation or median and
interquartile range [IQR]. Patients with and without VAP were compared using a non-paired t-
test or Wilcoxon test. Proportions were compared using a Chi2 test or Fisher’s exact test. In
C

bold type: significant differences between the two populations.

AC
 ACCEPTED MANUSCRIPT
Table 3. Bacteriological data from positive bronchoalveolar lavage

Pathogens on positive bronchoalveolar lavage (67) No. (%)

Gram negative bacteria 56 (86.2)

P. aeruginosa 19 (29.2)
E. cloacae / aerogenes 10 (15.4)
E. coli 9 (13.9)
K. pneumoniae 8 (12.3)

PT
H. influenzae 8 (12.3)
A. baumanii 4 (6.2)
Others 13 (20)

RI
Gram positive bacteria 16 (24.6)
S. aureus 14 (21.5)

SC
Others 2 (3.1)
17 bronchoalveolar lavages (25%) were positive for multiple pathogens.

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 4. Diagnostic performances for CPIS, tracheal aspirations and lung ultrasound for the diagnosis of VAP

N° of
Sensitivity Specificity PPV NPV +LR -LR
patients

PT
Clinical-laboratory findings
Purulent secretions 52 62 (49-73) 69 (50-84) 81 (68-90) 46 (31-61) 2.0 (1.1-3.4) 0.6 (0.4-0.8)

RI
CPIS ≥ 6 62 68 (55-78) 50 (32-68) 74(62-84) 42 (26-59) 1.4 (0.9-2.0) 0.7 (0.4-1.1)
CPIS ≥ 7 32 37 (25-49) 78 (60-91) 78 (60-91) 37 (25-49) 1.7 (0.8-3.5) 0.8 (0.6-1.0)

SC
PCT ≥ 0.5 67 74 (62-85) 7 (0-17) 63 (51-74) 12 (0-27) 0.8 (0.7-1.0) 3.6 (0.9-14.4)
PCT ≥ 1.0 51 49 (36-62) 15 (1--28) 55 (41-69) 12 (1-23) 0.58 (0.4-0.7) 3.43 (1.3-8.8)

U
AN
Endotracheal aspirate
Positive gram stain direct examination 40 57 (43-70) 77 (58-90) 83 (67-93) 48 (33-63) 2.4 (1.2-4.8) 0.6 (0.4-0.8)

M
Positive culture 52 83 (70-92) 71 (51-87) 85 (72-93) 69 (49-85) 2.9 (1.6-5.3) 0.2 (0.1-0.4)

Ultrasound findings

D
Lobar/hemilobar consolidations 94 93 (86-99) 0 (0-0) 66 (57-76) 0 (0-0) 0.9 (0.9-1.0) -

TE
Subpleural consolidations ≥ 1 73 81 (69-89) 41 (24-59) 74 (63-84) 50 (30-70) 1.4 (1.0-1.9) 0.5 (0.2-0.9)
Subpleural consolidations ≥ 2 57 65 (52-76) 56 (38-74) 76 (63-86) 43 (28-59) 1.5 (1.0-2.3) 0.6 (0.4-1.0)
EP
Dynamic linear air bronchograms ≥ 1 36 44 (32-57) 81 (64-93) 83 (67-94) 41 (29-54) 2.4 (1.1-5.1) 0.7 (0.5-0.9)
Dynamic linear air bronchograms ≥ 2 16 22 (13-33) 97 (84-100) 94 (70-100) 37 (27-48) 7.1 (1.0-51.1) 0.8 (0.7-0.9)
C

Dynamic linear air bronchograms ≥ 1 81

AC

90 (80-96) 34 (19-53) 74 (64-83) 61 (36-83) 1.4 (1.1-1.8) 0.3 (0.1-0.7)

or subpleural consolidations ≥ 1
Dynamic linear air bronchograms ≥ 1 28
35 (24-48) 88 (71-97) 86 (67-96) 39 (28-51) 2.8 (1.1-7.5) 0.7 (0.6-0.9)
and subpleural consolidations ≥ 1
 ACCEPTED MANUSCRIPT

VAP, Ventilator-Acquired Pneumonia; PPV, Positive Predictive Value; NPV, Negative Predictive Value; LR+, Positive Likelihood
Ratio; LR-, Negative Likelihood Ratio; PCT, Procalcitonine; Positive culture of quantitative endotracheal aspirate, culture ≥ 105
colony-forming units/mL; Lobar/hemilobar consolidations, at least one tissue-like pattern area; Subpleural consolidations, at least
one (≥ 1) or two (≥ 2) areas with small subpleural consolidations; Dynamic linear air bronchogram, at least one (≥ 1) or two (≥ 2)
consolidated areas with such an air-bronchogram in.

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 5. Diagnostic performances of lung ultrasound combined with endotracheal aspirate for the diagnosis of VAP

Sensitivity Specificity PPV NPV +LR -LR

Subpleural consolidation >1 and
positive EAgram 45 (10-33) 93 (80-99) 89 (72-98) 50 (37-63) 5.0 (1.7-15.3) 0.6 (0.5-0.8)

PT
Dynamic linear air bronchogram >1
20 (22-45) 88 (71-97) 85 (55-98) 41 530-53) 3.3 (0.8-14.0) 0.9 (0.7-1.0)

RI
and positive EAgram

Dynamic linear air bronchogram > 2

SC
and positive EAgram 18 (7-33) 100 (89-100) 100 (59-100) 50 (37-63) 0.8 (0.7-1.0)

Dynamic linear air bronchogram > 2 20 (10-33) 100 (89-100) 100 (69-100) 42 (31-55) 0.8 (0.7-0.9)

U
and positive EAquant

AN
Dynamic linear air bronchogram or
subpleural consolidation ≥ 1 and 45(32-59) 88 (72-97) 86 (68-96) 49 (36-62) 3.8 (1.4-8.8) 0.6 (0.5-0.8)

M
positive EAgram

Dynamic linear air bronchogram or

D
subpleural consolidation ≥ 1 and 59(46-7&) 91 (77-98) 93 (80-98) 55 (42-68) 6.9 (2.3-20.8) 0.4 (0.3-0.6)

TE
positive EAquant

Dynamic linear air bronchogram and

EP
subpleural consolidation ≥ 1 and 20(10-33) 97 (84-100) 92 (62-100) 42 (31-54) 6.6 (0.9-48.9) 0.8 (0.7-0.9)
positive EAgram
C

Dynamic linear air bronchogram and

20(11-32) 97 (85-100) 93 (66-100) 49 (29-54) 7.1 (1.0-52.9) 0.8 (0.7-0.9)
AC

subpleural consolidation ≥ 1 and

positive EAquant

VPLUS-EAquant >4
Purulent secretions and ≥ 1 Dynamic
48 (35-62) 97 (83-100) 97 (82-100) 49 (36-63) 14.5(2.1-101.3) 0.5 (0.4-0.7)
linear air bronchogram and ≥ 2
 ACCEPTED MANUSCRIPT

subpleural consolidations and

positive EAgram

VPLUS-EAgram >4
Purulent secretions and ≥ 1 Dynamic
linear air and ≥ 2 subpleural 57 (42-70) 96 (82-100) 97 (83-100) 54 (39-68) 15.9(2.3-110.2) 0.4 (0.3-0.6)

PT
consolidations and positive EAquant

VAP, Ventilator-acquired Pneumonia; PPV, Positive Predictive Value; NPV, Negative Predictive Value; LR+, Positive Likelihood

RI
Ratio; LR-, Negative Likelihood Ratio; PCT, Procalcitonine; EA, Endotracheal aspirate. Positive EAgram, positive direct gram
stain examination of endotracheal aspirate. Positive EAquant. culture ≥ 105 Colony-forming units/mL of endotracheal aspirate;

SC
Lobar/hemilobar consolidations, at least one tissue-like pattern area; Subpleural consolidations, at least one (≥ 1) or two (≥ 2)
areas with small subpleural consolidations; Dynamic linear air-bronchogram, at least one (≥ 1) or two (≥ 2) consolidated areas with
such a air-bronchogram in.

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Table 6. Diagnostic performances of CPIS, CPIS EAgram, CPIS EAquant and VPLUS, VPLUS EAgram, VPLUS EAquant

Sensitivity Specificity PPV NPV +LR -LR

PT
CPIS
CPIS ≥ 5 84 (72-91) 16 (5-33) 67 (56-77) 31 (11-59) 1.0 (0.8-1.2) 1.1 (0.4-2.8)

RI
CPIS ≥ 6 68 (55-78) 50 (32-68) 74(62-84) 42 (26-59) 1.4 (0.9-2.0) 0.7 (0.4-1.1)

SC
CPIS ≥ 7 37 (25-49) 78 (60-91) 78 (60-91) 37 (25-49) 1.7 (0.8-3.5) 0.8 (0.6-1.0)
CPIS EAgram ≥ 5 90 (79-96) 13 (4-31) 67 (55-77) 40 (12-74) 1.0 (0.9-1.2) 0.8 (0.2-2.5)

U
CPISEAgram ≥ 6 84 (73-93) 47 (28-66) 75 (63-85) 61 (38-80) 1.6 (1.1-2.3) 0.3 (0.2-0.7)
CPISEAgram ≥ 7 67 (54-79) 63 (44-80) 78 (64-88) 50 (33-67) 1.8 (1.1-3.0) 0.5 (0.3-0.8)

AN
CPIS EAquant≥ 6 94 (84-99) 43 (24-63) 76 (64-86) 80 (52-96) 1.6 (1.2-2.3) 0.1 (0.0-0.4)
CPISEAquant ≥ 7 74 (60-85) 68 (48-84) 81 (67-91) 58 (39-74) 2.3 (1.3-4.0) 0.4 (0.2-0.7)

M
CPIS EAquant≥ 8 55 (40-68) 75 (55-89) 81 (64-92) 47 (32-62) 2.2 (1.1-4.3) 0.6 (0.4-0.9)

D
VPLUS

VPLUS ≥ 1
TE
93 (84-98) 34 (19-53) 75 (64-84) 69 (41-89) 1.4 (1.1-1.8) 0.2 (0.1-0.6)
EP
VPLUS ≥ 2 71 (58-81) 69 (50-84) 83 (71-91) 52 (36-68) 2.3 (1.1-3.9) 0.4 (0.3-0.7)
VPLUS ≥ 3 41 (29-54) 84 (67-95) 85 (68-95) 40 (29-53) 2.6 (1.1-6.2) 0.7 (0.5-0.9)
C

VPLUSEAgram ≥ 2 90 (79-96) 50 (31-69) 78 (66-87) 71 (48-89) 1.8 (1.2-2.6) 0.2 (0.1-0.5)

AC

VPLUSEAgram ≥ 3 78 (65-88) 77 (58-90) 87 (74-94) 64 (46-79) 3.3 (1.7-6.5) 0.3 (0.2-0.5)

VPLUSEAgram ≥ 4 48 (35-62) 97 (83-100) 97 (82-100) 49 (36-63) 14.5 (2.1-101.3) 0.5 (0.4-0.7)
VPLUSEAquant ≥ 2 96 (87-100) 46 (28-66) 77 (65-87) 87 (60-98) 1.8 (1.3-2.5) 0.1 (0.0-0.3)
VPLUS EAquant≥ 3 83 (70-92) 79 (59-92) 88 (76-96) 71 (52-86) 3.9 (1.9-8.0) 0.2 (0.1-0.4)
 ACCEPTED MANUSCRIPT

VPLUS EAquant≥ 4 57 (42-70) 96 (82-100) 97 (83-100) 54 (39-68) 15.9 (2.3-110.2) 0.4 (0.3-0.6)
CPIS, Clinical Pulmonary Infection Score. VPLUS, Ventilator-associated Pneumonia Lung Ultrasound Score. EA, Endotracheal
aspirate. Positive EAgram, positive direct gram stain examination of endotracheal aspirate. Positive EAquant, culture ≥ 105 Colony-
forming units/mL of endotracheal aspirate. EAgram was used to calculate the CPIS-EAgram and VPLUS-EAgram

PT
RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC

View publication stats