Anda di halaman 1dari 9

Received: 14 April 2017 | Revised: 27 June 2017 | Accepted: 11 July 2017

DOI: 10.1002/jcu.22526

RESEARCH ARTICLE

Pulmonary ultrasound scoring system for intubated


critically ill patients and its association with clinical
metrics and mortality: A prospective cohort study

David M. Tierney MD1 | Lori L. Boland MPH2 | Josh D. Overgaard MD1 |


Joshua S. Huelster MD3 | Ann Jorgenson RN2 | James P. Normington MS2 |
Roman R. Melamed MD3

1
Department of Graduate Medical
Education, Abbott Northwestern Hospital,
Abstract
Minneapolis, Minnesota
Purpose: Pulmonary ultrasound (PU) examination at the point-of-care can rapidly identify the eti-
2
Division of Applied Research, Allina Health,
ology of acute respiratory failure (ARF) and assess treatment response. The often-subjective
Minneapolis, Minnesota
3
classification of PU abnormalities makes it difficult to document change over time and communi-
Department of Critical Care, Abbott
Northwestern Hospital, Minneapolis, cate findings across providers. The study goal was to develop a simple, PU scoring system that
Minnesota would allow for standardized documentation, have high interprovider agreement, and correlate
with clinical metrics.
Correspondence
David Tierney, Department of Medical Methods: In this prospective study of 250 adults intubated for ARF, a PU examination was per-
Education, Abbott Northwestern Hospital, formed at intubation, 48-hours later, and at extubation. A total lung score (TLS) was calculated.
800 East 28th Street, Minneapolis, MN
Clinical metrics and final diagnosis were extracted from the medical record.
55407.
Email: david.tierney@allina.com Results: TLS correlated positively with mortality (P 5 .03), ventilator hours (P 5 .003), intensive
care unit, and hospital length of stay (P 5 .003, P 5 .008), and decreasing PaO2/FiO2 (P < .001).
Agreement of PU findings was very good (kappa 5 0.83). Baseline TLS and subscores differed sig-
nificantly between ARF categories (nonpulmonary, obstructive, and parenchymal disease).

Conclusions: A quick, scored, PU examination was associated with clinical metrics, including mor-
tality among a diverse population of patients intubated for ARF. In addition to diagnostic and
prognostic information at the bedside, a standardized and quantifiable approach to PU provides
objectivity in serial assessment and may enhance communication of findings between providers.

KEYWORDS
intubation, point-of-care ultrasound, pulmonary ultrasonography, respiratory insufficiency

1 | INTRODUCTION Acquisition, interpretation, and integration of PU findings at an iso-


lated point in time are essential to prompt and accurate diagnosis.
Pulmonary ultrasound (PU) has become an essential tool for rapidly Tracking PU changes over time is equally important in confirming a
identifying the etiology of acute respiratory failure (ARF), following diagnosis and adjusting treatment. To do so requires a standardized
treatment progress, and clarifying nonspecific chest radiograph (CXR) approach to PU such that providers can document and agree not only
1–5
abnormalities among critically ill patients, and with test characteris- with themselves but also with each other over time.11–14
6,7
tics better than the clinical examination and CXR. When used in PU scoring models have been developed to meet the need for
combination with cardiac and vascular ultrasound, it can enhance the standardization and have been shown to correlate with various metrics
understanding of etiology3 and may reduce the need for CXR or chest in specific patient populations.5,11,15–17 Scoring systems correlate with
8–10
CT. mortality in patients with acute respiratory distress syndrome

14 | V
C 2017 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/jcu J Clin Ultrasound. 2018;46:14–22.
TIERNEY ET AL. | 15

(ARDS),17,18 on hemodialysis,19 with dyspnea and/or chest pain,15 and 2.3 | Lung zone classification criteria
with congestive heart failure (CHF),20 and PU can be used to quantify
The chest was divided into 9 zones (Figure 2), based on a modification
extravascular lung water,16,21–24 which predicts outcomes in critically ill
of previously established protocols,5,36 to approximate lobar anatomy
patients.4,15,16,25–35
of the lung and reflect our clinical workflow in intubated patients. Each
An ideal examination and scoring model would (1) have good intra-
zone was classified based on an overall impression from an area no
rater and interrater agreement, (2) be rooted in a routine clinical PU
larger than 7.5 3 5 cm. Classification of findings (Figure 3) could be
workflow, (3) be technically feasible, (4) be quick to perform and docu-
accomplished without advanced computer technology37 or the need to
ment, but (5) be sensitive enough to detect significant but subtle
freeze an image and count abnormalities.
pathology, and (6) yield a score that has clinical predictive value.
Similar to a previous study,15 each zone was classified as follows:
The primary objective of this study was to establish a PU workflow
and scoring system in a large, diagnostically diverse group of intubated  A 5 lung sliding, presence of A-lines, and lack of B-lines or
patients that met the above criteria and was associated with clinical consolidation
outcomes.
 B1 5 1 to 3 B-lines present per intercostal space
 B2 5 quantity of B-lines between B1 and B3
2 | MATERIALS AND METHODS  B3 5 confluent B-lines occupying >50% of an intercostal space

2.1 | Setting and study design Areas of nonaerated lung >3 cm in its axis perpendicular to pleura with
dynamic air bronchograms and lack of compressive reason for aeration
This prospective cohort study took place at a 640-bed (62 intensive
loss (eg, pleural effusion, elevated diaphragm) were classified as nona-
care unit [ICU] beds) quaternary care center and enrolled 250 adult
telectatic consolidation. Areas of nonaerated lung >3 cm in its axis per-
inpatients between December 2013 and March 2015. All intubated
pendicular to pleura with distinct features suggesting volume loss
ICU patients who had provided consent for use of their electronic
atelectasis, such as a pleural effusion size consistent with volume of
health record data for research purposes were screened for eligibility
tissue-like lung, lack of dynamic air bronchograms, typical atelectatic
by a research nurse. Patients were excluded if they were intubated
distribution in the inferior, lateral tip of lung with a smooth re-aeration
solely for airway protection, had a history of pleurodesis, pneumonec-
line, were classified as atelectasis. A combined classification of atelecta-
tomy, or video-assisted thoracoscopic surgery, had been extubated or
sis/consolidation was assigned when the area examined consisted of
had been intubated for >24 hours before a first PU examination, had tissue-like density and the physician was unable to see discriminating
been transitioned to comfort care, or were imminently dying (Figure 1). features indicative of either specific entity. Zones with small, focal, <3-
Because of physician availability, daily enrollment was limited to the 3 cm areas of peripheral consolidation were classified as small consolida-
most recently intubated patients each morning. The protocol was tions. The pleural effusion classification was added to each zone when
approved by an institutional review board (Schulman Associates IRB, present.
Inc, Approval No. 4080-2E).

2.4 | Scoring system


2.2 | Ultrasound examination protocol An individual lung zone received 0 points for an A classification, 1 point

PU examinations were performed at 3 time points during a patient’s for a B1, 2 points for a B2, and 3 points for any of the following: B3,
consolidation, atelectasis, or small consolidations. Each zone received
ICU stay: (1) as soon after intubation as possible, (2) 48 hours after the
an additional point if effusion was present. The maximum score per
initial examination, and (3) at extubation if the patient’s previous exami-
zone was 4 points. Isolated atelectasis without other parenchymal
nations did not already coincide with extubation. The patient was posi-
abnormalities existing in zone 3 or 8 only received 1 point (Figure 3).
tioned supine with the head of the bed elevated 208 to 408. All
Total lung score (TLS) was the summation of all points across the 9
examinations were performed with a SonoSite EDGE portable ultra-
lung zones (possible range 5 0–36). Total B score represented the sum
sound system equipped with a P21 (1–5 MHz) phased-array transducer
of only the B-line points (B1, B2, B3) (range 5 0–27), and total atelecta-
(FUJIFILM SonoSite Inc, Bothell, Washington).
sis/consolidation score was the sum of points assigned for atelectasis
Four physicians with greater than 3 years of PU experience per-
and consolidation classifications across zones. Total anterior score
formed all examinations. They participated in a prestudy 1-hour train-
included points from zones 1, 2, 6, 7 and total posterior score from
ing session to ensure uniformity of the pulmonary classifications and
zones 3, 5, 8, 9.
then independently scored 64 virtual patients with videoclips. Overall
and entity-specific interrater agreements were calculated.
2.5 | Data collection
Physicians performing the PU examinations were not involved in
the patient’s care and were blinded to history, laboratory data, imaging, Independent of previous examinations, the physician performing the
treatments, and diagnosis. They were instructed not to examine medi- PU examination entered findings into an electronic data collection tool
cation drips or ventilator settings during the ultrasound examination. after leaving the patient’s room. Relevant clinical data were abstracted
16 | TIERNEY ET AL.

FIGURE 1 Study patient flow. Abbreviations: ALI, acute lung injury; ARDS, acute respiratory distress syndrome; CHF, congestive heart
failure; COPD, chronic obstructive pulmonary disease; PU, pulmonary ultrasound; VATS, video-assisted thoracoscopic surgery

from the electronic health record by a research nurse. A final diagnosis 2.6 | Statistical analysis
for each patient’s ARF was determined via review of clinical notes, for-
Patient and event characteristics were described using means or
mal imaging, laboratory values, and the care team’s final diagnosis after
medians, and proportions. Subscores of TLS were expressed as mean
discharge, which was conducted by study physicians who did not per-
values and as the mean percentage of TLS and were examined overall
form the ultrasound examinations and were blinded to examination
results. and by etiology of ARF.
TIERNEY ET AL. | 17

FIGURE 2 Locations and numbering of the examination zones. Abbreviations: AAx, anterior axillary line; MC, midclavicular line; PAx,
posterior axillary line

Mortality, ICU, and hospital length of stay (LOS), PaO2/FiO2 ratio, pulmonary disease [COPD]), and parenchymal disease (eg, pneumonia,
and ventilator hours were examined across tertiles of TLS, with ARDS, aspiration, and CHF), with one-way analysis of variance used to
Cuzick’s trend test used to assess trends across tertiles. assess whether disease category explained a significant amount of the
Lung scores were also examined by the common clinical respira- variation in scores.
tory disease categories of no pulmonary disease (eg, altered level of Fleiss’s kappa coefficient was used to assess interrater agreement
consciousness, upper airway obstruction, and neuromuscular weak- in the prestudy training set. All analyses were conducted using Stata
ness), obstructive disease (eg, asthma and chronic obstructive 14.1 (StataCorp, LP, College Station, Texas).

FIGURE 3 Lung zone classification and associated point assignment


18 | TIERNEY ET AL.

T AB LE 1 Patient characteristics overall and by etiology of acute respiratory failure

% In-hospital
Variable N Age % Male Ventilator days P/F ratio at intubation ICU LOS Hospital LOS mortality

All patients 250 65 (23–93) 48 4.1 (0.1–34.4) 222.5 (63–584) 7.8 (1.0–40.8) 13 (2–58) 24

Diagnosis
Pneumonia 76 66 (28–93) 45 5.0 (0.5–15.8) 184 (88–494) 9.9 (1.7–38.6) 14 (3–41) 18
Aspiration 34 72 (49–91) 47 4.3 (0.1–34.4) 232 (86–483) 8.0 (1.3–30.8) 15 (2–39) 24
Sepsis 29 67 (30–87) 59 3.7 (0.5–13.2) 233 (158–530) 8.3 (1.4–25.4) 15 (2–58) 21
CHF 27 71 (50–89) 44 2.3 (0.8–15.0) 218 (68–417) 5.8 (1.7–21.8) 10 (2–25) 30
Cardiac arrest 22 67 (31–79) 45 4.6 (0.3–13.8) 254 (64–584) 7.0 (1.5–20.5) 12 (2–48) 45
ALI/ARDS 18 61 (30–78) 39 7.2 (0.6–13.5) 161 (63–336) 11.3 (3.1–36.9) 15 (7–37) 33
COPD/asthma 17 64 (55–87) 59 2.3 (1.3–11.6) 234 (120–348) 4.3 (1.9–12.1) 9 (2–12) 24
Othera 27 65 (23–88) 52 3.9 (0.9–10.8) 248 (75–474) 7.6 (1.0–40.8) 15 (2–51) 19

Abbreviations: ARF, acute respiratory failure; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; ICU, intensive care unit;
LOS, length of stay; P/F, PaO2/FiO2.
Results are expressed as median (range) or % as appropriate.
a
Other: acute myocardial infarction (1.6%), altered level of consciousness (2.0%), atelectasis (0.8%), neuromuscular weakness (0.8%), pulmonary embo-
lism (0.8%), pulmonary fibrosis (0.8%), blastomycosis (0.4%), cryptogenic organizing pneumonia (0.4%), Kaposi sarcoma (0.4%), obstructive sleep apnea
(0.4%), pleural effusion (0.4%), pneumothorax (0.4%), pulmonary aspergillosis (0.4%), upper airway obstruction (0.4%), and pulmonary vasculitis (0.4%).

3 | RESULTS respectively) with the highest scores in the parenchymal group (Table
3). The 8 patients in the no pulmonary disease group had a nonpulmo-
A total of 1308 patients were screened for enrollment, of which 1014 nary etiology for intubation on final chart review not apparent at the
(77%) did not meet inclusion criteria (Figure 1). The most common diag- time of initial inclusion.
nosis for ARF was pneumonia (Table 1). Average time to perform the PU examination on 10 randomly
TLS for baseline examinations ranged from 0 to 31 points (mean 5 selected days (n 5 28 examinations) was 135 seconds (range 5 88–
12), with the highest mean observed in those with ARDS and the low- 205). Among the 64 virtual patients scored by the 4 participating
est in patients with COPD/asthma (Table 2). Posterior lung scores con- physicians during prestudy training, interrater agreement was very
tributed a greater proportion to the TLS than anterior lung scores in all good for TLS (kappa 5 0.83), almost perfect for assignment of the A
diagnostic groups except ARDS. The contribution of B score to TLS classification (kappa 5 0.96), very good for atelectasis/consolidation
was greater than atelectasis/consolidation across all diagnostic catego- (kappa 5 0.84), and substantial for the B1, B2, and B3 classifications
ries, especially in CHF, ARDS, and aspiration (Table 2). (kappa 5 0.77, 0.61, 0.74, respectively).
Tertiles of baseline TLS (0–9, 9.5–18, 18.5–36) were positively
associated with mortality (P 5 .03), median ICU (P 5 .003), and hospital 4 | DISCUSSION
LOS (P 5 .008), and ICU ventilator hours (P 5 .003), and inversely asso-
ciated with baseline PaO2/FiO2 ratio (P < .001) (Figure 4). Respiratory failure is one of the most common conditions in the ICU
Patients in the 3 clinical groupings had significantly different total and often requires chest imaging for accurate diagnosis.38,39 Physicians
lung, B, anterior, and posterior scores (P < .001, .005, <.001, .002, need to make an accurate diagnosis, define disease severity, assess

T AB LE 2 Mean total and sub lung scores at baseline, overall, and by etiology of acute respiratory failure

Variable N Total lung score B score Atl/Consol score Anterior score Posterior score

All patients 250 12.0 8.6 (70) 2.5 (22) 5.8 (42) 6.1 (58)

Diagnosis
ARDS 18 20.3 16.1 (79) 3.2 (16) 12.1 (59) 8.2 (41)
CHF 27 14.0 10.5 (72) 2.1 (16) 6.3 (39) 7.6 (61)
Pneumonia 76 12.9 8.7 (66) 3.3 (27) 6.5 (47) 6.4 (53)
Cardiac arrest 22 11.2 7.5 (67) 3.0 (26) 5.6 (42) 5.5 (58)
Other 27 10.6 8.0 (71) 1.8 (21) 5.0 (35) 5.5 (65)
Aspiration 34 9.6 7.4 (78) 1.5 (14) 4.4 (40) 5.3 (60)
Sepsis 29 9.4 5.5 (54) 3.1 (37) 4.2 (34) 5.2 (66)
COPD/asthma 17 8.1 7.6 (96) 0.2 (2) 2.9 (33) 5.1 (67)

Abbreviations: ARDS, acute respiratory distress syndrome; Atl/Consol, atelectasis/consolidation; CHF, congestive heart failure; COPD, chronic obstruc-
tive pulmonary disease.
Results are expressed as mean or mean (% total lung score).
TIERNEY
ET AL.

FIGURE 4 Clinical metrics by total lung score tertile. Abbreviations: Hosp, hospital; ICU, intensive care unit; LOS, length of stay; P/F, PaO2/FiO2; Vent, ventilator
|
19
20 | TIERNEY ET AL.

response to treatment, and prognosticate, and PU represents an was no significant difference in the predictive power of our score when
emerging technique for accomplishing these in a safe, timely, and cost- atelectasis, consolidation, or small consolidations resulted in an extra
efficient manner. Portable chest radiograph has significant limitations7 point beyond the B3 classification (ie, the zone was given a total of 4
and potential risks (eg, accidental removal of lines/tubes), whereas points rather than 3) versus if each of these entities received 3 points
chest CT has notable drawbacks, such as increased radiation, cost, and for simplicity. By assigning 1 point for each zone with effusion (with
the need to transport the patient from the ICU.9,10,40 patients positioned semi-upright), a feasible and semi-quantitative
Small studies of PU scoring systems in disease-specific patient assessment of pleural effusion was incorporated into the TLS. The mod-
populations and their association to various clinical metrics ification of isolated atelectasis in the lateral caudal lung zone on the
exist,5,11,15–20 as does a large (260 patients) study proposing a standar- posterior axillary line receiving 1 point instead of 3 resulted in improved
dized diagnostic system based on lung ultrasound profiles.1 However, predictive power across metrics and is consistent with the limited sig-
to our knowledge, a large prospective study of a simple lung ultrasound nificance this isolated finding has in the clinical environment.
scoring system and its association to clinical outcomes including mor- The association of this PU scoring system to mortality in a large
tality across a wide spectrum of clinical disease has not been previously population of intubated ICU patients with ARF is important from a clin-
reported. The current approach was studied in 250 patients from medi- ical standpoint, because various small studies have only demonstrated
cal/surgical, neurological, and cardiovascular ICUs, and thus the results associations of specific PU findings with outcomes in specific disease
are highly applicable to a diverse ICU population. Furthermore, this 9- states.15,17–20 Non–ultrasound-based models predict mortality among
point examination, taking on average 2 to 3 minutes to perform, can be ICU patients; however, they typically rely on a combination of physio-
feasibly integrated into the daily physical examination of patients in logic variables and comorbid conditions, and few exist specifically for
their typical positioning. intubated patients.42,43 Beyond the observed association with mortal-
The 9 lung zones included in this algorithm reflect a clinical examina- ity, potential use of the TLS in predicting ventilator hours and LOS may
tion in an applied setting, rather than a workflow designed for a research be useful in guiding family expectations, patient flow, and quality meas-
protocol. There are no formally endorsed approaches to a clinical PU ures tied to LOS, but this requires further study.
examination, although both an 8- and 28-point examination have been Some limitations to our study exist. To preserve efficiency, we exam-
5
described. A clinical PU examination sequence often involves concep- ined a limited surface area of the lung, and this may have led to overesti-
tual lobar anatomy, and thus the right lateral caudal area was divided into mation of a process that only existed under the transducer, or the
both an anterior and a posterior zone over the right middle and lower inability to visualize a pulmonary process present just outside the area of
lobes, respectively. Designation of an “extra” zone in the right lung, in examination. Our interobserver agreement was calculated based solely
addition to being clinically relevant, can be physiologically rationalized on image review, thus dismissing the variability that can exist in image
based on the greater total lung capacity of the right lung.41 Of interest, acquisition. Physicians performing the examination were blinded to
3.9% of our patients with aspiration and pneumonia had findings present patient data, but there were opportunities for this blinding to have been
only in the right lateral caudal zone on the anterior axillary line that incomplete and possibly influence their assessment (eg, medication infu-
would have been missed if that zone was not included in the examina- sions visible at the bedside, other provider discussions in/around the
tion. A sensitivity analysis was conducted both with and without this patient room). Having a single provider perform the repeat examinations
additional zone included, and similar associations between lung scores on an individual patient, although meant to ensure consistency, may
and outcomes were observed with both methods. have introduced bias with the examiner knowing the previous location of
Many previous studies have examined the association of the pres- findings. Despite these instances of potential “unblinding,” they all do
ence of B-lines to various outcomes. However, in this study, ascending reflect the reality of PU in the clinical setting. Retrospective chart review
points (0 to 3) were assigned to the continuum of lung abnormality clas- was used to determine a final diagnosis, but, as can be the case in ARF
sifications from the normal, fully aerated A classification (0 points) to patients, more than 1 etiology may have been contributing, or the diag-
the least aerated consolidation and atelectasis (3 points). Of note, there nosis was not definitive. The number of extubation examinations more

T AB LE 3 Various lung scores by clinical groupings

Clinical grouping (n) Total lung score B score Anterior score Posterior score

No pulmonary disease (8) 4.3 2.5 0.9 3.4

Obstructive disease (17) 8.1 7.6 2.9 5.1

Parenchymal disease (155) 13.2 9.6 6.6 6.7

P value a
<.001 .005 <.001 .002

“No pulmonary disease” 5 altered level of consciousness, neuromuscular weakness, and upper airway obstruction. “Obstructive disease” 5 COPD and
asthma. “Parenchymal disease” 5 pneumonia, CHF, aspiration, ALI/ARDS.
a
One-way analysis of variance.
TIERNEY ET AL. | 21

than 48 hours after the initial examination was small (n 5 30) because of [9] Ott LK, Hoffman LA, Hravnak M. Intrahospital transport to the Radiol-
ICU workflow and difficulty coordinating a blinded US examination with ogy Department: risk for adverse events, nursing surveillance, utiliza-
tion of a MET and practice implications. J Radiol Nurs. 2011;30:49–52.
the treatment team. This prevented us from adequately assessing TLS
[10] Waydhas C. Intrahospital transport of critically ill patients. Crit
change from baseline to extubation.
Care. 1999;3:R83–89.
The differential diagnosis for ARF is often difficult to narrow
[11] Anderson KL, Fields JM, Panebianco NL, et al. Inter-rater reliability
based on history, physical examination, and laboratory data alone. PU
of quantifying pleural B-lines using multiple counting methods.
is immediately available at the bedside and can provide helpful diag- J Ultrasound Med. 2013;32:115–120.
nostic information in these patients, especially when combined with [12] Sperandeo M, Trovato GM, Catalano D. Quantifying B-lines on lung
cardiac and vascular US. This simple PU scoring system, which pro- sonography: insufficient evidence as an objective, constructive, and
vides for standardized quantification and communication of PU find- educational tool. J Ultrasound Med. 2014;33:362–365.
ings between examinations and providers, had good agreement [13] Cibinel GA, Casoli G, Elia F, et al. Diagnostic accuracy and reprodu-
among providers, was quick to perform, and correlated well with cibility of pleural and lung ultrasound in discriminating cardiogenic
causes of acute dyspnea in the emergency department. Intern
important clinical outcomes.
Emerg Med. 2012;7:65–70.
[14] Chiem AT, Chan CH, Ander DS, Kobylivker AN, Manson WC. Com-
AC KNOW LE DGME NT parison of expert and novice sonographers’ performance in focused
lung ultrasonography in dyspnea (FLUID) to diagnose patients with
Thanks to Anne Uttermark, RRT for coordination of the respiratory
acute heart failure syndrome. Acad Emerg Med. 2015;22:564–573.
therapy’s involvement in the study.
[15] Frassi F, Gargani L, Tesorio P, et al. Prognostic value of extravascu-
lar lung water assessed with ultrasound lung comets by chest
C ONFLICT OF INT E RE ST sonography in patients with dyspnea and/or chest pain. J Card Fail.
2007;13:830–835.
This manuscript has not been previously published and is not under
[16] Enghard P, Rademacher S, Nee J, et al. Simplified lung ultrasound
consideration in the same or substantially similar form in any other
protocol shows excellent prediction of extravascular lung water in
peer-reviewed media. All authors listed have contributed sufficiently ventilated intensive care patients. Crit Care. 2015;19:36.
to the project to be included as authors. No conflict of interest,
[17] Zhao Z, Jiang L, Xi X, et al. Prognostic value of extravascular lung
financial or otherwise, exists for any of the authors. water assessed with lung ultrasound score by chest sonography in
patients with acute respiratory distress syndrome. BMC Pulm Med.
2015;15:98.
OR CID
[18] Li L, Yang Q, Li L, et al. [The value of lung ultrasound score on eval-
David M. Tierney MD http://orcid.org/0000-0003-4202-1101 uating clinical severity and prognosis in patients with acute respira-
tory distress syndrome]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue.
R E FER E NCE S 2015;27:579–584.

zière GA. Relevance of lung ultrasound in the [19] Zoccali C, Torino C, Tripepi R, et al. Pulmonary congestion predicts car-
[1] Lichtenstein DA, Me
diac events and mortality in ESRD. J Am Soc Nephrol. 2013;24:639–
diagnosis of acute respiratory failure: the BLUE protocol. Chest.
646.
2008;134:117–125.
[20] Platz E, Lewis EF, Uno H, et al. Detection and prognostic value of
[2] Blaivas M. Lung ultrasound in evaluation of pneumonia. J Ultrasound
pulmonary congestion by lung ultrasound in ambulatory heart fail-
Med. 2012;31:823–826.
ure patients. Eur Heart J. 2016;37:1244–1251.
[3] Bataille B, Riu B, Ferre F, et al. Integrated use of bedside lung ultra-
[21] Noble VE, Murray AF, Capp R, et al. Ultrasound assessment for
sound and echocardiography in acute respiratory failure: a prospec-
extravascular lung water in patients undergoing hemodialysis. Time
tive observational study in ICU. Chest. 2014;146:1586–1593.
course for resolution. Chest. 2009;135:1433–1439.
[4] Bilotta F, Giudici LD, Zeppa IO, et al. Ultrasound imaging and use of zière G, Biderman P, Gepner A, Barre  O. The
[22] Lichtenstein D, Me
B-lines for functional lung evaluation in neurocritical care: a pro- comet-tail artifact: an ultrasound sign of alveolar-interstitial syn-
spective, observational study. Eur J Anaesthesiol. 2013;30:464–468.
drome. Am J Respir Crit Care Med. 1997;156:1640–1646.
[5] Volpicelli G, Elbarbary M, Blaivas M, et al. International evidence- [23] Picano E, Frassi F, Agricola E, et al. Ultrasound lung comets: a clini-
based recommendations for point-of-care lung ultrasound. Intensive cally useful sign of extravascular lung water. J Am Soc Echocar-
Care Med. 2012;38:577–591. diogr. 2006;19:356–363.
[6] Pivetta E, Goffi A, Lupia E, et al. Lung ultrasound-implemented diag- [24] Jambrik Z, Gargani L, Adamicza A,~ et al. B-lines quantify the lung
nosis of acute decompensated heart failure in the ED: a SIMEU water content: a lung ultrasound versus lung gravimetry study in
multicenter study. Chest. 2015;148:202–210. acute lung injury. Ultrasound Med Biol. 2010;36:2004–2010.
[7] Xirouchaki N, Magkanas E, Vaporidi K, et al. Lung ultrasound in crit- [25] Lubrano R, Cecchetti C, Elli M, et al. Prognostic value of extravascu-
ically ill patients: comparison with bedside chest radiography. Inten- lar lung water index in critically ill children with acute respiratory
sive Care Med. 2011;37:1488–1493. failure. Intensive Care Med. 2011;37:124–131.
[8] Fanara B, Manzon C, Barbot O, Desmettre T, Capellier G. Recom- [26] Sakka SG, Klein M, Reinhart K, Meier-Hellmann A. Prognostic value of
mendations for the intra-hospital transport of critically ill patients. extravascular lung water in critically ill patients. Chest. 2002;122:
Crit Care. 2010;14:R87. 2080–2086.
22 | TIERNEY ET AL.

[27] Yang CS, Qiu HB, Liu SQ, et al. [The prognostic value of extrava- [37] Brattain LJ, Telfer BA, Liteplo AS, Noble VE. Automated B-line scor-
scular lung water index in critically ill septic shock patients]. Zhong- ing on thoracic sonography. J Ultrasound Med. 2013;32:2185–
hua Nei Ke Za Zhi. 2006;45:192–195. 2190.
[28] Zhang Z, Lu B, Ni H. Prognostic value of extravascular lung water [38] Lichtenstein DA. BLUE-protocol and FALLS-protocol: two applica-
index in critically ill patients: a systematic review of the literature.
tions of lung ultrasound in the critically ill. Chest. 2015;147:1659–
J Crit Care. 2012;27:420e1-e8.
1670.
[29] Bognar Z, Foldi V, Rezman B, Bogar L, Csontos C. Extravascular
lung water index as a sign of developing sepsis in burns. Burns. [39] Via G, Storti E, Gulati G, et al. Lung ultrasound in the ICU: from
2010;36:1263–1270. diagnostic instrument to respiratory monitoring tool. Minerva Anes-
tesiol. 2012;78:1282–1296.
[30] Chung FT, Lin SM, Lin SY, Lin HC. Impact of extravascular lung
water index on outcomes of severe sepsis patients in a medical [40] Gargani L, Picano E. The risk of cumulative radiation exposure in
intensive care unit. Respir Med. 2008;102:956–961. chest imaging and the advantage of bedside ultrasound. Crit Ultra-
[31] Chung FT, Lin HC, Kuo CH, et al. Extravascular lung water corre- sound J. 2015;7:4.
lates multiorgan dysfunction syndrome and mortality in sepsis. PLoS
[41] Johansen B, Bjørtuft O, Boe J. Static lung volumes in healthy sub-
One. 2010;5:e15265.
jects assessed by helium dilution during occlusion of one mainstem
[32] Craig TR, Duffy MJ, Shyamsundar M, et al. Extravascular lung water bronchus. Thorax. 1993;48:381–384.
indexed to predicted body weight is a novel predictor of intensive
care unit mortality in patients with acute lung injury. Crit Care Med. [42] Pannu SR, Moreno Franco P, Li G, et al. Development and valida-
2010;38:114–120. tion of severe hypoxemia associated risk prediction model in 1,000
mechanically ventilated patients. Crit Care Med. 2015;43:308–317.
[33] Davey-Quinn A, Gedney JA, Whiteley SM, Bellamy MC. Extravascu-
lar lung water and acute respiratory distress syndrome–oxygenation [43] Umegaki T, Nishimura M, Tajimi K, et al. An in-hospital mortality
and outcome. Anaesth Intensive Care. 1999;27:357–362. equation for mechanically ventilated patients in intensive care units.
[34] Martin GS, Eaton S, Mealer M, Moss M. Extravascular lung water in J Anesth. 2013;27:541–549.
patients with severe sepsis: a prospective cohort study. Crit Care.
2005;9:R74–82.
[35] Phillips CR, Chesnutt MS, Smith SM. Extravascular lung water in How to cite this article: Tierney DM, Boland LL, Overgaard JD,
sepsis-associated acute respiratory distress syndrome: indexing with et al. Pulmonary ultrasound scoring system for intubated crit-
predicted body weight improves correlation with severity of illness
ically ill patients and its association with clinical metrics and
and survival. Crit Care Med. 2008;36:69–73.
mortality: A prospective cohort study. J Clin Ultrasound.
[36] Lichtenstein DA, Mezière GA. The BLUE-points: three standardized
points used in the BLUE-protocol for ultrasound assessment of the 2018;46:14–22. https://doi.org/10.1002/jcu.22526
lung in acute respiratory failure. Crit Ultrasound J. 2011;3:109–110.