Research ArticleA Simplified Ultrasound Comet Tail Grading Scoring toAssess Pulmonary Congestion in Patients with Heart Failure

Hong Li1 Yi-Dan Li2 Wei-Wei Zhu2 Ling-Yun Kong2 Xiao-Guang Ye2 Qi-Zhe Cai2

Lan-Lan Sun2 and Xiu-Zhang Lu 2

1Department of Ultrasound Shanxi Provincial Peoplersquos Hospital Taiyuan 030012 China2Department of Echocardiography Beijing Chaoyang Hospital Capital Medical University Beijing 100020 China

Correspondence should be addressed to Xiu-Zhang Lu lxz echo163com

Received 7 August 2017 Revised 23 November 2017 Accepted 3 December 2017 Published 2 January 2018

Academic Editor Gianluca Di Bella

Copyright copy 2018 Hong Li et al This is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Ultrasound lung comets (ULCs) are a nonionizing bedside approach to assess extravascular lung water We evaluated a protocolfor grading ULC score to estimate pulmonary congestion in heart failure patients and investigated clinical and echocardiographiccorrelates of the ULC score Ninety-three patients with congestive heart failure admitted to the emergency department underwentpulmonary ultrasound and echocardiography A ULC score was obtained by summing the ULC scores of 7 zones of anterolateralchest scans The results of ULC score were compared with echocardiographic results the New York Heart Association (NYHA)functional classification radiologic score andN-terminal pro-b-type natriuretic peptide (NT-proBNP) Positive linear correlationswere found between the 7-zone ULC score and the following Ee1015840 systolic pulmonary artery pressure severity of mitralregurgitation left ventricular global longitudinal strain NYHA functional classification radiologic score and NT-proBNPHowever there was no significant correlation between ULC score and left ventricular ejection fraction left ventricle diameterleft ventricular volume or left atrial volume A multivariate analysis identified the Ee1015840 systolic pulmonary artery pressure andradiologic score as the only independent variables associated with ULC score increase The simplified 7-zone ULC score is arapid and noninvasive method to assess lung congestion Diastolic rather than systolic performance may be the most importantdeterminant of the degree of lung congestion in patients with heart failure

1 Introduction

In patients with congestive heart failure or intravascular vol-ume overload redistribution of fluids within the lungs leadsto pulmonary edema Excessive extravascular lung water(EVLW) accumulates in the interstitial and alveolar spacesdue to elevated left ventricular (LV) filling pressures [1] Themost accurate methods for assessing EVLW are expensiveinvasive and impractical for routine clinical practice [2]Thus EVLW is often assessed by chest X-ray althoughthe sensitivity is low [3] and includes the risk of radiationexposure

Lung ultrasound is a noninvasive and nonionizing imag-ing technique that has been previously proposed as a bedsidetool for evaluating pulmonary congestion in patients withheart failure [4] Lichtenstein and Meziere [5] reported thatmultiple anterior diffuse B-lines with lung sliding indicated

pulmonary edema with 97 sensitivity and 95 specificityHowever a standard method to quantify the volume ofEVLW by lung ultrasound has not been established sincethere is no defined method for grading the severity of signsthat are typical on lung ultrasound The 28-sector approachhas been applied to partially quantify lung congestion andshowed positive linear correlation with the radiologic lungwater score [6ndash9] Yet this comprehensive scoring method isinconvenient in daily clinical practice and counting B-lines(also known as ultrasound lung comets or ULCs) in heartfailure patients with lung congestion is confusing because B-lines often merge Therefore a simpler method is needed

Echocardiography is a noninvasive tool which providesa good understanding of the functional and hemodynamicinformation in heart failure patients The ability to evaluateULCs on lung ultrasound may provide further insight intothe clinical and pathophysiological involvement of the lungs

HindawiBioMed Research InternationalVolume 2018 Article ID 8474839 10 pageshttpsdoiorg10115520188474839

2 BioMed Research International

(a) (b) (c)

Figure 1 A linear probe was used to exclude noncardiac ULCs (a) Normal pleura line and cardiac ULCs (b c) The abnormal pleuralline could also generate ULCs (which are best visible under real-time examination) and they were confirmed by high-resolution computedtomography as interstitial lung disease and pneumonic infiltrate respectively

The present study assessed the performance of a simplifiedULC scoring system for evaluating pulmonary congestionand investigated clinical and echocardiographic correlates ofthe ULC score

2 Methods

21 Study Protocol This was a prospective study The EthicsCommittee of Beijing Chaoyang Hospital approved the pro-tocol and all patients provided informed consent

During a 6-month period from August 2015 to January2016 we studied patients with congestive heart failure whosatisfied the modified Framingham criteria [10] and wereadmitted to our emergency department All the patientspresentedwith shortness of breath andwere suspected ofNewYork Heart Association (NYHA) classification II III or IV[11]

Patients with the following were excluded age lt 18 yearsatrial fibrillation mitral stenosis and pulmonary disease Adiagnosis of pulmonary disease was based mainly on clinicalsymptoms chest radiograph and blood tests Also excludedwere patients with an abnormal pleural line observed bylinear probe because B-lines can arise due to interstitial lungdisease [12] or pneumonia [13] (Figure 1)

Cardiac and lung echographic examinations were per-formed before intravenous diuretic therapy All patients wereanalyzed in the supine near-to-supine or lateral positionAn experienced operator (HL) with 8 years of echographicexamination experience and 2 years of lung ultrasoundexperience performed the examinations using Philips CX50(Philips Ultrasound Bothell Washington USA) with S5-1 phased-array probe (1ndash5MHz) and L12-3 linear probe(3ndash12MHz)

Before the echographic examinations a bedside antero-posterior chest X-ray was performed with the patient inthe supine position Scoring of pulmonary congestion wasdetermined through a previously validated radiologic scoreincorporating assessment of variables (Table 1) [14] Thefilm was read by an experienced radiologist blinded to theultrasound and clinical findingsThe intra- and interobserver

Table 1 Radiologic score variables

Variables ScoreMild Moderate Severe

Hilar vesselsEnlarged 1 2 3Increased in density 2 4 6Blurred 3 6 9

Kerley linesA 4 8B 4 8C 4 8

Micronodules 4 8Widening of interlobar fissures 4 8 12Peribronchial and perivascularcuffs 4 8 12

Extensive perihilar haze 4 8 12Subpleural effusion 5 10Diffuse increase in density 5 10 15

reproducibility of the radiologic scoring among experiencedobservers were very high as previously described In addi-tion a sample of blood (5ml) was collected for blindedmeasurement of NT-proBNP

22 Transthoracic Echocardiographic Study All patientsunderwent transthoracic echocardiography examination atbedside In accordance with the recommendations of theAmerican Society of Echocardiography the LV end-diastolicand end-systolic diameters (LVEDD and LVESD resp) weremeasured from the M-mode trace obtained via a parasternallong-axis view Left ventricular end-diastolic and end-systolicvolumes (LVEDV and LVESV) ejection fraction (LVEF)and left atrial volume were obtained from 2-chamber and4-chamber views using the biplane Simpsonrsquos method andindexed to body surface area The peak Doppler velocitiesof early (E) and late (A) diastolic flow and the ratio of

BioMed Research International 3

Alveolus

Normal Flood-filled interstitium

Alveolar flooding

Capillary

Interstitium

(a) (b)

(c) (d)

Increased hydrostatic pressure

Figure 2 Increasing severity of interstitial or alveoli involvement (a) Normal lung B-lines are absent (b) Septal syndrome B-lines areabout 7mm apart corresponding to subpleural septa (c) Interstitial-alveolar syndrome B-lines are confluent (d) White lung B-lines havecoalesced resulting in an echographic lung field that is almost completely white

E to A (119864119860) were measured from the apical 4-chamberview A 15mm sample volume was placed at the septaland lateral corner of the mitral annulus An analysis wasalso performed for early (e1015840) and late diastolic velocityand the average Ee1015840 ratio was calculated The severity ofmitral regurgitation was assessed semiquantitatively (iemild moderate or severe) by color flow Doppler [15] Thesystolic pulmonary artery pressure (SPAP) was calculatedas the sum of the maximum systolic tricuspid regurgitationpressure gradient and the right atrial pressure The rightatrial pressure was estimated on the basis of the diameterand inspiratory collapse index of the inferior vena cava[16] Tricuspid annular plane systolic excursion (TAPSE) wasmeasured by M-mode echocardiography in the apical 4-chamber view as the longitudinal systolic excursion of thetricuspid annulus [17] Right ventricular dysfunction wasdefined as TAPSE lt 17mm [17] LV global longitudinal strain(GLS) analysis was performed offline using commerciallyavailable software (QLAB version 103 Philips UltrasoundSeattle USA) averaging the peak longitudinal strain of the3 apical views GLS data are expressed as absolute values

23 Transthoracic Lung Ultrasound After the transtho-racic echocardiography examination all patients underwenttransthoracic lung ultrasonography with the same phased-array transducer Seven zones were considered in our sim-plified ULC scoring method The anterior chest wall wasdelineated from the sternum to the anterior axillary line andwas subdivided into upper and lower halves from the clavicleto the diaphragm The lateral zone was delineated from theanterior to the posterior axillary line and was subdividedinto upper and lower halves (the area above the fourthintercostal space was defined as the upper area) We initiallyadopted an 8-zone protocol but inclusion of the anteriorlower area on the left side was subsequently removed because

most of the study population had an enlarged heart whichintervened with the area Therefore the 7-zone protocol wasadopted

The elementary findings that were evaluated were theULCs (also known as B-lines) defined as hyperechogenicvertical comet tail artifacts with a narrow base spreadingfrom the pleural line to the further border of the screen[9]

According to the increasing order of severity of interstitialor alveoli involvement images were classified as zero septalsyndrome interstitial-alveolar syndrome or white lung [18](Figure 2) Zero was defined as the absence of B-lines Septalsyndrome was defined as B-lines at regular distances corre-sponding to pleural projection of the subpleural septa (equalto about 7mm) In interstitial-alveolar syndrome B-linesbecomemore confluent separated bylt7mmWhite lungwasdesignated for B-lines that coalesced resulting in an almostcompletely white echographic lung field (confluent B-lines gt80 Figure 2) (Videos 1ndash4 in Supplementary Materials) Asimplified 7-zone ULC score was then calculated accordingto the grades 0 = zero 1 = septal syndrome 2 = interstitial-alveolar syndrome and 3 = white lung Each intercostalspace was examined thoroughly and the images recordedin each zone were those with the highest score All clipsof transthoracic lung ultrasonography were recorded andreviewed blinded to echocardiographic data

The stored images of each patient were scored one monthafter the baseline assessment by the same investigator (HL)who performed the corresponding transthoracic ultrasonog-raphy examination Interobserver reliability was assessed byanother observer (WZ) by dynamic clips in a set of 20 casesEach investigator was blinded to the previous results

We also scanned the pleural line with a high-resolutionlinear probe to exclude pneumogenic ULCs (Videos 5ndash7 inSupplementary Materials)

4 BioMed Research International

Table 2 Patientsrsquo clinical characteristics

Variables Mean plusmn SD or number ()Subjects 119899 93Age y 67 plusmn 14

Gender femalemale 3261Body surface area m2 18 plusmn 02

Hypercholesterolemia 40 (43)Diabetes 20 (22)Previous MIa 16 (17)PCIb 10 (11)CABGc 4 (4)NT-proBNP pgml 11645 plusmn 11070

Radiologic score 13 plusmn 7

NYHA functional classII 28 (30)III 56 (60)IV 9 (10)

Cause of heart failureCoronary artery disease 64 (69)Hypertension 14 (15)Dilated cardiomyopathy 7 (8)Myocarditis 4 (4)Perinatal cardiomyopathy 1 (1)Autoimmunity cardiomyopathy 1 (1)Alcoholic cardiomyopathy 2 (2)

aMI myocardial infarction bPCI percutaneous coronary interventioncCABG coronary artery bypass grafting

24 Statistical Analysis Standard descriptive results areexpressed as mean and standard deviation and categori-cal data are expressed as percentage Correlations betweenvariables were assessed by Spearmanrsquos 2-tailed methodIndependent correlates of ULC score were identified bymultiple linear regression analyses after logarithmic transfor-mation of NT-proBNP The coefficient of differences among3 groups was compared using one-way analysis of varianceTo assess intraobserver and interobserver reliability theULC scores were calculated by a weighted kappa statisticThe diagnostic utility of transthoracic echocardiography indetecting moderate or severe pulmonary congestion symp-toms was determined using receiver-operating character-istic (ROC) curves The best threshold was obtained byselecting the point on the ROC curve that maximized bothsensitivity and specificity Comparisons of the values oftwo groups were performed using the independent-samplesStudentrsquos 119905-test and MannndashWhitney nonparametric testA 119875 value lt 005 was considered statistically significantThe statistical analyses were performed using SPSS soft-ware (version 200 IBM Chicago IL USA) and GraphPad50

3 Results

During the study period 93 heart failure patients withdyspnea were enrolled (Table 2)

Table 3 Patientsrsquo echocardiographic characteristics

Variables Mean plusmn SDLV ejection fraction 357 plusmn 78

LV end-diastolic diametermm 601 plusmn 74

LV end-systolic diametermm 474 plusmn 86

LVEDV mLLVEDVindexmLm2 1663 plusmn 475934 plusmn 255

LVESV mLLVESVindexmLm2 1083 plusmn 387609 plusmn 214

LAV mLLAVindex mLm2 781 plusmn 21244 plusmn 12SPAP mmHg 422 plusmn 103

TAPSE mm 177 plusmn 43

GLS 92 plusmn 27

LV Left ventricular LVEDV left ventricle end-diastolic volume LVESV leftventricle end-systolic diameter LAV left atrial volume index divided byBSA (body surface area) SPAP systolic pulmonary artery pressure TAPSEtricuspid annular plane systolic excursion GLS global longitudinal strain

31 Transthoracic Lung Ultrasound Assessments of ULCswere performed in all patients (feasibility = 100) Bilateraldiffuse B-lines were identified in all patients by lung ultra-sound [5] The duration of the 7-zone lung ultrasoundexamination was 27 plusmn 05 minutes The median ULC scorewas 9 (range 2ndash20)

The kappa values for the intra- and interobserver relia-bilities of the simplified ULC score assessment were 092 and090 respectively

32 Comparisons between NYHA Functional Class RadiologicScore NT-proBNP and ULC Score Significant linear corre-lations were found between the simplified ULC score andradiologic score (119903 = 060 119875 lt 00001) NT-proBNP values(119903 = 050 119875 lt 00001) and NYHA functional class (119903 = 044119875 lt 00001)

33 Comparison between Echocardiographic Parameters andULC Score The mean LVEF was 36 (range 20ndash55Table 2) Among the 93 patients 8 (9) 44 (47) and 41(44) conformed to diastolic dysfunction grades I II andIII respectively [19] Right ventricular dysfunctionwas foundin 39 (42) patients Mitral regurgitation was recognizedas mild in 46 (50) moderate in 33 (35) and severe in14 (15) patients Tricuspid regurgitation was present in 81patients GLS analysis was feasible in 87 of the 93 patients(94) who had optimal image quality for analysis Theechocardiographic characteristics of patients are reported inTable 3

There was a significant correlation between the simplifiedULC score and each of the following average Ee1015840 ratio (119903 =060 119875 lt 00001) SPAP (119903 = 057 119875 lt 00001 Figures3(a)-3(b)) and severity of mitral regurgitation (119903 = 047119875 lt 00001) The correlation between simplified ULC scoreand GLS was also significant (119903 = minus029 119875 lt 001) butrather weak No significant correlation was found betweenULC score and any of the following parameters LVEF

BioMed Research International 5

r = 060P lt 00001

0

5

10

15

20

25U

LC sc

ore

10 20 30 40 500Ee

(a)

r = 057P lt 00001

0

5

10

15

20

25

ULC

scor

e

20 40 60 800SPAP

(b)

Figure 3 Correlation between ULC score and Ee1015840 (a) and systolic pulmonary artery pressure (SPAP) (b)

end-diastolic or end-systolic diameter volume or volumeindex There was also no significant correlation betweenULC score and left atrial volume left atrial volume index orTAPSE

Figure 4 shows a patient with obvious pulmonary edemaon lung ultrasound in whom Ee1015840 was significantly reducedSPAP was increased and LV GLS was mildly reduced butLVEF was normal

The multivariate analysis showed that the only variablesindependently associated with ULC score were Ee1015840 (beta =024 119875 lt 001) SPAP (beta = 032 119875 lt 001) and theradiologic score (beta = 042 119875 lt 00001)

The patients were stratified into 3 groups by LVEF ge 40(119899 = 28 30) 25ndash39 (119899 = 58 62) or lt25 (119899 = 7 8)Among these groups theULC scores were statistically similar(119875 gt 05)However theULC scores of the 3 grades of diastolicdysfunction (grades I II and III) were significantly different(119875 lt 001 Table 4)

The optimal cutoff value of ULC score according tothe ROC curve was 8 for predicting NYHA ge 3 (iemoderate or severe congestion symptoms according toNYHA [11]) The area under the curve was 085 (119875 lt00001) and the sensitivity and specificity was 786 and769 respectively Compared with patients with ULC lt8 those with ULC ge 8 were older with higher Ee1015840 SPAPradiologic score NT-proBNP LV diastolic function andNYHA functional class and more severe mitral regurgitation(Table 5)

An ROC curve was plotted to predict further formoderate-to-severe pulmonary congestion symptoms (ieULC score ge 8 Figure 5) The optimal cutoff value for Ee1015840according to the ROC curve was 184 (area under the curve082 119875 lt 00001 Figure 5(a)) The sensitivity and specificitywere 732 and 784 respectively The optimal cutoffvalue for SPAP based on the ROC curve was 425mmHgfor predicting moderate-to-severe pulmonary congestion(area under the curve 079 119875 lt 00001 Figure 5(b))with a sensitivity and specificity of 628 and 828respectively

Table 4 ULC scores by diastolic function grade and left ventricleejection fraction (LVEF)

ULC score 119875

LV diastolic function gradeGrade I 56 plusmn 24

lt0001Grade II 84 plusmn 32

Grade III 103 plusmn 37

LV ejection fraction (LVEF)LVEF ge 40 85 plusmn 30

052LVEF 25ndash39 91 plusmn 38

LFEF lt 25 101 plusmn 44

4 Discussion

The primary objective of the present study was to evaluate asimplified protocol for scoring ULCs to estimate pulmonarycongestion in heart failure patients This 7-zone protocolwas compared with the echocardiographic results NYHAfunctional classification radiologic score and NT-proBNPWe found that the 7-zone ULC score correlated with LVdiastolic functional parameters SPAP GLS the severity ofmitral regurgitation NYHA functional classification radio-logic score and NT-proBNP Multivariate analysis identifiedthe Ee1015840 systolic pulmonary artery pressure and radiologicscore as the only independent variables associated withULC score increase among echocardiographic results It isimportant to note that the patients enrolled in the study werepatients with heart failure and varying degrees of dyspnea

ULCs are multiple comet tails and a simple echographicsign of EVLW [9] Normal extravascular lung water islt500mL or lt10mLkg [20 21] while excessive EVLW leadsto interstitial and alveolar edema that is lung congestion[22] The distribution of ULCs is a reflection of the vol-ume of pulmonary congestion [18] When decompensatedcongestive heart failure occurs increased LV end-diastolicpressure and left atrial pressure lead to elevated pulmonaryvenous pressure and then increased hydrostatic pressurein the pulmonary capillaries Mild elevation of left atrialpressure (18ndash25mmHg) causes edema in the interstitial

6 BioMed Research International

(a) (b) (c)

(d) (e) (f)

Figure 4 Lung ultrasound and echocardiographic parameters of a patient with congestive heart failure (a) Interstitial-alveolar syndromewas detected by lung ultrasound (b) Mitral inflow showed 119864119860 gt 2 (c) Tissue Doppler early (e1015840) and late (a1015840) diastolic velocities weremarkedly reduced (d) Peak TR velocity by CW Doppler peak right ventricle to right atrial systolic pressure gradient is 47mmHg (e f)Global longitudinal strain analysis was minus161 Note that the left ventricular ejection fraction was 553 (f)

spaces and thickened subpleural interlobular septa Whenleft atrial pressure rises further (gt25mmHg) the lymphaticresorption capacity is exceeded and edema fluid breaksthrough the lung epithelium and pours into the alveoli [23]The density and distribution of ULCs will vary accordingto the pathologic states described above scattered septalsyndrome represents thickened subpleural interlobular septaInterstitial-alveolar syndrome and white lung are a moresevere form of interstitial lung syndrome that is alveolarflooding [18] In the latter stage gas exchange is impaired anddyspnea becomes more serious Therefore the correlationbetween ULC score and severity of dyspnea in accordancewith the NYHA functional classification makes rationalsense

Serum NT-proBNP is a biomarker that is widely used fordiagnosing heart failure although the sensitivity and speci-ficity are not perfect The use of NT-proBNP coupled withlung ultrasound could significantly improve the diagnosticaccuracy in determining heart failure [24] The presence ofB-profile or the number of B-lines in lung ultrasound thatcorrelate with higher BNPs levels was previously studied [2526] We used a 7-zone scanning protocol that was comparedwith NT-proBNP and also detected the positive correlationbetween them

A recent study used a simplified 4-sector method andreported good correlation with EVLW values derived fromtranspulmonary thermodilution [27] Restricting the exami-nation to the anterior pulmonary surface is likely sufficient forcritically ill patients in intensive care However in our opin-ion applying transthoracic ultrasonography in cardiologicalsettings requires the anterolateral pulmonary surface as wellIt was also suggested by Liteplo et al [28] that an 8-zoneprotocol is strongly predictive for patients with congestiveheart failure and another simplified method with an 8-zone protocol using ultrasound was reportedly predictive ofcardiogenic lung edema [29] In the present study we adopteda 7-zone protocol because we found that most of the patientsenrolled in our study had enlarged heart which interferedwith the view of the anterior lower area on the left side

For patients with suspected heart failure echocardiogra-phy is an essential imaging tool that can be used to measureLV systolic and diastolic function estimate pulmonary cap-illary wedge pressure (PCWP) and SPAP and evaluate LVfilling pressure [30] Existing literature is sparse regardingthe association between lung congestion on lung ultrasoundand cardiac function and structure and in most studies thepopulation was not restricted or GLS assessment of LV wasnot included A previous study found a correlation between

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

BioMed Research International 3

Alveolus

Normal Flood-filled interstitium

Alveolar flooding

Capillary

Interstitium

(a) (b)

(c) (d)

Increased hydrostatic pressure

Figure 2 Increasing severity of interstitial or alveoli involvement (a) Normal lung B-lines are absent (b) Septal syndrome B-lines areabout 7mm apart corresponding to subpleural septa (c) Interstitial-alveolar syndrome B-lines are confluent (d) White lung B-lines havecoalesced resulting in an echographic lung field that is almost completely white

E to A (119864119860) were measured from the apical 4-chamberview A 15mm sample volume was placed at the septaland lateral corner of the mitral annulus An analysis wasalso performed for early (e1015840) and late diastolic velocityand the average Ee1015840 ratio was calculated The severity ofmitral regurgitation was assessed semiquantitatively (iemild moderate or severe) by color flow Doppler [15] Thesystolic pulmonary artery pressure (SPAP) was calculatedas the sum of the maximum systolic tricuspid regurgitationpressure gradient and the right atrial pressure The rightatrial pressure was estimated on the basis of the diameterand inspiratory collapse index of the inferior vena cava[16] Tricuspid annular plane systolic excursion (TAPSE) wasmeasured by M-mode echocardiography in the apical 4-chamber view as the longitudinal systolic excursion of thetricuspid annulus [17] Right ventricular dysfunction wasdefined as TAPSE lt 17mm [17] LV global longitudinal strain(GLS) analysis was performed offline using commerciallyavailable software (QLAB version 103 Philips UltrasoundSeattle USA) averaging the peak longitudinal strain of the3 apical views GLS data are expressed as absolute values

23 Transthoracic Lung Ultrasound After the transtho-racic echocardiography examination all patients underwenttransthoracic lung ultrasonography with the same phased-array transducer Seven zones were considered in our sim-plified ULC scoring method The anterior chest wall wasdelineated from the sternum to the anterior axillary line andwas subdivided into upper and lower halves from the clavicleto the diaphragm The lateral zone was delineated from theanterior to the posterior axillary line and was subdividedinto upper and lower halves (the area above the fourthintercostal space was defined as the upper area) We initiallyadopted an 8-zone protocol but inclusion of the anteriorlower area on the left side was subsequently removed because

most of the study population had an enlarged heart whichintervened with the area Therefore the 7-zone protocol wasadopted

The elementary findings that were evaluated were theULCs (also known as B-lines) defined as hyperechogenicvertical comet tail artifacts with a narrow base spreadingfrom the pleural line to the further border of the screen[9]

According to the increasing order of severity of interstitialor alveoli involvement images were classified as zero septalsyndrome interstitial-alveolar syndrome or white lung [18](Figure 2) Zero was defined as the absence of B-lines Septalsyndrome was defined as B-lines at regular distances corre-sponding to pleural projection of the subpleural septa (equalto about 7mm) In interstitial-alveolar syndrome B-linesbecomemore confluent separated bylt7mmWhite lungwasdesignated for B-lines that coalesced resulting in an almostcompletely white echographic lung field (confluent B-lines gt80 Figure 2) (Videos 1ndash4 in Supplementary Materials) Asimplified 7-zone ULC score was then calculated accordingto the grades 0 = zero 1 = septal syndrome 2 = interstitial-alveolar syndrome and 3 = white lung Each intercostalspace was examined thoroughly and the images recordedin each zone were those with the highest score All clipsof transthoracic lung ultrasonography were recorded andreviewed blinded to echocardiographic data

The stored images of each patient were scored one monthafter the baseline assessment by the same investigator (HL)who performed the corresponding transthoracic ultrasonog-raphy examination Interobserver reliability was assessed byanother observer (WZ) by dynamic clips in a set of 20 casesEach investigator was blinded to the previous results

We also scanned the pleural line with a high-resolutionlinear probe to exclude pneumogenic ULCs (Videos 5ndash7 inSupplementary Materials)

4 BioMed Research International

Table 2 Patientsrsquo clinical characteristics

Variables Mean plusmn SD or number ()Subjects 119899 93Age y 67 plusmn 14

Gender femalemale 3261Body surface area m2 18 plusmn 02

Hypercholesterolemia 40 (43)Diabetes 20 (22)Previous MIa 16 (17)PCIb 10 (11)CABGc 4 (4)NT-proBNP pgml 11645 plusmn 11070

Radiologic score 13 plusmn 7

NYHA functional classII 28 (30)III 56 (60)IV 9 (10)

Cause of heart failureCoronary artery disease 64 (69)Hypertension 14 (15)Dilated cardiomyopathy 7 (8)Myocarditis 4 (4)Perinatal cardiomyopathy 1 (1)Autoimmunity cardiomyopathy 1 (1)Alcoholic cardiomyopathy 2 (2)

aMI myocardial infarction bPCI percutaneous coronary interventioncCABG coronary artery bypass grafting

24 Statistical Analysis Standard descriptive results areexpressed as mean and standard deviation and categori-cal data are expressed as percentage Correlations betweenvariables were assessed by Spearmanrsquos 2-tailed methodIndependent correlates of ULC score were identified bymultiple linear regression analyses after logarithmic transfor-mation of NT-proBNP The coefficient of differences among3 groups was compared using one-way analysis of varianceTo assess intraobserver and interobserver reliability theULC scores were calculated by a weighted kappa statisticThe diagnostic utility of transthoracic echocardiography indetecting moderate or severe pulmonary congestion symp-toms was determined using receiver-operating character-istic (ROC) curves The best threshold was obtained byselecting the point on the ROC curve that maximized bothsensitivity and specificity Comparisons of the values oftwo groups were performed using the independent-samplesStudentrsquos 119905-test and MannndashWhitney nonparametric testA 119875 value lt 005 was considered statistically significantThe statistical analyses were performed using SPSS soft-ware (version 200 IBM Chicago IL USA) and GraphPad50

3 Results

During the study period 93 heart failure patients withdyspnea were enrolled (Table 2)

Table 3 Patientsrsquo echocardiographic characteristics

Variables Mean plusmn SDLV ejection fraction 357 plusmn 78

LV end-diastolic diametermm 601 plusmn 74

LV end-systolic diametermm 474 plusmn 86

LVEDV mLLVEDVindexmLm2 1663 plusmn 475934 plusmn 255

LVESV mLLVESVindexmLm2 1083 plusmn 387609 plusmn 214

LAV mLLAVindex mLm2 781 plusmn 21244 plusmn 12SPAP mmHg 422 plusmn 103

TAPSE mm 177 plusmn 43

GLS 92 plusmn 27

LV Left ventricular LVEDV left ventricle end-diastolic volume LVESV leftventricle end-systolic diameter LAV left atrial volume index divided byBSA (body surface area) SPAP systolic pulmonary artery pressure TAPSEtricuspid annular plane systolic excursion GLS global longitudinal strain

31 Transthoracic Lung Ultrasound Assessments of ULCswere performed in all patients (feasibility = 100) Bilateraldiffuse B-lines were identified in all patients by lung ultra-sound [5] The duration of the 7-zone lung ultrasoundexamination was 27 plusmn 05 minutes The median ULC scorewas 9 (range 2ndash20)

The kappa values for the intra- and interobserver relia-bilities of the simplified ULC score assessment were 092 and090 respectively

32 Comparisons between NYHA Functional Class RadiologicScore NT-proBNP and ULC Score Significant linear corre-lations were found between the simplified ULC score andradiologic score (119903 = 060 119875 lt 00001) NT-proBNP values(119903 = 050 119875 lt 00001) and NYHA functional class (119903 = 044119875 lt 00001)

33 Comparison between Echocardiographic Parameters andULC Score The mean LVEF was 36 (range 20ndash55Table 2) Among the 93 patients 8 (9) 44 (47) and 41(44) conformed to diastolic dysfunction grades I II andIII respectively [19] Right ventricular dysfunctionwas foundin 39 (42) patients Mitral regurgitation was recognizedas mild in 46 (50) moderate in 33 (35) and severe in14 (15) patients Tricuspid regurgitation was present in 81patients GLS analysis was feasible in 87 of the 93 patients(94) who had optimal image quality for analysis Theechocardiographic characteristics of patients are reported inTable 3

There was a significant correlation between the simplifiedULC score and each of the following average Ee1015840 ratio (119903 =060 119875 lt 00001) SPAP (119903 = 057 119875 lt 00001 Figures3(a)-3(b)) and severity of mitral regurgitation (119903 = 047119875 lt 00001) The correlation between simplified ULC scoreand GLS was also significant (119903 = minus029 119875 lt 001) butrather weak No significant correlation was found betweenULC score and any of the following parameters LVEF

BioMed Research International 5

r = 060P lt 00001

0

5

10

15

20

25U

LC sc

ore

10 20 30 40 500Ee

(a)

r = 057P lt 00001

0

5

10

15

20

25

ULC

scor

e

20 40 60 800SPAP

(b)

Figure 3 Correlation between ULC score and Ee1015840 (a) and systolic pulmonary artery pressure (SPAP) (b)

end-diastolic or end-systolic diameter volume or volumeindex There was also no significant correlation betweenULC score and left atrial volume left atrial volume index orTAPSE

Figure 4 shows a patient with obvious pulmonary edemaon lung ultrasound in whom Ee1015840 was significantly reducedSPAP was increased and LV GLS was mildly reduced butLVEF was normal

The multivariate analysis showed that the only variablesindependently associated with ULC score were Ee1015840 (beta =024 119875 lt 001) SPAP (beta = 032 119875 lt 001) and theradiologic score (beta = 042 119875 lt 00001)

The patients were stratified into 3 groups by LVEF ge 40(119899 = 28 30) 25ndash39 (119899 = 58 62) or lt25 (119899 = 7 8)Among these groups theULC scores were statistically similar(119875 gt 05)However theULC scores of the 3 grades of diastolicdysfunction (grades I II and III) were significantly different(119875 lt 001 Table 4)

The optimal cutoff value of ULC score according tothe ROC curve was 8 for predicting NYHA ge 3 (iemoderate or severe congestion symptoms according toNYHA [11]) The area under the curve was 085 (119875 lt00001) and the sensitivity and specificity was 786 and769 respectively Compared with patients with ULC lt8 those with ULC ge 8 were older with higher Ee1015840 SPAPradiologic score NT-proBNP LV diastolic function andNYHA functional class and more severe mitral regurgitation(Table 5)

An ROC curve was plotted to predict further formoderate-to-severe pulmonary congestion symptoms (ieULC score ge 8 Figure 5) The optimal cutoff value for Ee1015840according to the ROC curve was 184 (area under the curve082 119875 lt 00001 Figure 5(a)) The sensitivity and specificitywere 732 and 784 respectively The optimal cutoffvalue for SPAP based on the ROC curve was 425mmHgfor predicting moderate-to-severe pulmonary congestion(area under the curve 079 119875 lt 00001 Figure 5(b))with a sensitivity and specificity of 628 and 828respectively

Table 4 ULC scores by diastolic function grade and left ventricleejection fraction (LVEF)

ULC score 119875

LV diastolic function gradeGrade I 56 plusmn 24

lt0001Grade II 84 plusmn 32

Grade III 103 plusmn 37

LV ejection fraction (LVEF)LVEF ge 40 85 plusmn 30

052LVEF 25ndash39 91 plusmn 38

LFEF lt 25 101 plusmn 44

4 Discussion

The primary objective of the present study was to evaluate asimplified protocol for scoring ULCs to estimate pulmonarycongestion in heart failure patients This 7-zone protocolwas compared with the echocardiographic results NYHAfunctional classification radiologic score and NT-proBNPWe found that the 7-zone ULC score correlated with LVdiastolic functional parameters SPAP GLS the severity ofmitral regurgitation NYHA functional classification radio-logic score and NT-proBNP Multivariate analysis identifiedthe Ee1015840 systolic pulmonary artery pressure and radiologicscore as the only independent variables associated withULC score increase among echocardiographic results It isimportant to note that the patients enrolled in the study werepatients with heart failure and varying degrees of dyspnea

ULCs are multiple comet tails and a simple echographicsign of EVLW [9] Normal extravascular lung water islt500mL or lt10mLkg [20 21] while excessive EVLW leadsto interstitial and alveolar edema that is lung congestion[22] The distribution of ULCs is a reflection of the vol-ume of pulmonary congestion [18] When decompensatedcongestive heart failure occurs increased LV end-diastolicpressure and left atrial pressure lead to elevated pulmonaryvenous pressure and then increased hydrostatic pressurein the pulmonary capillaries Mild elevation of left atrialpressure (18ndash25mmHg) causes edema in the interstitial

6 BioMed Research International

(a) (b) (c)

(d) (e) (f)

Figure 4 Lung ultrasound and echocardiographic parameters of a patient with congestive heart failure (a) Interstitial-alveolar syndromewas detected by lung ultrasound (b) Mitral inflow showed 119864119860 gt 2 (c) Tissue Doppler early (e1015840) and late (a1015840) diastolic velocities weremarkedly reduced (d) Peak TR velocity by CW Doppler peak right ventricle to right atrial systolic pressure gradient is 47mmHg (e f)Global longitudinal strain analysis was minus161 Note that the left ventricular ejection fraction was 553 (f)

spaces and thickened subpleural interlobular septa Whenleft atrial pressure rises further (gt25mmHg) the lymphaticresorption capacity is exceeded and edema fluid breaksthrough the lung epithelium and pours into the alveoli [23]The density and distribution of ULCs will vary accordingto the pathologic states described above scattered septalsyndrome represents thickened subpleural interlobular septaInterstitial-alveolar syndrome and white lung are a moresevere form of interstitial lung syndrome that is alveolarflooding [18] In the latter stage gas exchange is impaired anddyspnea becomes more serious Therefore the correlationbetween ULC score and severity of dyspnea in accordancewith the NYHA functional classification makes rationalsense

Serum NT-proBNP is a biomarker that is widely used fordiagnosing heart failure although the sensitivity and speci-ficity are not perfect The use of NT-proBNP coupled withlung ultrasound could significantly improve the diagnosticaccuracy in determining heart failure [24] The presence ofB-profile or the number of B-lines in lung ultrasound thatcorrelate with higher BNPs levels was previously studied [2526] We used a 7-zone scanning protocol that was comparedwith NT-proBNP and also detected the positive correlationbetween them

A recent study used a simplified 4-sector method andreported good correlation with EVLW values derived fromtranspulmonary thermodilution [27] Restricting the exami-nation to the anterior pulmonary surface is likely sufficient forcritically ill patients in intensive care However in our opin-ion applying transthoracic ultrasonography in cardiologicalsettings requires the anterolateral pulmonary surface as wellIt was also suggested by Liteplo et al [28] that an 8-zoneprotocol is strongly predictive for patients with congestiveheart failure and another simplified method with an 8-zone protocol using ultrasound was reportedly predictive ofcardiogenic lung edema [29] In the present study we adopteda 7-zone protocol because we found that most of the patientsenrolled in our study had enlarged heart which interferedwith the view of the anterior lower area on the left side

For patients with suspected heart failure echocardiogra-phy is an essential imaging tool that can be used to measureLV systolic and diastolic function estimate pulmonary cap-illary wedge pressure (PCWP) and SPAP and evaluate LVfilling pressure [30] Existing literature is sparse regardingthe association between lung congestion on lung ultrasoundand cardiac function and structure and in most studies thepopulation was not restricted or GLS assessment of LV wasnot included A previous study found a correlation between

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

4 BioMed Research International

Table 2 Patientsrsquo clinical characteristics

Variables Mean plusmn SD or number ()Subjects 119899 93Age y 67 plusmn 14

Gender femalemale 3261Body surface area m2 18 plusmn 02

Hypercholesterolemia 40 (43)Diabetes 20 (22)Previous MIa 16 (17)PCIb 10 (11)CABGc 4 (4)NT-proBNP pgml 11645 plusmn 11070

Radiologic score 13 plusmn 7

NYHA functional classII 28 (30)III 56 (60)IV 9 (10)

Cause of heart failureCoronary artery disease 64 (69)Hypertension 14 (15)Dilated cardiomyopathy 7 (8)Myocarditis 4 (4)Perinatal cardiomyopathy 1 (1)Autoimmunity cardiomyopathy 1 (1)Alcoholic cardiomyopathy 2 (2)

aMI myocardial infarction bPCI percutaneous coronary interventioncCABG coronary artery bypass grafting

24 Statistical Analysis Standard descriptive results areexpressed as mean and standard deviation and categori-cal data are expressed as percentage Correlations betweenvariables were assessed by Spearmanrsquos 2-tailed methodIndependent correlates of ULC score were identified bymultiple linear regression analyses after logarithmic transfor-mation of NT-proBNP The coefficient of differences among3 groups was compared using one-way analysis of varianceTo assess intraobserver and interobserver reliability theULC scores were calculated by a weighted kappa statisticThe diagnostic utility of transthoracic echocardiography indetecting moderate or severe pulmonary congestion symp-toms was determined using receiver-operating character-istic (ROC) curves The best threshold was obtained byselecting the point on the ROC curve that maximized bothsensitivity and specificity Comparisons of the values oftwo groups were performed using the independent-samplesStudentrsquos 119905-test and MannndashWhitney nonparametric testA 119875 value lt 005 was considered statistically significantThe statistical analyses were performed using SPSS soft-ware (version 200 IBM Chicago IL USA) and GraphPad50

3 Results

During the study period 93 heart failure patients withdyspnea were enrolled (Table 2)

Table 3 Patientsrsquo echocardiographic characteristics

Variables Mean plusmn SDLV ejection fraction 357 plusmn 78

LV end-diastolic diametermm 601 plusmn 74

LV end-systolic diametermm 474 plusmn 86

LVEDV mLLVEDVindexmLm2 1663 plusmn 475934 plusmn 255

LVESV mLLVESVindexmLm2 1083 plusmn 387609 plusmn 214

LAV mLLAVindex mLm2 781 plusmn 21244 plusmn 12SPAP mmHg 422 plusmn 103

TAPSE mm 177 plusmn 43

GLS 92 plusmn 27

LV Left ventricular LVEDV left ventricle end-diastolic volume LVESV leftventricle end-systolic diameter LAV left atrial volume index divided byBSA (body surface area) SPAP systolic pulmonary artery pressure TAPSEtricuspid annular plane systolic excursion GLS global longitudinal strain

31 Transthoracic Lung Ultrasound Assessments of ULCswere performed in all patients (feasibility = 100) Bilateraldiffuse B-lines were identified in all patients by lung ultra-sound [5] The duration of the 7-zone lung ultrasoundexamination was 27 plusmn 05 minutes The median ULC scorewas 9 (range 2ndash20)

The kappa values for the intra- and interobserver relia-bilities of the simplified ULC score assessment were 092 and090 respectively

32 Comparisons between NYHA Functional Class RadiologicScore NT-proBNP and ULC Score Significant linear corre-lations were found between the simplified ULC score andradiologic score (119903 = 060 119875 lt 00001) NT-proBNP values(119903 = 050 119875 lt 00001) and NYHA functional class (119903 = 044119875 lt 00001)

33 Comparison between Echocardiographic Parameters andULC Score The mean LVEF was 36 (range 20ndash55Table 2) Among the 93 patients 8 (9) 44 (47) and 41(44) conformed to diastolic dysfunction grades I II andIII respectively [19] Right ventricular dysfunctionwas foundin 39 (42) patients Mitral regurgitation was recognizedas mild in 46 (50) moderate in 33 (35) and severe in14 (15) patients Tricuspid regurgitation was present in 81patients GLS analysis was feasible in 87 of the 93 patients(94) who had optimal image quality for analysis Theechocardiographic characteristics of patients are reported inTable 3

There was a significant correlation between the simplifiedULC score and each of the following average Ee1015840 ratio (119903 =060 119875 lt 00001) SPAP (119903 = 057 119875 lt 00001 Figures3(a)-3(b)) and severity of mitral regurgitation (119903 = 047119875 lt 00001) The correlation between simplified ULC scoreand GLS was also significant (119903 = minus029 119875 lt 001) butrather weak No significant correlation was found betweenULC score and any of the following parameters LVEF

BioMed Research International 5

r = 060P lt 00001

0

5

10

15

20

25U

LC sc

ore

10 20 30 40 500Ee

(a)

r = 057P lt 00001

0

5

10

15

20

25

ULC

scor

e

20 40 60 800SPAP

(b)

Figure 3 Correlation between ULC score and Ee1015840 (a) and systolic pulmonary artery pressure (SPAP) (b)

end-diastolic or end-systolic diameter volume or volumeindex There was also no significant correlation betweenULC score and left atrial volume left atrial volume index orTAPSE

Figure 4 shows a patient with obvious pulmonary edemaon lung ultrasound in whom Ee1015840 was significantly reducedSPAP was increased and LV GLS was mildly reduced butLVEF was normal

The multivariate analysis showed that the only variablesindependently associated with ULC score were Ee1015840 (beta =024 119875 lt 001) SPAP (beta = 032 119875 lt 001) and theradiologic score (beta = 042 119875 lt 00001)

The patients were stratified into 3 groups by LVEF ge 40(119899 = 28 30) 25ndash39 (119899 = 58 62) or lt25 (119899 = 7 8)Among these groups theULC scores were statistically similar(119875 gt 05)However theULC scores of the 3 grades of diastolicdysfunction (grades I II and III) were significantly different(119875 lt 001 Table 4)

The optimal cutoff value of ULC score according tothe ROC curve was 8 for predicting NYHA ge 3 (iemoderate or severe congestion symptoms according toNYHA [11]) The area under the curve was 085 (119875 lt00001) and the sensitivity and specificity was 786 and769 respectively Compared with patients with ULC lt8 those with ULC ge 8 were older with higher Ee1015840 SPAPradiologic score NT-proBNP LV diastolic function andNYHA functional class and more severe mitral regurgitation(Table 5)

An ROC curve was plotted to predict further formoderate-to-severe pulmonary congestion symptoms (ieULC score ge 8 Figure 5) The optimal cutoff value for Ee1015840according to the ROC curve was 184 (area under the curve082 119875 lt 00001 Figure 5(a)) The sensitivity and specificitywere 732 and 784 respectively The optimal cutoffvalue for SPAP based on the ROC curve was 425mmHgfor predicting moderate-to-severe pulmonary congestion(area under the curve 079 119875 lt 00001 Figure 5(b))with a sensitivity and specificity of 628 and 828respectively

Table 4 ULC scores by diastolic function grade and left ventricleejection fraction (LVEF)

ULC score 119875

LV diastolic function gradeGrade I 56 plusmn 24

lt0001Grade II 84 plusmn 32

Grade III 103 plusmn 37

LV ejection fraction (LVEF)LVEF ge 40 85 plusmn 30

052LVEF 25ndash39 91 plusmn 38

LFEF lt 25 101 plusmn 44

4 Discussion

The primary objective of the present study was to evaluate asimplified protocol for scoring ULCs to estimate pulmonarycongestion in heart failure patients This 7-zone protocolwas compared with the echocardiographic results NYHAfunctional classification radiologic score and NT-proBNPWe found that the 7-zone ULC score correlated with LVdiastolic functional parameters SPAP GLS the severity ofmitral regurgitation NYHA functional classification radio-logic score and NT-proBNP Multivariate analysis identifiedthe Ee1015840 systolic pulmonary artery pressure and radiologicscore as the only independent variables associated withULC score increase among echocardiographic results It isimportant to note that the patients enrolled in the study werepatients with heart failure and varying degrees of dyspnea

ULCs are multiple comet tails and a simple echographicsign of EVLW [9] Normal extravascular lung water islt500mL or lt10mLkg [20 21] while excessive EVLW leadsto interstitial and alveolar edema that is lung congestion[22] The distribution of ULCs is a reflection of the vol-ume of pulmonary congestion [18] When decompensatedcongestive heart failure occurs increased LV end-diastolicpressure and left atrial pressure lead to elevated pulmonaryvenous pressure and then increased hydrostatic pressurein the pulmonary capillaries Mild elevation of left atrialpressure (18ndash25mmHg) causes edema in the interstitial

6 BioMed Research International

(a) (b) (c)

(d) (e) (f)

Figure 4 Lung ultrasound and echocardiographic parameters of a patient with congestive heart failure (a) Interstitial-alveolar syndromewas detected by lung ultrasound (b) Mitral inflow showed 119864119860 gt 2 (c) Tissue Doppler early (e1015840) and late (a1015840) diastolic velocities weremarkedly reduced (d) Peak TR velocity by CW Doppler peak right ventricle to right atrial systolic pressure gradient is 47mmHg (e f)Global longitudinal strain analysis was minus161 Note that the left ventricular ejection fraction was 553 (f)

spaces and thickened subpleural interlobular septa Whenleft atrial pressure rises further (gt25mmHg) the lymphaticresorption capacity is exceeded and edema fluid breaksthrough the lung epithelium and pours into the alveoli [23]The density and distribution of ULCs will vary accordingto the pathologic states described above scattered septalsyndrome represents thickened subpleural interlobular septaInterstitial-alveolar syndrome and white lung are a moresevere form of interstitial lung syndrome that is alveolarflooding [18] In the latter stage gas exchange is impaired anddyspnea becomes more serious Therefore the correlationbetween ULC score and severity of dyspnea in accordancewith the NYHA functional classification makes rationalsense

Serum NT-proBNP is a biomarker that is widely used fordiagnosing heart failure although the sensitivity and speci-ficity are not perfect The use of NT-proBNP coupled withlung ultrasound could significantly improve the diagnosticaccuracy in determining heart failure [24] The presence ofB-profile or the number of B-lines in lung ultrasound thatcorrelate with higher BNPs levels was previously studied [2526] We used a 7-zone scanning protocol that was comparedwith NT-proBNP and also detected the positive correlationbetween them

A recent study used a simplified 4-sector method andreported good correlation with EVLW values derived fromtranspulmonary thermodilution [27] Restricting the exami-nation to the anterior pulmonary surface is likely sufficient forcritically ill patients in intensive care However in our opin-ion applying transthoracic ultrasonography in cardiologicalsettings requires the anterolateral pulmonary surface as wellIt was also suggested by Liteplo et al [28] that an 8-zoneprotocol is strongly predictive for patients with congestiveheart failure and another simplified method with an 8-zone protocol using ultrasound was reportedly predictive ofcardiogenic lung edema [29] In the present study we adopteda 7-zone protocol because we found that most of the patientsenrolled in our study had enlarged heart which interferedwith the view of the anterior lower area on the left side

For patients with suspected heart failure echocardiogra-phy is an essential imaging tool that can be used to measureLV systolic and diastolic function estimate pulmonary cap-illary wedge pressure (PCWP) and SPAP and evaluate LVfilling pressure [30] Existing literature is sparse regardingthe association between lung congestion on lung ultrasoundand cardiac function and structure and in most studies thepopulation was not restricted or GLS assessment of LV wasnot included A previous study found a correlation between

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

BioMed Research International 5

r = 060P lt 00001

0

5

10

15

20

25U

LC sc

ore

10 20 30 40 500Ee

(a)

r = 057P lt 00001

0

5

10

15

20

25

ULC

scor

e

20 40 60 800SPAP

(b)

Figure 3 Correlation between ULC score and Ee1015840 (a) and systolic pulmonary artery pressure (SPAP) (b)

end-diastolic or end-systolic diameter volume or volumeindex There was also no significant correlation betweenULC score and left atrial volume left atrial volume index orTAPSE

Figure 4 shows a patient with obvious pulmonary edemaon lung ultrasound in whom Ee1015840 was significantly reducedSPAP was increased and LV GLS was mildly reduced butLVEF was normal

The multivariate analysis showed that the only variablesindependently associated with ULC score were Ee1015840 (beta =024 119875 lt 001) SPAP (beta = 032 119875 lt 001) and theradiologic score (beta = 042 119875 lt 00001)

The patients were stratified into 3 groups by LVEF ge 40(119899 = 28 30) 25ndash39 (119899 = 58 62) or lt25 (119899 = 7 8)Among these groups theULC scores were statistically similar(119875 gt 05)However theULC scores of the 3 grades of diastolicdysfunction (grades I II and III) were significantly different(119875 lt 001 Table 4)

The optimal cutoff value of ULC score according tothe ROC curve was 8 for predicting NYHA ge 3 (iemoderate or severe congestion symptoms according toNYHA [11]) The area under the curve was 085 (119875 lt00001) and the sensitivity and specificity was 786 and769 respectively Compared with patients with ULC lt8 those with ULC ge 8 were older with higher Ee1015840 SPAPradiologic score NT-proBNP LV diastolic function andNYHA functional class and more severe mitral regurgitation(Table 5)

An ROC curve was plotted to predict further formoderate-to-severe pulmonary congestion symptoms (ieULC score ge 8 Figure 5) The optimal cutoff value for Ee1015840according to the ROC curve was 184 (area under the curve082 119875 lt 00001 Figure 5(a)) The sensitivity and specificitywere 732 and 784 respectively The optimal cutoffvalue for SPAP based on the ROC curve was 425mmHgfor predicting moderate-to-severe pulmonary congestion(area under the curve 079 119875 lt 00001 Figure 5(b))with a sensitivity and specificity of 628 and 828respectively

Table 4 ULC scores by diastolic function grade and left ventricleejection fraction (LVEF)

ULC score 119875

LV diastolic function gradeGrade I 56 plusmn 24

lt0001Grade II 84 plusmn 32

Grade III 103 plusmn 37

LV ejection fraction (LVEF)LVEF ge 40 85 plusmn 30

052LVEF 25ndash39 91 plusmn 38

LFEF lt 25 101 plusmn 44

4 Discussion

The primary objective of the present study was to evaluate asimplified protocol for scoring ULCs to estimate pulmonarycongestion in heart failure patients This 7-zone protocolwas compared with the echocardiographic results NYHAfunctional classification radiologic score and NT-proBNPWe found that the 7-zone ULC score correlated with LVdiastolic functional parameters SPAP GLS the severity ofmitral regurgitation NYHA functional classification radio-logic score and NT-proBNP Multivariate analysis identifiedthe Ee1015840 systolic pulmonary artery pressure and radiologicscore as the only independent variables associated withULC score increase among echocardiographic results It isimportant to note that the patients enrolled in the study werepatients with heart failure and varying degrees of dyspnea

ULCs are multiple comet tails and a simple echographicsign of EVLW [9] Normal extravascular lung water islt500mL or lt10mLkg [20 21] while excessive EVLW leadsto interstitial and alveolar edema that is lung congestion[22] The distribution of ULCs is a reflection of the vol-ume of pulmonary congestion [18] When decompensatedcongestive heart failure occurs increased LV end-diastolicpressure and left atrial pressure lead to elevated pulmonaryvenous pressure and then increased hydrostatic pressurein the pulmonary capillaries Mild elevation of left atrialpressure (18ndash25mmHg) causes edema in the interstitial

6 BioMed Research International

(a) (b) (c)

(d) (e) (f)

Figure 4 Lung ultrasound and echocardiographic parameters of a patient with congestive heart failure (a) Interstitial-alveolar syndromewas detected by lung ultrasound (b) Mitral inflow showed 119864119860 gt 2 (c) Tissue Doppler early (e1015840) and late (a1015840) diastolic velocities weremarkedly reduced (d) Peak TR velocity by CW Doppler peak right ventricle to right atrial systolic pressure gradient is 47mmHg (e f)Global longitudinal strain analysis was minus161 Note that the left ventricular ejection fraction was 553 (f)

spaces and thickened subpleural interlobular septa Whenleft atrial pressure rises further (gt25mmHg) the lymphaticresorption capacity is exceeded and edema fluid breaksthrough the lung epithelium and pours into the alveoli [23]The density and distribution of ULCs will vary accordingto the pathologic states described above scattered septalsyndrome represents thickened subpleural interlobular septaInterstitial-alveolar syndrome and white lung are a moresevere form of interstitial lung syndrome that is alveolarflooding [18] In the latter stage gas exchange is impaired anddyspnea becomes more serious Therefore the correlationbetween ULC score and severity of dyspnea in accordancewith the NYHA functional classification makes rationalsense

Serum NT-proBNP is a biomarker that is widely used fordiagnosing heart failure although the sensitivity and speci-ficity are not perfect The use of NT-proBNP coupled withlung ultrasound could significantly improve the diagnosticaccuracy in determining heart failure [24] The presence ofB-profile or the number of B-lines in lung ultrasound thatcorrelate with higher BNPs levels was previously studied [2526] We used a 7-zone scanning protocol that was comparedwith NT-proBNP and also detected the positive correlationbetween them

A recent study used a simplified 4-sector method andreported good correlation with EVLW values derived fromtranspulmonary thermodilution [27] Restricting the exami-nation to the anterior pulmonary surface is likely sufficient forcritically ill patients in intensive care However in our opin-ion applying transthoracic ultrasonography in cardiologicalsettings requires the anterolateral pulmonary surface as wellIt was also suggested by Liteplo et al [28] that an 8-zoneprotocol is strongly predictive for patients with congestiveheart failure and another simplified method with an 8-zone protocol using ultrasound was reportedly predictive ofcardiogenic lung edema [29] In the present study we adopteda 7-zone protocol because we found that most of the patientsenrolled in our study had enlarged heart which interferedwith the view of the anterior lower area on the left side

For patients with suspected heart failure echocardiogra-phy is an essential imaging tool that can be used to measureLV systolic and diastolic function estimate pulmonary cap-illary wedge pressure (PCWP) and SPAP and evaluate LVfilling pressure [30] Existing literature is sparse regardingthe association between lung congestion on lung ultrasoundand cardiac function and structure and in most studies thepopulation was not restricted or GLS assessment of LV wasnot included A previous study found a correlation between

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

6 BioMed Research International

(a) (b) (c)

(d) (e) (f)

Figure 4 Lung ultrasound and echocardiographic parameters of a patient with congestive heart failure (a) Interstitial-alveolar syndromewas detected by lung ultrasound (b) Mitral inflow showed 119864119860 gt 2 (c) Tissue Doppler early (e1015840) and late (a1015840) diastolic velocities weremarkedly reduced (d) Peak TR velocity by CW Doppler peak right ventricle to right atrial systolic pressure gradient is 47mmHg (e f)Global longitudinal strain analysis was minus161 Note that the left ventricular ejection fraction was 553 (f)

spaces and thickened subpleural interlobular septa Whenleft atrial pressure rises further (gt25mmHg) the lymphaticresorption capacity is exceeded and edema fluid breaksthrough the lung epithelium and pours into the alveoli [23]The density and distribution of ULCs will vary accordingto the pathologic states described above scattered septalsyndrome represents thickened subpleural interlobular septaInterstitial-alveolar syndrome and white lung are a moresevere form of interstitial lung syndrome that is alveolarflooding [18] In the latter stage gas exchange is impaired anddyspnea becomes more serious Therefore the correlationbetween ULC score and severity of dyspnea in accordancewith the NYHA functional classification makes rationalsense

Serum NT-proBNP is a biomarker that is widely used fordiagnosing heart failure although the sensitivity and speci-ficity are not perfect The use of NT-proBNP coupled withlung ultrasound could significantly improve the diagnosticaccuracy in determining heart failure [24] The presence ofB-profile or the number of B-lines in lung ultrasound thatcorrelate with higher BNPs levels was previously studied [2526] We used a 7-zone scanning protocol that was comparedwith NT-proBNP and also detected the positive correlationbetween them

A recent study used a simplified 4-sector method andreported good correlation with EVLW values derived fromtranspulmonary thermodilution [27] Restricting the exami-nation to the anterior pulmonary surface is likely sufficient forcritically ill patients in intensive care However in our opin-ion applying transthoracic ultrasonography in cardiologicalsettings requires the anterolateral pulmonary surface as wellIt was also suggested by Liteplo et al [28] that an 8-zoneprotocol is strongly predictive for patients with congestiveheart failure and another simplified method with an 8-zone protocol using ultrasound was reportedly predictive ofcardiogenic lung edema [29] In the present study we adopteda 7-zone protocol because we found that most of the patientsenrolled in our study had enlarged heart which interferedwith the view of the anterior lower area on the left side

For patients with suspected heart failure echocardiogra-phy is an essential imaging tool that can be used to measureLV systolic and diastolic function estimate pulmonary cap-illary wedge pressure (PCWP) and SPAP and evaluate LVfilling pressure [30] Existing literature is sparse regardingthe association between lung congestion on lung ultrasoundand cardiac function and structure and in most studies thepopulation was not restricted or GLS assessment of LV wasnot included A previous study found a correlation between

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

BioMed Research International 7

Table 5 Patients with ULC scores lt 8 and ge8lowast

Variables ULC score119875

lt8 (119899 = 37) ge8 (119899 = 56)Ee1015840 146 plusmn 45 227 plusmn 81 lt00001SPAP mmHg 352 plusmn 86 458 plusmn 93 lt00001GLS 99 plusmn 22 91 plusmn 27 014LV ejection fraction 363 plusmn 76 353 plusmn 80 055LVEDD mm 593 plusmn 76 606 plusmn 73 046LVEDVindex mLm2 916 plusmn 228 946 plusmn 273 057LAVindex mLm2 434 plusmn 134 446 plusmn 113 066TAPSE mm 189 plusmn 46 170 plusmn 40 006LV diastolic function grade lt005

Grade I 7 (19) 1 (2)Grade II 18 (49) 26 (46)Grade III 12 (32) 29 (52)

Mitral regurgitation lt0001Mild 27 (73) 19 (34)Moderate 8 (22) 25 (45)Severe 2 (5) 12 (21)

NYHA functional class lt0001II 20 (54) 11 (20)III 15 (41) 33 (59)IV 2 (5) 12 (21)

Age y 636 plusmn 137 701 plusmn 134 lt005NT-proBNP pgml 50462 plusmn 43253 154263 plusmn 121402 lt00001Radiologic score 82 plusmn 44 157 plusmn 65 lt00001lowastData are mean plusmn SD or number () SPAP systolic pulmonary artery pressure GLS global longitudinal strain LVEDD left ventricle end-systolic diameterLVEDV left ventricle end-diastolic volume LAV left atrial volume index divided by body surface area TAPSE tricuspid annular plane systolic excursionNYHA New York Heart Association

Area under curve 08295 CI 074ndash090

Sensitivity 732 specificity 784

184

0

20

40

60

80

100

Sens

itivi

ty (

)

20 40 60 80 1000100 minus specificity

Ee

(a)

Area under curve 07995 CI 069ndash089

Sensitivity 628 specificity 828

SPAP

0

20

40

60

80

100

Sens

itivi

ty (

)

425 mmHg

20 40 60 80 1000100 minus specificity

(b)

Figure 5 ROC curves showing the diagnostic performance of Ee1015840 and SPAP for predicting ULC score ge 8 (a) ROC curve of the Ee1015840 and(b) ROC curve of the SPAP SPAP systolic pulmonary artery pressure

the number of B-lines and LVEF or B-lines and the degreeof diastolic dysfunction in patients with suspected heartfailure [6] In a study of 72 patients (53 and 19 with LVsystolic dysfunction and normal function resp) Agricolaet al [31] reported a positive linear correlation between

B-lines and Ee1015840 estimated PCWP SPAP and LVEF Anotherreport corroborated this finding in addition to a significantlinear correlation between B-lines and left atrial volume andpulmonary pressure in a cohort of dialysis patients [1] Thesedata are in broad agreement with the present study in which

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

8 BioMed Research International

Ee1015840 a surrogate marker for left-side filling pressures andSPAP due to pulmonary venous congestion had the strongestassociation with ULCs

However unlike the above studies we found that theLVEF was somewhat less informative for predicting thedegree of lung congestion The discrepancy in results couldbe due to different study populations Ours was limitedexclusively to a cohort of identified left heart failure patientswith pulmonary congestion In some contexts it has beenshown that the signs and symptoms of congestive heartfailure correlate poorly with LVEF [32 33] Gandhi et al [34]proposed that in patients with acute hypertensive pulmonaryedema the edema was due to the exacerbation of diastolicdysfunction but not to systolic dysfunction It was alsoreported in a study of patients with cardiogenic pulmonaryedema detected by lung ultrasound that up to 154 had anLVEF gt 50 and it was thought that diastolic dysfunctionmay be the cause of pulmonary edema [29] Actually as aparameter of LV systolic function LVEF is load-dependentlinked to the quality of imaging and LV geometry and maynot correlate with functional status Interestingly althoughGLS weakly correlated with ULC score in the present studythe multivariate analysis showed no independent associa-tion Therefore this finding further validates that systolicperformance may not be better than the diastolic one fordetermining the degree of lung congestion in patients withheart failure

An enlarged left heart is suggestive of chronically elevatedLV filling pressure A normal left heart volume is often notedin patients with acute increase in LV filling pressures orin the earliest stage of diastolic dysfunction However ourstudy population included patients with a history of chronicheart failure and this may be the reason for the inconsis-tency between left heart volume and pulmonary congestiondegree

In the present study we offer a simplified ULC scoringmethod to estimate the degree of congestion that consideredonly 7 thoracic zones This tool is less refined than countingall the B-lines in 28 sectors but provides easy-to-acquire datain an emergency setting and it is much easier to differentiate4 types of ULC patterns than it is to count B-lines

The study may be considered limited by the lack ofpatients with atrial fibrillation and mitral stenosis A goodlyproportion of patients with heart failure experience theseconditions especially thosewith preserved LVEF [35] On theother hand our relatively homogeneous population avoidedthe possibility of outpatients with poor echocardiographicindices or patients receiving intravenous diuretics

Our conclusions are based only on imaging evaluationsand echocardiographic indices While B-lines are thought toreflect EVLW there is no reference standard available to verifythe EVLW volume and left atrial pressure (LAP) by invasivecatheter

5 Conclusion

This novel simplified ULC scoring method is a rapid non-invasive and reliable tool to assess pulmonary congestionin patients with heart failure Diastolic rather than systolic

performance may be the most important determinant of thedegree of lung congestion in these patients

Conflicts of Interest

The authors declare that they have no conflicts of interest

Acknowledgments

This work was supported by the NSFC (National NaturalScience Foundation of China no 81571683) the Fund ofClinical Application of Capital Project (Z141107002514074)and the Fund of Health Development and Scientific Researchof Capital Projects (no 2014-2-1061)

Supplementary Materials

Videos 1ndash4 Normal lung and three types of ULCs in ultra-sound images which indicate increasing severity of interstitialor alveoli involvementSupplementary 1 Video 1 Normal lung B-lines are absentSupplementary 2 Video 2 Septal syndrome B-lines are scat-tered (about 7mm apart) corresponding to the distance ofsubpleural septaSupplementary 3 Video 3 Interstitial-alveolar syndrome B-lines are confluentSupplementary 4 Video 4 White lung B-lines have coa-lesced resulting in an echographic lung field that is almostcompletely whiteNote Classification of images is according to the distributionof B-lines with the ldquohighestrdquo score in the respiratory cycle(Videos 2ndash4)Videos 5ndash7 Linear probe was used to exclude noncardiacULCs (abnormal pleural line could also generate ULCs)Supplementary 5 Video 5 Normal pleura line and cardiacULCsSupplementary 6 Video 6 Fringed pleural line indicates in-terstitial lung diseaseSupplementary 7 Video 7 Irregular pleural line with micro-consolidations

References

[1] F Mallamaci F A Benedetto R Tripepi et al ldquoDetection ofpulmonary congestion by chest ultrasound in dialysis patientsrdquoJACC Cardiovascular Imaging vol 3 no 6 pp 586ndash594 2010

[2] N R Lange and D P Schuster ldquoThe measurement of lungwaterrdquo Critical Care vol 3 no 2 pp R19ndashR24 1999

[3] S P Collins C J Lindsell A B Storrow and W T AbrahamldquoPrevalence of negative chest radiography results in the emer-gency department patient with decompensated heart failurerdquoAnnals of Emergency Medicine vol 47 no 1 pp 13ndash18 2006

[4] M Gheorghiade F Follath P Ponikowski et al ldquoAssessing andgrading congestion in acute heart failure a scientific statementfrom the acute heart failure committee of the heart failure

BioMed Research International 9

association of the European society of cardiology and endorsedby the European society of intensive care medicinerdquo EuropeanJournal of Heart Failure vol 12 no 5 pp 423ndash433 2010

[5] D A Lichtenstein and G A Meziere ldquoRelevance of lungultrasound in the diagnosis of acute respiratory failure theBLUE protocolrdquo CHEST vol 134 no 1 pp 117ndash125 2008

[6] F Frassi L Gargani S Gligorova Q Ciampi GMottola and EPicano ldquoClinical and echocardiographic determinants of ultra-sound lung cometsrdquo European Heart Journal - CardiovascularImaging vol 8 no 6 pp 474ndash479 2007

[7] L Gargani F Frassi G Soldati P Tesorio M Gheorghiadeand E Picano ldquoUltrasound lung comets for the differentialdiagnosis of acute cardiogenic dyspnoea A comparison withnatriuretic peptidesrdquo European Journal of Heart Failure vol 10no 1 pp 70ndash77 2008

[8] E Agricola T Bove M Oppizzi et al ldquorsquoUltrasound comet-tailimagesrsquo a marker of pulmonary edemamdashA Comparative StudyWith Wedge Pressure And Extravascular Lung Waterrdquo CHESTvol 127 no 5 pp 1690ndash1695 2005

[9] Z Jambrik S Monti V Coppola et al ldquoUsefulness of ultra-sound lung comets as a nonradiologic sign of extravascular lungwaterrdquo American Journal of Cardiology vol 93 no 10 pp 1265ndash1270 2004

[10] K K Ho J L Pinsky W B Kannel and D Levy ldquoTheepidemiology of heart failure the Framingham studyrdquo Journalof the American College of Cardiology vol 22 supplement A no4 pp 6Andash13A 1993

[11] J J McMurray S Adamopoulos S D Anker and et alldquoESC guidelines for the diagnosis and treatment of acute andchronic heart failure 2012 The Task Force for the Diagnosisand Treatment of Acute and Chronic Heart Failurerdquo EuropeanJournal of Heart Failure vol 14 no 8 pp 803-69 2012

[12] A A Hasan and H A Makhlouf ldquoB-lines Transthoracicchest ultrasound signs useful in assessment of interstitial lungdiseasesrdquo Annals of Thoracic Medicine vol 9 no 2 pp 99ndash1032014

[13] M Sperandeo A Varriale G Sperandeo et al ldquoAssessment ofultrasound acoustic artifacts in patients with acute dyspnea Amulticenter studyrdquo Acta Radiologica vol 53 no 8 pp 885ndash8922012

[14] C Giuntini M Pistolesi M Miniati and F Fazio ldquoTheoreticaland practical considerations of measuring extravascular lungwaterrdquo Journal ofThoracic Imaging vol 3 no 3 pp 36ndash43 1988

[15] WA ZoghbiM Enriquez-Sarano E Foster et al ldquoRecommen-dations for evaluation of the severity of native valvular regurgi-tation with two-dimensional and Doppler echocardiographyrdquoJournal of the American Society of Echocardiography vol 16 no7 pp 777ndash802 2003

[16] JM Brennan J E Blair S Goonewardena et al ldquoReappraisal ofthe use of inferior vena cava for estimating right atrial pressurerdquoJournal of the American Society of Echocardiography vol 20 no7 pp 857ndash861 2007

[17] R M Lang L P Badano V Mor-Avi et al ldquoRecommendationsfor cardiac chamber quantification by echocardiography inadults an update from the American Society of Echocardiogra-phy and the European Association of Cardiovascular ImagingrdquoEuropean Heart JournalmdashCardiovascular Imaging vol 28 no 1pp 1ndash39 2015

[18] G Soldati R Copetti and S Sher ldquoSonographic interstitialsyndrome the sound of lung waterrdquo Journal of Ultrasound inMedicine vol 28 no 2 pp 163ndash174 2009

[19] S F Nagueh O A Smiseth C P Appleton et al ldquoRec-ommendations for the evaluation of left ventricular diastolicfunction by echocardiography an update from the americansociety of echocardiography and the european association ofcardiovascular imagingrdquo Journal of the American Society ofEchocardiography vol 29 no 4 pp 277ndash314 2016

[20] F R Lewis V B Elings and J A Sturm ldquoBedsidemeasurementof lung waterrdquo Journal of Surgical Research vol 27 no 4 pp250ndash261 1979

[21] T Tagami M Sawabe S Kushimoto et al ldquoQuantitativediagnosis of diffuse alveolar damage using extravascular lungwaterrdquoCritical CareMedicine vol 41 no 9 pp 2144ndash2150 2013

[22] R E Fromm Jr J Varon and L R Gibbs ldquoCongestive heartfailure and pulmonary edema for the emergency physicianrdquoTheJournal of Emergency Medicine vol 13 no 1 pp 71ndash87 1995

[23] N C Staub ldquoPulmonary edemardquo Physiological Reviews vol 54pp 678ndash811 1974

[24] G Prosen P Klemen M Strnad and S Grmec ldquoCombinationof lung ultrasound (a comet-tail sign) andN-terminal pro-brainnatriuretic peptide in differentiating acute heart failure fromchronic obstructive pulmonary disease and asthma as cause ofacute dyspnea in prehospital emergency settingrdquo Critical Carevol 15 no 2 article no R114 2011

[25] W C Manson J W Bonz K Carmody M Osborne and CL Moore ldquoIdentification of sonographic B-lines with lineartransducer predicts elevated B-type natriuretic peptide levelrdquoWestern Journal of Emergency Medicine vol 12 no 1 pp 102ndash106 2011

[26] E Platz E F Lewis H Uno et al ldquoDetection and prognosticvalue of pulmonary congestion by lung ultrasound in ambula-tory heart failure patientsrdquo European Heart Journal vol 37 no15 pp 1244ndash1251 2016

[27] P Enghard S Rademacher J Nee et al ldquoSimplified lungultrasound protocol shows excellent prediction of extravascularlung water in ventilated intensive care patientsrdquo Critical Carevol 19 no 1 article no 36 2015

[28] A S Liteplo K A Marill T Villen et al ldquoEmergency thoracicultrasound in the differentiation of the etiology of shortnessof breath (ETUDES) sonographic B-lines and N-terminal Pro-brain-type natriuretic peptide in diagnosing congestive heartfailurerdquo Academic Emergency Medicine vol 16 no 3 pp 201ndash210 2009

[29] Z Bitar O Maadarani and K Almerri ldquoSonographic chestB-lines anticipate elevated B-type natriuretic peptide levelirrespective of ejection fractionrdquo Annals of Intensive Care vol5 no 1 article no 56 pp 1ndash6 2015

[30] Y Hadano K Murata J Liu et al ldquoCan transthoracic Dopplerechocardiography predict the discrepancy between left ven-tricular end-diastolic pressure and mean pulmonary capillarywedge pressure in patients with heart failurerdquo CirculationJournal vol 69 no 4 pp 432ndash438 2005

[31] E Agricola E Picano M Oppizzi et al ldquoAssessment of Stress-induced Pulmonary Interstitial Edema by Chest UltrasoundDuring Exercise Echocardiography and its Correlation withLeft Ventricular Functionrdquo Journal of the American Society ofEchocardiography vol 19 no 4 pp 457ndash463 2006

[32] M Packer ldquoAbnormalities of diastolic function as a potentialcause of exercise intolerance in chronic heart failurerdquo Circula-tion vol 81 no 2 pp 78ndash86 1990

[33] R S Vasan and D Levy ldquoDefining diastolic heart failure a callfor standardized diagnostic criteriardquo Circulation vol 101 no 17pp 2118ndash2121 2000

10 BioMed Research International

[34] S K Gandhi J C Powers A-M Nomeir et al ldquoThe pathogen-esis of acute pulmonary edema associated with hypertensionrdquoThe New England Journal of Medicine vol 344 no 1 pp 17ndash222001

[35] A Varela-Roman L Grigorian E Barge P Bassante MG De La Pena and J R Gonzalez-Juanatey ldquoHeart failurein patients with preserved and deteriorated left ventricularejection fractionrdquo Heart vol 91 no 4 pp 489ndash494 2005

Stem Cells International

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Disease Markers

Hindawiwwwhindawicom Volume 2018

BioMed Research International

OncologyJournal of

Hindawiwwwhindawicom Volume 2013

Hindawiwwwhindawicom Volume 2018

Oxidative Medicine and Cellular Longevity

Hindawiwwwhindawicom Volume 2018

PPAR Research

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwwwhindawicom Volume 2018

Journal of

ObesityJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Computational and Mathematical Methods in Medicine

Hindawiwwwhindawicom Volume 2018

Behavioural Neurology

OphthalmologyJournal of

Hindawiwwwhindawicom Volume 2018

Diabetes ResearchJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Research and TreatmentAIDS

Hindawiwwwhindawicom Volume 2018

Gastroenterology Research and Practice

Hindawiwwwhindawicom Volume 2018

Parkinsonrsquos Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwwwhindawicom

Submit your manuscripts atwwwhindawicom

10 BioMed Research International

[34] S K Gandhi J C Powers A-M Nomeir et al ldquoThe pathogen-esis of acute pulmonary edema associated with hypertensionrdquoThe New England Journal of Medicine vol 344 no 1 pp 17ndash222001

[35] A Varela-Roman L Grigorian E Barge P Bassante MG De La Pena and J R Gonzalez-Juanatey ldquoHeart failurein patients with preserved and deteriorated left ventricularejection fractionrdquo Heart vol 91 no 4 pp 489ndash494 2005

Stem Cells International

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Disease Markers

Hindawiwwwhindawicom Volume 2018

BioMed Research International

OncologyJournal of

Hindawiwwwhindawicom Volume 2013

Hindawiwwwhindawicom Volume 2018

Oxidative Medicine and Cellular Longevity

Hindawiwwwhindawicom Volume 2018

PPAR Research

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwwwhindawicom Volume 2018

Journal of

ObesityJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Computational and Mathematical Methods in Medicine

Hindawiwwwhindawicom Volume 2018

Behavioural Neurology

OphthalmologyJournal of

Hindawiwwwhindawicom Volume 2018

Diabetes ResearchJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Research and TreatmentAIDS

Hindawiwwwhindawicom Volume 2018

Gastroenterology Research and Practice

Hindawiwwwhindawicom Volume 2018

Parkinsonrsquos Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwwwhindawicom

Submit your manuscripts atwwwhindawicom

Stem Cells International

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Disease Markers

Hindawiwwwhindawicom Volume 2018

BioMed Research International

OncologyJournal of

Hindawiwwwhindawicom Volume 2013

Hindawiwwwhindawicom Volume 2018

Oxidative Medicine and Cellular Longevity

Hindawiwwwhindawicom Volume 2018

PPAR Research

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwwwhindawicom Volume 2018

Journal of

ObesityJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Computational and Mathematical Methods in Medicine

Hindawiwwwhindawicom Volume 2018

Behavioural Neurology

OphthalmologyJournal of

Hindawiwwwhindawicom Volume 2018

Diabetes ResearchJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Research and TreatmentAIDS

Hindawiwwwhindawicom Volume 2018

Gastroenterology Research and Practice

Hindawiwwwhindawicom Volume 2018

Parkinsonrsquos Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwwwhindawicom

Submit your manuscripts atwwwhindawicom