- Email: [email protected]
Lung ultrasound and short-term prognosis in heart failure patients
Lung ultrasound and short-term prognosis in heart failure patients Chiara Cogliati, Giovanni Casazza, Elisa Ceriani, Daniela Torzillo, Stefano Furlotti, Ilaria Bossi, Tarcisio Vago, Giorgio Costantino, Nicola Montano PII: DOI: Reference:
S0167-5273(16)30912-3 doi: 10.1016/j.ijcard.2016.05.010 IJCA 22566
To appear in:
International Journal of Cardiology
Received date: Revised date: Accepted date:
26 December 2015 20 April 2016 12 May 2016
Please cite this article as: Cogliati Chiara, Casazza Giovanni, Ceriani Elisa, Torzillo Daniela, Furlotti Stefano, Bossi Ilaria, Vago Tarcisio, Costantino Giorgio, Montano Nicola, Lung ultrasound and short-term prognosis in heart failure patients, International Journal of Cardiology (2016), doi: 10.1016/j.ijcard.2016.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title page
PT
Lung ultrasound and short-term prognosis in heart failure patients. Chiara Cogliatia, Giovanni Casazzab, Elisa Cerianic, Daniela Torzillod, Stefano Furlottie, Ilaria
SC RI
Bossif, Tarcisio Vagog, Giorgio Costantinoh, Nicola Montanoi
NU
Total word count: 1944
MA
Running Title: lung ultrasound and prognosis in heart failure patients
Authors’ affiliation and statement of authorship:
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Email:
ED
a
PT
[email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. Department of Biomedical and Clinical Sciences, University of Milan. Italy. Email:
CE
b
AC
[email protected]; This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. c
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
Department of Health and Community Sciences, University of Milan, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. d
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Email:
[email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. e
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Present address:
Department of Medical, Surgical and Health Sciences, University of Trieste, Cattinara Teaching
ACCEPTED MANUSCRIPT Hospital, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. f
PT
Emergency Medicine, L.Sacco Hospital University of Milan. Italy.. Email: [email protected].
This author takes responsibility for all aspects of the reliability and freedom from bias of the data
SC RI
presented and their discussed interpretation. g
Clinical Endocrinology Laboratory, L.Sacco Hospital University of Milan. Italy. Email:
[email protected]. This author takes responsibility for all aspects of the reliability and
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
MA
h
NU
freedom from bias of the data presented and their discussed interpretation.
Department of Health and Community Sciences, University of Milan, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
PT
i
ED
freedom from bias of the data presented and their discussed interpretation.
Department of Health and Community Sciences, University of Milan, Italy. Email:
CE
[email protected]. This author takes responsibility for all aspects of the reliability and
AC
freedom from bias of the data presented and their discussed interpretation.
Adress for correspondence: Elisa Ceriani MD Internal Medicine Department Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico, University of Milan Via F.Sforza 35, Milan, Italy Tel +39 3393055428 E mail: [email protected]
ACCEPTED MANUSCRIPT
PT
Financial support: none
SC RI
Conflict of interest: none
AC
CE
PT
ED
MA
NU
Keywords: lung ultrasound, heart failure, prognosis, B-lines
ACCEPTED MANUSCRIPT
Abstract
PT
Background: Heart failure (HF) is the leading cause of hospitalization for patients older than 65 years, with a 30-day readmission rate of 20-25%. Although several markers have been evaluated to
SC RI
stratify timing of follow-up after an acute decompensation is mostly based on clinical judgment. Lung ultrasound (LUS) has been demonstrated to be a valid tool for the assessment and monitoring of pulmonary congestion. Aim of our study was to evaluate if LUS performed in HF patients at
NU
discharge could predict 100-day hospital readmission or death.
Methods: One-hundred fifty patients were enrolled. The anterolateral chest was scanned to evaluate
MA
the presence of B-lines. A sonographic score was calculated attributing 1 to each positive (≥3 Blines) sector. Clinical, biochemical and echocardiographic data were recorded. A Cox proportional
ED
hazard regression analysis was performed to evaluate the association between variables and 100-
PT
day events.
Results: Follow-up was obtained in 149 patients. Thirty-four events were recorded. Sonographic
CE
score was significantly associated with events (HR 1.19; CI 1.05 to 1.34; p=0.005). On average, the
AC
increase of 1 point in the sonographic score was associated with an increase of approximately 24% in the risk of event within 100 days. At multivariate analysis NTproBNP remained the only independent prognostic factor. Conclusions: We confirmed that B-lines at discharge are a prognostic marker for hospital readmission and death at 100 days in HF patients. Nevertheless, further randomized clinical studies are needed to definitely support the routine use of LUS in the clinical management of HF patients, in combination or not with NT-proBNP
ACCEPTED MANUSCRIPT
Introduction
PT
Heart failure (HF) is the leading cause of hospitalization in industrialized countries and affects 10% of males and 8% of women over 60 years. Even though overall survival rate improved among HF
SC RI
patients in the last decades, 5-year mortality rate remains roughly 50%, worse than many cancers [1]. On short term, 30-day all-cause mortality rate after an episode of acute HF is about 10%, while 30-day all-cause readmission rate is approximately 20-25% [2–4].
NU
Main risk factors affecting prognosis of HF patients are known to be age, sex (better prognosis in
MA
women), underlying cardiomyopathies, New York Heart Association (NYHA) functional class, VO2 peak, left ventricular ejection fraction (EF), diastolic dysfunction, cardiac volumes, right ventricular function, and natriuretic peptides (NP) [5].
ED
Currently, NPs are considered the most reliable risk stratification tools for HF patients. BNP or NT-
PT
proBNP support diagnosis at presentation in HF patients and help in first assessment and treatment, particularly in the setting of dyspnea with unclear etiology. Prognosis estimation is well-established
CE
by pre-discharge value in hospitalized patients, predicting following admissions and deaths [5–7].
AC
Lung ultrasound (LUS) has been demonstrated to be a valid tool for the assessment of pulmonary congestion. Evidence in literature has shown that LUS represents a useful diagnostic tool, especially in the emergency setting [8,9]. B-lines, formerly lung comets, are artefacts identifying the interstitial pattern, which detects pulmonary edema with a high sensitivity and specificity [10]. Concentration of B-lines relates to interstitial burden; the more the interstitial burden increases, the more the B-lines become numerous, bilateral and confluent, up to the so-called “white lung” echographic pattern [11]. A significant correlation has been reported between the number of B-lines and the interstitial edema documented with different techniques, such as CT scan, dilution method and wedge pressure [12,13]. The prognostic value of LUS in hospitalized patients was firstly explored in a study by Frassi and colleagues, in patients admitted to the emergency department for dyspnea or chest pain. They found
ACCEPTED MANUSCRIPT that absence of events was significantly higher in patients without B-lines; on the contrary, more than 30 B-lines were associated with a higher risk of events [14].
PT
Aim of our study was to evaluate if LUS performed at patient’s discharge could predict 100-day hospital readmission or death in a cohort of unselected patients admitted to an internal medicine
SC RI
department for HF from any cause.
Material and methods
NU
Population
MA
We consecutively enrolled 150 patients admitted to the internal medicine department for newly diagnosed or decompensated HF from October 2012 to January 2014. Exclusion criteria were: concomitant acute coronary syndrome; pneumonia in the last month; interstitial lung disease;
ED
pleural effusion more than moderate at the time of discharge; cerebrovascular accident in the
PT
previous two months; bedridden patient. Protocol
CE
For each patient complete clinical history, estimated glomerular filtration rate (eGFR), haemoglobin
AC
and electrolytes values (at discharge) have been recorded. Patients underwent a complete transthoracic color Doppler echocardiography (Philips iE33). All the measurements have been performed according to the recommendations of the American Society of Echocardiography [15]. On the day of discharge from hospital, a physical examination, in particular lung, jugular vein turgescence (JVT) and peripheral edema assessment, was performed. Lung ultrasound was performed by an expert operator (C.C., D.T.) blinded to patient's course, using a 2-5MHz convex transducer (Philips Envisor). A blood sample for NT-proBNP determination was collected after lung ultrasound and dosed at the end of the study. The study protocol was approved by the institutional ethics board of L. Sacco Hospital, University of Milan (385/2011/52/AP) and conforms to the ethical guidelines of the 1975 Declaration of Helsinki. Patients gave their written consent for the enrolment in the study.
ACCEPTED MANUSCRIPT LUS B-lines assessment LUS was performed scanning the anterolateral chest. Eight areas were considered as previously
PT
described in literature -2 anterior and 2 lateral in each side- and B-lines were evaluated [16]. An area was considered as “positive” when at least three B-lines were detected in the same scan and a
SC RI
sonographic score was calculated attributing the value of 1 to each positive sector (sonographic score ranging from 0 to 8).
Moreover, in order to compare the results with those obtained with a more ‘quantitative’ method
NU
previously described in literature [14] anterior chest walls were explored through longitudinal scans
MA
following vertical anatomical lines and 28 segments were considered. B-lines present in each segment were then counted and the total B-lines number was calculated. Moreover, posterior chest was scanned for additional analysis [17].
PT
quantification was 0.96 [18].
ED
As previously reported, in our laboratory the r-value for interobserver variability for B-line
Follow up
CE
After discharge patients were in charge to the outpatients physician, blinded to the ultrasound
AC
results.
Follow-up was performed by contacting patients or their caregivers over the phone 100 days after the date of discharge. All the successive readmissions for HF or deaths from any cause have been considered as “events”. Electronic clinical records have been checked by two physicians to verify the readmission events. Statistical methods For descriptive purposes, quantitative continuous variables were expressed as mean (± standard deviation, SD), or median (interquartile range, IQR), when appropriate. Categorical variables were expressed as count (percentage). The primary endpoint was the composite of readmission for HF and death from any cause at 100 days. In order to assess predictive value of sonographic score on the risk of the primary endpoint, a Cox proportional hazard regression analysis was performed. At
ACCEPTED MANUSCRIPT univariate analysis the following variables were considered as continuous and/or categorized: the sonographic score, total count of B-lines, age, haemoglobin, eGFR, oxygen saturation, presence of
PT
crackles, JVT and peripheral edema at physical examination, presence of pleural effusion, EF, E/E’ ratio, NYHA Class, left atrial (LA) volume, inferior vena cava (IVC) collapsibility, NT-proBNP.
SC RI
Subsequently, the factors that showed a statistically significant association with the endpoint in univariate analysis were entered in a multivariate model. A backward elimination procedure was used to find the best predictive model. For all the fitted Cox models, the proportional hazard
NU
assumption was checked and found to be met. Results were reported as hazard ratios (HR) and 95%
MA
confidence intervals (CI).
We planned to enrol 150 patients. The sample size was calculated assuming that, with an incidence of the primary endpoint of at least 20%, we would have observed at least 30 patients experiencing
ED
the primary endpoint by enrolling 150 patients. With at least 30 events we would have been able to
PT
assess, in our multivariate regression analysis, the effect of at least three covariates simultaneously. P values less than 0.05, two tailed, were considered statistically significant. Statistical analysis was
Results
AC
CE
performed using SAS statistical software (release 9.2).
One hundred and fifty patients were enrolled; only one patient was missed during the follow-up because of erroneous data record. Population characteristics are summarized in Table 1. LUS was performed in all patients. Median of positive sectors was 2 (IQR 0-4). Pleural effusion was present in 55 (36,9%) patients. NT-proBNP was assessed in 99 patients (66.4%). During the 100-days follow up, 34 events were recorded (22.8%), 23 readmissions for HF (15.4%), 11 deaths (7.4%). Data analysis showed significant correlation between the sonographic score and event occurrence (HR 1.19; CI 1.05 to 1.34; p=0.005) showing that, on average, the increase of 1 point in the sonographic score was associated with an increase of approximately 24% in the risk of event
ACCEPTED MANUSCRIPT within 100 days. Moreover, similar results were observed when considering the absolute total number of B-lines (HR 1.005; CI 1.002 to 1.008; p=0,003). A significant correlation with events
bilaterally, total B-lines >15 and total B-lines >30; tab. 2).
Evaluation of the whole thorax,
SC RI
including also posterior sectors, did not change the results.
PT
was found also for LUS dichotomized variables (number of positive sectors ≥ 1 bilaterally, ≥ 2
As to the other considered parameters, eGFR was significantly correlated with events, as well as NYHA and EF, but only when dicotomized (NYHA ≥ 3; EF < 40%; tab. 2). No correlation was
NU
found for age, haemoglobin, oxygen saturation, presence of crackles, JVT and peripheral edema at
MA
physical examination, presence of pleural effusion, E/E’ ratio, LA volume and IVC collapsibility. (Tab. 2).
NT-proBNP was also significantly correlated with events and when performing a multivariate
PT
ED
analysis it remained the only independent prognostic factor (Tab. 2).
Discussion
CE
Our study shows that B-lines represent a prognostic factor for readmission or death in HF patients.
AC
Moreover, multivariate analysis indicated that only NT-proBNP was an independent prognostic factor in this cohort. We also observed that, as expected, eGFR, NYHA and EF were significantly correlated with events, just indicating that renal failure and HFrEF comorbidities are related to a worst outcome. Clinical congestion is the most common cause of HF patients admission and readmission [19]. Residual congestion might play a major role in the evolution and prognosis of decompensated heart failure [20]. However the EVEREST trial showed that even patients with absent or minimal signs and symptoms of congestion at discharge experienced a high mortality and readmission rate [21]. On the other hand in these patients NT-proBNP values were still increased at discharge, revealing a discrepancy between clinical and hemodynamic congestion. This dissociation could be in part the result of a substantial lack of sensitivity of physical examination in detecting signs of residual
ACCEPTED MANUSCRIPT congestion, as confirmed in our study by the lack of correlation with events for peripheral edema, JVT and pulmonary crackles.
PT
Several studies collected evidences showing that the number of B-lines is related to interstitial water burden in HF [13,18] and it has been reported that B-lines are significantly correlated with
SC RI
more established parameters of decompensation among HF outpatients [22]. A first study evaluating the prognostic value of LUS in patients admitted to a cardiologic department for dyspnea or chest pain, found a correlation between the number of B-lines and events [14]. However LUS
NU
was performed the day of admission and diagnosis was heterogeneous.
MA
Two recent papers evaluated the prognostic role of LUS in HF hospitalized patients at discharge [23,24]. Coiro et al studied 60 patients and demonstrated that LUS signs of residual pulmonary congestion, assessed by scanning the anterior chest, were correlated with 90-day readmission or
ED
death. Similar results were reported by Gargani and colleagues. [23]. We confirmed the prognostic
PT
value of B-lines burden in a larger cohort of HF patients showing that the increase of one point in the sonographic score was on average associated with an increase of approximately 24% in the risk
CE
of event within 100 days after discharge. The prognostic value of LUS was confirmed
AC
independently of the method used to evaluate B-lines burden (when considering a 8-zone method or a semi-quantitative method based on counting B-lines). Moreover, due to the clinical importance of evaluating residual congestion in the posterior lung fields, we performed also posterior chest scanning without any change in the results. Nevertheless, in our study B-lines did not maintain the association with events at the multivariate analysis, being NT-proBNP the only independent prognostic factor. This discrepancy could be at least partly due to the differences in the studied population between our and previous studies. First, in our study patients were admitted to an internal medicine department and were characterized by a higher mean age; both these characteristics might account for a higher rate of comorbidities that could have affected the specificity of B-lines. In older patients, frequently presenting with associated chronic pulmonary diseases, pleural and parenchymal scars relative to previous diseases,
ACCEPTED MANUSCRIPT lung bronchiectasis of focal fibrosis may generate B-lines, which in these cases are not expression of extravascular lung water [25]. Second, mean EF in our cohort was higher and almost two thirds
PT
of patients had EF>40. Recently, volume overload profiles were studied in patients with HF with preserved EF (HFpEF) and HF with reduced EF (HFrEF) [26]. The authors demonstrated that in
SC RI
response to diuretic therapy, patients with HFrEF revealed a greater reduction in intravascular volume, however with less mobilization of total body fluid compared with HFpEF. In these latter, the majority of body fluid loss was derived from the interstitial compartment with an unchanged
NU
intravascular volume overload. In this perspective, HFpEF patients may present a real discrepancy
MA
between residual pulmonary congestion as assessed by LUS and hemodynamic and intravascular volume related congestion as better assessed by NT-proBNP, thus making this last more sensitive in this subgroup of patients.
ED
Clinical perspectives
PT
LUS represent a quick to perform and easy to learn tool in the clinician hands for the evaluation of residual pulmonary congestion. We confirm that it allows a prognostic risk stratification at
CE
discharge in patient hospitalized for HF. In this perspective LUS potentially could help to guide the
AC
timing of discharge from AHF hospitalization, the follow up scheduling and the therapy tailoring [27,28]. Further randomized clinical studies are needed to definitely support the routine use of LUS in the clinical management of HF patients, in combination or not with NT-proBNP .
ACCEPTED MANUSCRIPT References M. Hoshijima, K.R. Chien, Mixed signals in heart failure: cancer rules., J. Clin. Invest. 109 (2002) 849–55. doi:10.1172/JCI15380.
[2]
E. Braunwald, Heart failure, JACC Hear. Fail. 1 (2013) 1–20. doi:10.1016/j.jchf.2012.10.002.
[3]
D. Mozaffarian, E.J. Benjamin, A.S. Go, D.K. Arnett, M.J. Blaha, M. Cushman, et al., Heart Disease and Stroke Statistics-2015 Update: A Report From the American Heart Association., Circulation. 131 (2014) e29–322. http://circ.ahajournals.org/content/131/4/e29.short?rss=1&ssource=mfr (accessed December 19, 2014).
[4]
J.S. Ross, J. Chen, Z. Lin, H. Bueno, J.P. Curtis, P.S. Keenan, et al., Recent national trends in readmission rates after heart failure hospitalization, Circ. Hear. Fail. 3 (2010) 97–103. doi:10.1161/CIRCHEARTFAILURE.109.885210.
[5]
E.S. Ketchum, W.C. Levy, Establishing Prognosis in Heart Failure: A Multimarker Approach, Prog. Cardiovasc. Dis. 54 (2011) 86–96. doi:10.1016/j.pcad.2011.03.003.
[6]
J.L. Januzzi, A.S. Maisel, M. Silver, Y. Xue, C. DeFilippi, Natriuretic Peptide Testing for Predicting Adverse Events Following Heart Failure Hospitalization, Congest. Hear. Fail. 18 (2012) S9–S13. doi:10.1111/j.1751-7133.2012.00306.x.
[7]
C.W. Yancy, M. Jessup, B. Bozkurt, J. Butler, D.E. Casey, M.H. Drazner, et al., 2013 ACCF/AHA guideline for the management of heart failure: A report of the american college of cardiology foundation/american heart association task force on practice guidelines, Circulation. 128 (2013) 240–327. doi:10.1161/CIR.0b013e31829e8776.
[8]
G. Volpicelli, M. Elbarbary, M. Blaivas, D.A. Lichtenstein, G. Mathis, A.W. Kirkpatrick, et al., International evidence-based recommendations for point-of-care lung ultrasound, Intensive Care Med. 38 (2012) 577–591. doi:10.1007/s00134-012-2513-4.
[9]
E. Pivetta, A. Goffi, E. Lupia, M. Tizzani, G. Porrino, E. Ferreri, et al., Lung UltrasoundImplemented Diagnosis of Acute Decompensated Heart Failure in the ED: A SIMEU Multicenter Study., Chest. 148 (2015) 202–10. doi:10.1378/chest.14-2608.
AC
CE
PT
ED
MA
NU
SC RI
PT
[1]
[10] D. Lichtenstein, G. Mezière, Relevance of lung ultrasound in the diagnosis of acute respiratory failure the BLUE protocol, Chest. 134 (2008) 117–125. doi:10.1378/chest.072800. [11] G. Volpicelli, V. Caramello, L. Cardinale, A. Mussa, F. Bar, M.F. Frascisco, Bedside ultrasound of the lung for the monitoring of acute decompensated heart failure, Am. J. Emerg. Med. 26 (2008) 585–591. doi:10.1016/j.ajem.2007.09.014. [12] D. Lichtenstein, G. Meziere, P. Biderman, A. Gepner, O. Barre, The comet-tail artefact: an ultrasound sign of alveolar-interstitial syndrome., Am J Respir Crit Care Med. 156 (1997) 1640–1646. doi:10.1164/ajrccm.156.5.96-07096. [13] E. Agricola, T. Bove, M. Oppizzi, G. Marino, A. Zangrillo, A. Margonato, et al., “Ultrasound comet-tail images”: a marker of pulmonary edema: a comparative study with wedge pressure and extravascular lung water., Chest. 127 (2005) 1690–5. doi:10.1378/chest.127.5.1690. [14] F. Frassi, L. Gargani, P. Tesorio, M. Raciti, G. Mottola, E. Picano, Prognostic Value of Extravascular Lung Water Assessed With Ultrasound Lung Comets by Chest Sonography in Patients With Dyspnea and/or Chest Pain, J. Card. Fail. 13 (2007) 830–835. doi:10.1016/j.cardfail.2007.07.003.
ACCEPTED MANUSCRIPT [15] R.M. Lang, L.P. Badano, V. Mor-Avi, J. Afilalo, A. Armstrong, L. Ernande, et al., Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging., Eur. Heart J. Cardiovasc. Imaging. 16 (2015) 233–70. doi:10.1093/ehjci/jev014.
SC RI
PT
[16] G. Volpicelli, A. Mussa, G. Garofalo, L. Cardinale, G. Casoli, F. Perotto, et al., Bedside lung ultrasound in the assessment of alveolar-interstitial syndrome, Am. J. Emerg. Med. 24 (2006) 689–696. doi:10.1016/j.ajem.2006.02.013. [17] L. Gargani, M. Doveri, L. D’Errico, F. Frassi, M.L. Bazzichi, A. Delle Sedie, et al., Ultrasound lung comets in systemic sclerosis: a chest sonography hallmark of pulmonary interstitial fibrosis., Rheumatology (Oxford). 48 (2009) 1382–1387. doi:10.1093/rheumatology/kep263.
MA
NU
[18] M. Trezzi, D. Torzillo, E. Ceriani, G. Costantino, S. Caruso, P.T. Damavandi, et al., Lung ultrasonography for the assessment of rapid extravascular water variation: Evidence from hemodialysis patients, Intern. Emerg. Med. 8 (2013) 409–415. doi:10.1007/s11739-0110625-4. [19] M. Gheorghiade, G. Filippatos, L. De Luca, J. Burnett, Congestion in acute heart failure syndromes: an essential target of evaluation and treatment., Am. J. Med. 119 (2006) S3–S10. doi:10.1016/j.amjmed.2006.09.011.
PT
ED
[20] K.F. Adams, G.C. Fonarow, C.L. Emerman, T.H. LeJemtel, M.R. Costanzo, W.T. Abraham, et al., Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE)., Am. Heart J. 149 (2005) 209–16. doi:10.1016/j.ahj.2004.08.005.
CE
[21] A.P. Ambrosy, P.S. Pang, S. Khan, M.A. Konstam, G.C. Fonarow, B. Traver, et al., Clinical course and predictive value of congestion during hospitalization in patients admitted for worsening signs and symptoms of heart failure with reduced ejection fraction: findings from the EVEREST trial, Eur. Heart J. 34 (2013) 835–843. doi:10.1093/eurheartj/ehs444.
AC
[22] M.H. Miglioranza, L. Gargani, R.T. Sant’Anna, M.M. Rover, V.M. Martins, A. Mantovani, et al., Lung ultrasound for the evaluation of pulmonary congestion in outpatients: A comparison with clinical assessment, natriuretic peptides, and echocardiography, JACC Cardiovasc. Imaging. 6 (2013) 1141–1151. doi:10.1016/j.jcmg.2013.08.004. [23] L. Gargani, P.S. Pang, F. Frassi, M.H. Miglioranza, F.L. Dini, P. Landi, et al., Persistent pulmonary congestion before discharge predicts rehospitalization in heart failure: a lung ultrasound study., Cardiovasc. Ultrasound. 13 (2015) 40. /pmc/articles/PMC4558829/?report=abstract. [24] S. Coiro, P. Rossignol, G. Ambrosio, E. Carluccio, G. Alunni, A. Murrone, et al., Prognostic value of residual pulmonary congestion at discharge assessed by lung ultrasound imaging in heart failure., Eur. J. Heart Fail. 17 (2015) 1172–81. doi:10.1002/ejhf.344. [25] F. Ciccarese, A.M. Chiesa, F. Feletti, L. Vizioli, M. Pasquali, P. Forti, et al., The Senile Lung as a Possible Source of Pitfalls on Chest Ultrasonography and Computed Tomography, Respiration. 90 (2015) 56–62. http://www.ncbi.nlm.nih.gov/pubmed/26044398. [26] W.L. Miller, B.P. Mullan, Volume Overload Profiles in Patients With Preserved (HFpEF) and Reduced (HFrEF) Ejection Fraction Chronic Heart Failure, JACC Hear. Fail. (2016). http://www.ncbi.nlm.nih.gov/pubmed/26970830.
ACCEPTED MANUSCRIPT [27] L. Gargani, Prognosis in heart failure: look at the lungs, Eur. J. Heart Fail. 17 (2015) 1086– 1088. http://www.ncbi.nlm.nih.gov/pubmed/26470753.
AC
CE
PT
ED
MA
NU
SC RI
PT
[28] S. Coiro, T. Chouihed, N. Girerd, Lung ultrasound - The extension of clinical examination in patients with acute heart failure: Reply., Eur. J. Heart Fail. 18 (2016) 215. http://www.ncbi.nlm.nih.gov/pubmed/26726954.
ACCEPTED MANUSCRIPT Figure legends Figure 1:
PT
Example of two different US scans in the same patient: in the upper panel, no B-lines are detectable
AC
CE
PT
ED
MA
NU
SC RI
while multiple and confluent B-lines are clearly seen in the lower panel.
ACCEPTED MANUSCRIPT Tables
AC
CE
PT
ED
MA
NU
SC RI
Age years 80.7 (75.9 – 84.7)* Sex F % (n) 57.7% (86) NYHA Class % (n) I 14.9% (22) II 58.8% (87) III 23.0% (34) IV 3.4% (5) Etiology of current HF hospitalization % (n) Coronary heart disease 41.6% (62) Hypertensive Heart Disease 20.1% (30) Valvular Heart Disease 32.9% (49) Hypertrophic cardiomyopathy 2.0% (3) Primitive dilated cardiomyopathy 6.7% (10) Toxic Cardiomyopathy 0.7% (1) Metabolic Cardiomyopathy 0.7% (1) Comorbidities % (n) Hypertension 62.4% (93) Atrial fibrillation 20.8% (31) Chronic obstructive pulmonary 16.1% (24) disease Chronic kidney disease 24.2% (36) Diabetes Mellitus 34.2% (51) Vitals and Chemistry Heart rate bpm 72.0 (10.5)§ Systolic blood pressure mmHg 121.0 (15.9)§ diastolic blood pressure mmHg 68.9 (10.4)§ Oxigen saturation % 96.1 (2.1)§ Hemoglobin g/dl 11.6 (1.7)§ Na+ mEq/l 139.2 (3.1)§ GFR ml/min 55.1 (42.6-76.1)* NT-proBNP pg/ml 2407 (1032-5273) * NT-proBNP categorical %(n) <1500 pg/ml 37.4% (37) 1500-5000 pg/ml 35.3% (35) >5000 pg/ml 27.3% (27) Drugs % (n) β-blockers 65.5% (97) ACE-i/ARBs 68.9% (102) Furosemide 95.7% (141) Aldosterone receptor antagonists 38.5% (57) Digoxin 23.6% (35) Antiarrythmic agents 12.2% (18) Pace Maker 18.8% (28) Implantable cardioverter-defibrillator 6.0% (9) Cardiac resynchronization therapy 5.4% (8) Echocardiography EF % 47.7 (16.9)§
PT
Table 1: Baseline population characteristic
ACCEPTED MANUSCRIPT
MA
NU
SC RI
PT
EF<40 % (n) 36.2% (54) 2 LVEDVi ml/m 66.6 (28.7)§ LVESVi ml/m2 38.5 (27.0)§ LA vol i ml/m2 68.4 (27.7)§ 2 RA vol i ml/m 35.5 (20.4)§ E/A 1.8 (1.3)§ E/E’ # 14.0 (7.2)§ E/E’ >15 % (n) # 41.6% (49) RVd mm 34.5 (7.5)§ TAPSE mm 16.5 (4.8)§ RV-RA gr mmHg 32.4 (11.2) § IVC-CI % 49.5 (28.1)§ Lung Ultrasound Positive sectors 2 (0-4)* Bilaterally ≥1 40.3% (60) Bilaterally ≥2 24.2% (36) * data reported as median (IQR) § data reported as mean (SD) # E/E’ not feasible in 31 patients (anular calcifications, prosthetic valves).
ED
Abbreviations
SD:standard deviation
PT
IQR: interquartile range
CE
GFR: Glomerular filtration rate
ACE-I: angiotensin coverting enzyme inhibitors
AC
ARBs: angiotensin receptor blockers; EF: ejection fraction
LVESV: left ventricular end-systolic volume (indexed) LVEDVi: left ventricular end-diastolic volume (indexed) LA vol i: left atrium volume (indexed) RA vol i: right atrium volume (indexed) RVd: right ventricle diameter RV-RA gr: right ventricular-right atrial gradient IVC-CI: inferior vena cava collapsibility index
ACCEPTED MANUSCRIPT Table 2: Prediction of readmission for HF and death from any cause at 100 days. Results of univariate and multivariate Cox analysis.
PT
Multivariate HR (95% CI) -
p value
SC RI
p value 0.116 0.009
0.034 0.028 0.692 0.502 0.0953
NU
1 2.44 (1.24-4.82) 0.98 2.14 1.04 (0.85 to 1.26) 1.05 (0.89 to 1.25)
MA
1 1.79 (0.90 to 3.55)
ED
1 1.01 (0.49 to 2.08) 0.98 (0.97 to 1.00)
PT
1 2.19 (1.12 to 4.29) 1.03 (0.98 to 1.08) 1 1.35 (0.62 to 2.97) 1.00 (0.99 to 1.00)
AC
CE
Age, years NYHA <3 ≥3 GFR GFR <50ml/min Hemoglobin Oxigen saturation Peripheral edema Absent Present Presence of crackles Absent Present EF EF categorical ≥40% <40% E/E’ E/E’ categorical ≤15 >15 LA vol i Pleural effusion Absent Present IVC-CI IVC-CI categorical ≥50% <50% JV turgescence Absent Present NT-proBNP NT-proBNP categorical <1500 pg/ml 1500-5000 pg/ml >5000 pg/ml Sonographic score
Univariate HR (95% CI) 1.04 (0.99 to 1.09)
Positive sectors bilaterally ≥1 ≥2
1 1.78 (0.91 to 3.49) 0.99 (0.97 to 1.00)
0.978
0.148 0.022
-
0.210 0.449
0.556 0.0925
0.133 0.540
1 1.24 (0.61 to 2.52) 0.360 1 1.51 (0.63 to 3.65) 1.18 (1.07 to 1.31) 1 3.92 (0.81 to 18.88) 9.95 (2.22 to 44.53) 1.19 (1.05 to 1.34)
3.62 (1.76 to 7.43) 2.27 (1.12 to 4.59)
0.001 0.005
0.005 0.022 0.0005 0.022
1 0.005 3.92 (0.81 to 18.88) 9.95 (2.22 to 44.53) -
ACCEPTED MANUSCRIPT 1.005 (1.002 to 1.008)
Total B-lines B-lines categorical ≤15 >15 B-lines categorical ≤30 >30
0.003 0.019
1 3.10 (1.20 to 8.02) 0.005
NU
SC RI
PT
1 2.85 (1.36 to 5.96)
Abbreviations
GFR: Glomerular filtration rate
ED
LA vol i: left atrium volume (indexed)
MA
CI: confidence interval
IVC-CI: inferior vena cava collapsibility index
AC
CE
PT
JV: jugular vein
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI
PT
Figure 1
S0167-5273(16)30912-3 doi: 10.1016/j.ijcard.2016.05.010 IJCA 22566
To appear in:
International Journal of Cardiology
Received date: Revised date: Accepted date:
26 December 2015 20 April 2016 12 May 2016
Please cite this article as: Cogliati Chiara, Casazza Giovanni, Ceriani Elisa, Torzillo Daniela, Furlotti Stefano, Bossi Ilaria, Vago Tarcisio, Costantino Giorgio, Montano Nicola, Lung ultrasound and short-term prognosis in heart failure patients, International Journal of Cardiology (2016), doi: 10.1016/j.ijcard.2016.05.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title page
PT
Lung ultrasound and short-term prognosis in heart failure patients. Chiara Cogliatia, Giovanni Casazzab, Elisa Cerianic, Daniela Torzillod, Stefano Furlottie, Ilaria
SC RI
Bossif, Tarcisio Vagog, Giorgio Costantinoh, Nicola Montanoi
NU
Total word count: 1944
MA
Running Title: lung ultrasound and prognosis in heart failure patients
Authors’ affiliation and statement of authorship:
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Email:
ED
a
PT
[email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. Department of Biomedical and Clinical Sciences, University of Milan. Italy. Email:
CE
b
AC
[email protected]; This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. c
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
Department of Health and Community Sciences, University of Milan, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. d
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Email:
[email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. e
Department of Internal Medicine, L.Sacco Hospital University of Milan. Italy. Present address:
Department of Medical, Surgical and Health Sciences, University of Trieste, Cattinara Teaching
ACCEPTED MANUSCRIPT Hospital, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. f
PT
Emergency Medicine, L.Sacco Hospital University of Milan. Italy.. Email: [email protected].
This author takes responsibility for all aspects of the reliability and freedom from bias of the data
SC RI
presented and their discussed interpretation. g
Clinical Endocrinology Laboratory, L.Sacco Hospital University of Milan. Italy. Email:
[email protected]. This author takes responsibility for all aspects of the reliability and
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
MA
h
NU
freedom from bias of the data presented and their discussed interpretation.
Department of Health and Community Sciences, University of Milan, Italy. Email: [email protected]. This author takes responsibility for all aspects of the reliability and
Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico,
PT
i
ED
freedom from bias of the data presented and their discussed interpretation.
Department of Health and Community Sciences, University of Milan, Italy. Email:
CE
[email protected]. This author takes responsibility for all aspects of the reliability and
AC
freedom from bias of the data presented and their discussed interpretation.
Adress for correspondence: Elisa Ceriani MD Internal Medicine Department Department of Internal Medicine, Ca Granda Foundation IRCCS, Ospedale Maggiore Policlinico, University of Milan Via F.Sforza 35, Milan, Italy Tel +39 3393055428 E mail: [email protected]
ACCEPTED MANUSCRIPT
PT
Financial support: none
SC RI
Conflict of interest: none
AC
CE
PT
ED
MA
NU
Keywords: lung ultrasound, heart failure, prognosis, B-lines
ACCEPTED MANUSCRIPT
Abstract
PT
Background: Heart failure (HF) is the leading cause of hospitalization for patients older than 65 years, with a 30-day readmission rate of 20-25%. Although several markers have been evaluated to
SC RI
stratify timing of follow-up after an acute decompensation is mostly based on clinical judgment. Lung ultrasound (LUS) has been demonstrated to be a valid tool for the assessment and monitoring of pulmonary congestion. Aim of our study was to evaluate if LUS performed in HF patients at
NU
discharge could predict 100-day hospital readmission or death.
Methods: One-hundred fifty patients were enrolled. The anterolateral chest was scanned to evaluate
MA
the presence of B-lines. A sonographic score was calculated attributing 1 to each positive (≥3 Blines) sector. Clinical, biochemical and echocardiographic data were recorded. A Cox proportional
ED
hazard regression analysis was performed to evaluate the association between variables and 100-
PT
day events.
Results: Follow-up was obtained in 149 patients. Thirty-four events were recorded. Sonographic
CE
score was significantly associated with events (HR 1.19; CI 1.05 to 1.34; p=0.005). On average, the
AC
increase of 1 point in the sonographic score was associated with an increase of approximately 24% in the risk of event within 100 days. At multivariate analysis NTproBNP remained the only independent prognostic factor. Conclusions: We confirmed that B-lines at discharge are a prognostic marker for hospital readmission and death at 100 days in HF patients. Nevertheless, further randomized clinical studies are needed to definitely support the routine use of LUS in the clinical management of HF patients, in combination or not with NT-proBNP
ACCEPTED MANUSCRIPT
Introduction
PT
Heart failure (HF) is the leading cause of hospitalization in industrialized countries and affects 10% of males and 8% of women over 60 years. Even though overall survival rate improved among HF
SC RI
patients in the last decades, 5-year mortality rate remains roughly 50%, worse than many cancers [1]. On short term, 30-day all-cause mortality rate after an episode of acute HF is about 10%, while 30-day all-cause readmission rate is approximately 20-25% [2–4].
NU
Main risk factors affecting prognosis of HF patients are known to be age, sex (better prognosis in
MA
women), underlying cardiomyopathies, New York Heart Association (NYHA) functional class, VO2 peak, left ventricular ejection fraction (EF), diastolic dysfunction, cardiac volumes, right ventricular function, and natriuretic peptides (NP) [5].
ED
Currently, NPs are considered the most reliable risk stratification tools for HF patients. BNP or NT-
PT
proBNP support diagnosis at presentation in HF patients and help in first assessment and treatment, particularly in the setting of dyspnea with unclear etiology. Prognosis estimation is well-established
CE
by pre-discharge value in hospitalized patients, predicting following admissions and deaths [5–7].
AC
Lung ultrasound (LUS) has been demonstrated to be a valid tool for the assessment of pulmonary congestion. Evidence in literature has shown that LUS represents a useful diagnostic tool, especially in the emergency setting [8,9]. B-lines, formerly lung comets, are artefacts identifying the interstitial pattern, which detects pulmonary edema with a high sensitivity and specificity [10]. Concentration of B-lines relates to interstitial burden; the more the interstitial burden increases, the more the B-lines become numerous, bilateral and confluent, up to the so-called “white lung” echographic pattern [11]. A significant correlation has been reported between the number of B-lines and the interstitial edema documented with different techniques, such as CT scan, dilution method and wedge pressure [12,13]. The prognostic value of LUS in hospitalized patients was firstly explored in a study by Frassi and colleagues, in patients admitted to the emergency department for dyspnea or chest pain. They found
ACCEPTED MANUSCRIPT that absence of events was significantly higher in patients without B-lines; on the contrary, more than 30 B-lines were associated with a higher risk of events [14].
PT
Aim of our study was to evaluate if LUS performed at patient’s discharge could predict 100-day hospital readmission or death in a cohort of unselected patients admitted to an internal medicine
SC RI
department for HF from any cause.
Material and methods
NU
Population
MA
We consecutively enrolled 150 patients admitted to the internal medicine department for newly diagnosed or decompensated HF from October 2012 to January 2014. Exclusion criteria were: concomitant acute coronary syndrome; pneumonia in the last month; interstitial lung disease;
ED
pleural effusion more than moderate at the time of discharge; cerebrovascular accident in the
PT
previous two months; bedridden patient. Protocol
CE
For each patient complete clinical history, estimated glomerular filtration rate (eGFR), haemoglobin
AC
and electrolytes values (at discharge) have been recorded. Patients underwent a complete transthoracic color Doppler echocardiography (Philips iE33). All the measurements have been performed according to the recommendations of the American Society of Echocardiography [15]. On the day of discharge from hospital, a physical examination, in particular lung, jugular vein turgescence (JVT) and peripheral edema assessment, was performed. Lung ultrasound was performed by an expert operator (C.C., D.T.) blinded to patient's course, using a 2-5MHz convex transducer (Philips Envisor). A blood sample for NT-proBNP determination was collected after lung ultrasound and dosed at the end of the study. The study protocol was approved by the institutional ethics board of L. Sacco Hospital, University of Milan (385/2011/52/AP) and conforms to the ethical guidelines of the 1975 Declaration of Helsinki. Patients gave their written consent for the enrolment in the study.
ACCEPTED MANUSCRIPT LUS B-lines assessment LUS was performed scanning the anterolateral chest. Eight areas were considered as previously
PT
described in literature -2 anterior and 2 lateral in each side- and B-lines were evaluated [16]. An area was considered as “positive” when at least three B-lines were detected in the same scan and a
SC RI
sonographic score was calculated attributing the value of 1 to each positive sector (sonographic score ranging from 0 to 8).
Moreover, in order to compare the results with those obtained with a more ‘quantitative’ method
NU
previously described in literature [14] anterior chest walls were explored through longitudinal scans
MA
following vertical anatomical lines and 28 segments were considered. B-lines present in each segment were then counted and the total B-lines number was calculated. Moreover, posterior chest was scanned for additional analysis [17].
PT
quantification was 0.96 [18].
ED
As previously reported, in our laboratory the r-value for interobserver variability for B-line
Follow up
CE
After discharge patients were in charge to the outpatients physician, blinded to the ultrasound
AC
results.
Follow-up was performed by contacting patients or their caregivers over the phone 100 days after the date of discharge. All the successive readmissions for HF or deaths from any cause have been considered as “events”. Electronic clinical records have been checked by two physicians to verify the readmission events. Statistical methods For descriptive purposes, quantitative continuous variables were expressed as mean (± standard deviation, SD), or median (interquartile range, IQR), when appropriate. Categorical variables were expressed as count (percentage). The primary endpoint was the composite of readmission for HF and death from any cause at 100 days. In order to assess predictive value of sonographic score on the risk of the primary endpoint, a Cox proportional hazard regression analysis was performed. At
ACCEPTED MANUSCRIPT univariate analysis the following variables were considered as continuous and/or categorized: the sonographic score, total count of B-lines, age, haemoglobin, eGFR, oxygen saturation, presence of
PT
crackles, JVT and peripheral edema at physical examination, presence of pleural effusion, EF, E/E’ ratio, NYHA Class, left atrial (LA) volume, inferior vena cava (IVC) collapsibility, NT-proBNP.
SC RI
Subsequently, the factors that showed a statistically significant association with the endpoint in univariate analysis were entered in a multivariate model. A backward elimination procedure was used to find the best predictive model. For all the fitted Cox models, the proportional hazard
NU
assumption was checked and found to be met. Results were reported as hazard ratios (HR) and 95%
MA
confidence intervals (CI).
We planned to enrol 150 patients. The sample size was calculated assuming that, with an incidence of the primary endpoint of at least 20%, we would have observed at least 30 patients experiencing
ED
the primary endpoint by enrolling 150 patients. With at least 30 events we would have been able to
PT
assess, in our multivariate regression analysis, the effect of at least three covariates simultaneously. P values less than 0.05, two tailed, were considered statistically significant. Statistical analysis was
Results
AC
CE
performed using SAS statistical software (release 9.2).
One hundred and fifty patients were enrolled; only one patient was missed during the follow-up because of erroneous data record. Population characteristics are summarized in Table 1. LUS was performed in all patients. Median of positive sectors was 2 (IQR 0-4). Pleural effusion was present in 55 (36,9%) patients. NT-proBNP was assessed in 99 patients (66.4%). During the 100-days follow up, 34 events were recorded (22.8%), 23 readmissions for HF (15.4%), 11 deaths (7.4%). Data analysis showed significant correlation between the sonographic score and event occurrence (HR 1.19; CI 1.05 to 1.34; p=0.005) showing that, on average, the increase of 1 point in the sonographic score was associated with an increase of approximately 24% in the risk of event
ACCEPTED MANUSCRIPT within 100 days. Moreover, similar results were observed when considering the absolute total number of B-lines (HR 1.005; CI 1.002 to 1.008; p=0,003). A significant correlation with events
bilaterally, total B-lines >15 and total B-lines >30; tab. 2).
Evaluation of the whole thorax,
SC RI
including also posterior sectors, did not change the results.
PT
was found also for LUS dichotomized variables (number of positive sectors ≥ 1 bilaterally, ≥ 2
As to the other considered parameters, eGFR was significantly correlated with events, as well as NYHA and EF, but only when dicotomized (NYHA ≥ 3; EF < 40%; tab. 2). No correlation was
NU
found for age, haemoglobin, oxygen saturation, presence of crackles, JVT and peripheral edema at
MA
physical examination, presence of pleural effusion, E/E’ ratio, LA volume and IVC collapsibility. (Tab. 2).
NT-proBNP was also significantly correlated with events and when performing a multivariate
PT
ED
analysis it remained the only independent prognostic factor (Tab. 2).
Discussion
CE
Our study shows that B-lines represent a prognostic factor for readmission or death in HF patients.
AC
Moreover, multivariate analysis indicated that only NT-proBNP was an independent prognostic factor in this cohort. We also observed that, as expected, eGFR, NYHA and EF were significantly correlated with events, just indicating that renal failure and HFrEF comorbidities are related to a worst outcome. Clinical congestion is the most common cause of HF patients admission and readmission [19]. Residual congestion might play a major role in the evolution and prognosis of decompensated heart failure [20]. However the EVEREST trial showed that even patients with absent or minimal signs and symptoms of congestion at discharge experienced a high mortality and readmission rate [21]. On the other hand in these patients NT-proBNP values were still increased at discharge, revealing a discrepancy between clinical and hemodynamic congestion. This dissociation could be in part the result of a substantial lack of sensitivity of physical examination in detecting signs of residual
ACCEPTED MANUSCRIPT congestion, as confirmed in our study by the lack of correlation with events for peripheral edema, JVT and pulmonary crackles.
PT
Several studies collected evidences showing that the number of B-lines is related to interstitial water burden in HF [13,18] and it has been reported that B-lines are significantly correlated with
SC RI
more established parameters of decompensation among HF outpatients [22]. A first study evaluating the prognostic value of LUS in patients admitted to a cardiologic department for dyspnea or chest pain, found a correlation between the number of B-lines and events [14]. However LUS
NU
was performed the day of admission and diagnosis was heterogeneous.
MA
Two recent papers evaluated the prognostic role of LUS in HF hospitalized patients at discharge [23,24]. Coiro et al studied 60 patients and demonstrated that LUS signs of residual pulmonary congestion, assessed by scanning the anterior chest, were correlated with 90-day readmission or
ED
death. Similar results were reported by Gargani and colleagues. [23]. We confirmed the prognostic
PT
value of B-lines burden in a larger cohort of HF patients showing that the increase of one point in the sonographic score was on average associated with an increase of approximately 24% in the risk
CE
of event within 100 days after discharge. The prognostic value of LUS was confirmed
AC
independently of the method used to evaluate B-lines burden (when considering a 8-zone method or a semi-quantitative method based on counting B-lines). Moreover, due to the clinical importance of evaluating residual congestion in the posterior lung fields, we performed also posterior chest scanning without any change in the results. Nevertheless, in our study B-lines did not maintain the association with events at the multivariate analysis, being NT-proBNP the only independent prognostic factor. This discrepancy could be at least partly due to the differences in the studied population between our and previous studies. First, in our study patients were admitted to an internal medicine department and were characterized by a higher mean age; both these characteristics might account for a higher rate of comorbidities that could have affected the specificity of B-lines. In older patients, frequently presenting with associated chronic pulmonary diseases, pleural and parenchymal scars relative to previous diseases,
ACCEPTED MANUSCRIPT lung bronchiectasis of focal fibrosis may generate B-lines, which in these cases are not expression of extravascular lung water [25]. Second, mean EF in our cohort was higher and almost two thirds
PT
of patients had EF>40. Recently, volume overload profiles were studied in patients with HF with preserved EF (HFpEF) and HF with reduced EF (HFrEF) [26]. The authors demonstrated that in
SC RI
response to diuretic therapy, patients with HFrEF revealed a greater reduction in intravascular volume, however with less mobilization of total body fluid compared with HFpEF. In these latter, the majority of body fluid loss was derived from the interstitial compartment with an unchanged
NU
intravascular volume overload. In this perspective, HFpEF patients may present a real discrepancy
MA
between residual pulmonary congestion as assessed by LUS and hemodynamic and intravascular volume related congestion as better assessed by NT-proBNP, thus making this last more sensitive in this subgroup of patients.
ED
Clinical perspectives
PT
LUS represent a quick to perform and easy to learn tool in the clinician hands for the evaluation of residual pulmonary congestion. We confirm that it allows a prognostic risk stratification at
CE
discharge in patient hospitalized for HF. In this perspective LUS potentially could help to guide the
AC
timing of discharge from AHF hospitalization, the follow up scheduling and the therapy tailoring [27,28]. Further randomized clinical studies are needed to definitely support the routine use of LUS in the clinical management of HF patients, in combination or not with NT-proBNP .
ACCEPTED MANUSCRIPT References M. Hoshijima, K.R. Chien, Mixed signals in heart failure: cancer rules., J. Clin. Invest. 109 (2002) 849–55. doi:10.1172/JCI15380.
[2]
E. Braunwald, Heart failure, JACC Hear. Fail. 1 (2013) 1–20. doi:10.1016/j.jchf.2012.10.002.
[3]
D. Mozaffarian, E.J. Benjamin, A.S. Go, D.K. Arnett, M.J. Blaha, M. Cushman, et al., Heart Disease and Stroke Statistics-2015 Update: A Report From the American Heart Association., Circulation. 131 (2014) e29–322. http://circ.ahajournals.org/content/131/4/e29.short?rss=1&ssource=mfr (accessed December 19, 2014).
[4]
J.S. Ross, J. Chen, Z. Lin, H. Bueno, J.P. Curtis, P.S. Keenan, et al., Recent national trends in readmission rates after heart failure hospitalization, Circ. Hear. Fail. 3 (2010) 97–103. doi:10.1161/CIRCHEARTFAILURE.109.885210.
[5]
E.S. Ketchum, W.C. Levy, Establishing Prognosis in Heart Failure: A Multimarker Approach, Prog. Cardiovasc. Dis. 54 (2011) 86–96. doi:10.1016/j.pcad.2011.03.003.
[6]
J.L. Januzzi, A.S. Maisel, M. Silver, Y. Xue, C. DeFilippi, Natriuretic Peptide Testing for Predicting Adverse Events Following Heart Failure Hospitalization, Congest. Hear. Fail. 18 (2012) S9–S13. doi:10.1111/j.1751-7133.2012.00306.x.
[7]
C.W. Yancy, M. Jessup, B. Bozkurt, J. Butler, D.E. Casey, M.H. Drazner, et al., 2013 ACCF/AHA guideline for the management of heart failure: A report of the american college of cardiology foundation/american heart association task force on practice guidelines, Circulation. 128 (2013) 240–327. doi:10.1161/CIR.0b013e31829e8776.
[8]
G. Volpicelli, M. Elbarbary, M. Blaivas, D.A. Lichtenstein, G. Mathis, A.W. Kirkpatrick, et al., International evidence-based recommendations for point-of-care lung ultrasound, Intensive Care Med. 38 (2012) 577–591. doi:10.1007/s00134-012-2513-4.
[9]
E. Pivetta, A. Goffi, E. Lupia, M. Tizzani, G. Porrino, E. Ferreri, et al., Lung UltrasoundImplemented Diagnosis of Acute Decompensated Heart Failure in the ED: A SIMEU Multicenter Study., Chest. 148 (2015) 202–10. doi:10.1378/chest.14-2608.
AC
CE
PT
ED
MA
NU
SC RI
PT
[1]
[10] D. Lichtenstein, G. Mezière, Relevance of lung ultrasound in the diagnosis of acute respiratory failure the BLUE protocol, Chest. 134 (2008) 117–125. doi:10.1378/chest.072800. [11] G. Volpicelli, V. Caramello, L. Cardinale, A. Mussa, F. Bar, M.F. Frascisco, Bedside ultrasound of the lung for the monitoring of acute decompensated heart failure, Am. J. Emerg. Med. 26 (2008) 585–591. doi:10.1016/j.ajem.2007.09.014. [12] D. Lichtenstein, G. Meziere, P. Biderman, A. Gepner, O. Barre, The comet-tail artefact: an ultrasound sign of alveolar-interstitial syndrome., Am J Respir Crit Care Med. 156 (1997) 1640–1646. doi:10.1164/ajrccm.156.5.96-07096. [13] E. Agricola, T. Bove, M. Oppizzi, G. Marino, A. Zangrillo, A. Margonato, et al., “Ultrasound comet-tail images”: a marker of pulmonary edema: a comparative study with wedge pressure and extravascular lung water., Chest. 127 (2005) 1690–5. doi:10.1378/chest.127.5.1690. [14] F. Frassi, L. Gargani, P. Tesorio, M. Raciti, G. Mottola, E. Picano, Prognostic Value of Extravascular Lung Water Assessed With Ultrasound Lung Comets by Chest Sonography in Patients With Dyspnea and/or Chest Pain, J. Card. Fail. 13 (2007) 830–835. doi:10.1016/j.cardfail.2007.07.003.
ACCEPTED MANUSCRIPT [15] R.M. Lang, L.P. Badano, V. Mor-Avi, J. Afilalo, A. Armstrong, L. Ernande, et al., Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging., Eur. Heart J. Cardiovasc. Imaging. 16 (2015) 233–70. doi:10.1093/ehjci/jev014.
SC RI
PT
[16] G. Volpicelli, A. Mussa, G. Garofalo, L. Cardinale, G. Casoli, F. Perotto, et al., Bedside lung ultrasound in the assessment of alveolar-interstitial syndrome, Am. J. Emerg. Med. 24 (2006) 689–696. doi:10.1016/j.ajem.2006.02.013. [17] L. Gargani, M. Doveri, L. D’Errico, F. Frassi, M.L. Bazzichi, A. Delle Sedie, et al., Ultrasound lung comets in systemic sclerosis: a chest sonography hallmark of pulmonary interstitial fibrosis., Rheumatology (Oxford). 48 (2009) 1382–1387. doi:10.1093/rheumatology/kep263.
MA
NU
[18] M. Trezzi, D. Torzillo, E. Ceriani, G. Costantino, S. Caruso, P.T. Damavandi, et al., Lung ultrasonography for the assessment of rapid extravascular water variation: Evidence from hemodialysis patients, Intern. Emerg. Med. 8 (2013) 409–415. doi:10.1007/s11739-0110625-4. [19] M. Gheorghiade, G. Filippatos, L. De Luca, J. Burnett, Congestion in acute heart failure syndromes: an essential target of evaluation and treatment., Am. J. Med. 119 (2006) S3–S10. doi:10.1016/j.amjmed.2006.09.011.
PT
ED
[20] K.F. Adams, G.C. Fonarow, C.L. Emerman, T.H. LeJemtel, M.R. Costanzo, W.T. Abraham, et al., Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE)., Am. Heart J. 149 (2005) 209–16. doi:10.1016/j.ahj.2004.08.005.
CE
[21] A.P. Ambrosy, P.S. Pang, S. Khan, M.A. Konstam, G.C. Fonarow, B. Traver, et al., Clinical course and predictive value of congestion during hospitalization in patients admitted for worsening signs and symptoms of heart failure with reduced ejection fraction: findings from the EVEREST trial, Eur. Heart J. 34 (2013) 835–843. doi:10.1093/eurheartj/ehs444.
AC
[22] M.H. Miglioranza, L. Gargani, R.T. Sant’Anna, M.M. Rover, V.M. Martins, A. Mantovani, et al., Lung ultrasound for the evaluation of pulmonary congestion in outpatients: A comparison with clinical assessment, natriuretic peptides, and echocardiography, JACC Cardiovasc. Imaging. 6 (2013) 1141–1151. doi:10.1016/j.jcmg.2013.08.004. [23] L. Gargani, P.S. Pang, F. Frassi, M.H. Miglioranza, F.L. Dini, P. Landi, et al., Persistent pulmonary congestion before discharge predicts rehospitalization in heart failure: a lung ultrasound study., Cardiovasc. Ultrasound. 13 (2015) 40. /pmc/articles/PMC4558829/?report=abstract. [24] S. Coiro, P. Rossignol, G. Ambrosio, E. Carluccio, G. Alunni, A. Murrone, et al., Prognostic value of residual pulmonary congestion at discharge assessed by lung ultrasound imaging in heart failure., Eur. J. Heart Fail. 17 (2015) 1172–81. doi:10.1002/ejhf.344. [25] F. Ciccarese, A.M. Chiesa, F. Feletti, L. Vizioli, M. Pasquali, P. Forti, et al., The Senile Lung as a Possible Source of Pitfalls on Chest Ultrasonography and Computed Tomography, Respiration. 90 (2015) 56–62. http://www.ncbi.nlm.nih.gov/pubmed/26044398. [26] W.L. Miller, B.P. Mullan, Volume Overload Profiles in Patients With Preserved (HFpEF) and Reduced (HFrEF) Ejection Fraction Chronic Heart Failure, JACC Hear. Fail. (2016). http://www.ncbi.nlm.nih.gov/pubmed/26970830.
ACCEPTED MANUSCRIPT [27] L. Gargani, Prognosis in heart failure: look at the lungs, Eur. J. Heart Fail. 17 (2015) 1086– 1088. http://www.ncbi.nlm.nih.gov/pubmed/26470753.
AC
CE
PT
ED
MA
NU
SC RI
PT
[28] S. Coiro, T. Chouihed, N. Girerd, Lung ultrasound - The extension of clinical examination in patients with acute heart failure: Reply., Eur. J. Heart Fail. 18 (2016) 215. http://www.ncbi.nlm.nih.gov/pubmed/26726954.
ACCEPTED MANUSCRIPT Figure legends Figure 1:
PT
Example of two different US scans in the same patient: in the upper panel, no B-lines are detectable
AC
CE
PT
ED
MA
NU
SC RI
while multiple and confluent B-lines are clearly seen in the lower panel.
ACCEPTED MANUSCRIPT Tables
AC
CE
PT
ED
MA
NU
SC RI
Age years 80.7 (75.9 – 84.7)* Sex F % (n) 57.7% (86) NYHA Class % (n) I 14.9% (22) II 58.8% (87) III 23.0% (34) IV 3.4% (5) Etiology of current HF hospitalization % (n) Coronary heart disease 41.6% (62) Hypertensive Heart Disease 20.1% (30) Valvular Heart Disease 32.9% (49) Hypertrophic cardiomyopathy 2.0% (3) Primitive dilated cardiomyopathy 6.7% (10) Toxic Cardiomyopathy 0.7% (1) Metabolic Cardiomyopathy 0.7% (1) Comorbidities % (n) Hypertension 62.4% (93) Atrial fibrillation 20.8% (31) Chronic obstructive pulmonary 16.1% (24) disease Chronic kidney disease 24.2% (36) Diabetes Mellitus 34.2% (51) Vitals and Chemistry Heart rate bpm 72.0 (10.5)§ Systolic blood pressure mmHg 121.0 (15.9)§ diastolic blood pressure mmHg 68.9 (10.4)§ Oxigen saturation % 96.1 (2.1)§ Hemoglobin g/dl 11.6 (1.7)§ Na+ mEq/l 139.2 (3.1)§ GFR ml/min 55.1 (42.6-76.1)* NT-proBNP pg/ml 2407 (1032-5273) * NT-proBNP categorical %(n) <1500 pg/ml 37.4% (37) 1500-5000 pg/ml 35.3% (35) >5000 pg/ml 27.3% (27) Drugs % (n) β-blockers 65.5% (97) ACE-i/ARBs 68.9% (102) Furosemide 95.7% (141) Aldosterone receptor antagonists 38.5% (57) Digoxin 23.6% (35) Antiarrythmic agents 12.2% (18) Pace Maker 18.8% (28) Implantable cardioverter-defibrillator 6.0% (9) Cardiac resynchronization therapy 5.4% (8) Echocardiography EF % 47.7 (16.9)§
PT
Table 1: Baseline population characteristic
ACCEPTED MANUSCRIPT
MA
NU
SC RI
PT
EF<40 % (n) 36.2% (54) 2 LVEDVi ml/m 66.6 (28.7)§ LVESVi ml/m2 38.5 (27.0)§ LA vol i ml/m2 68.4 (27.7)§ 2 RA vol i ml/m 35.5 (20.4)§ E/A 1.8 (1.3)§ E/E’ # 14.0 (7.2)§ E/E’ >15 % (n) # 41.6% (49) RVd mm 34.5 (7.5)§ TAPSE mm 16.5 (4.8)§ RV-RA gr mmHg 32.4 (11.2) § IVC-CI % 49.5 (28.1)§ Lung Ultrasound Positive sectors 2 (0-4)* Bilaterally ≥1 40.3% (60) Bilaterally ≥2 24.2% (36) * data reported as median (IQR) § data reported as mean (SD) # E/E’ not feasible in 31 patients (anular calcifications, prosthetic valves).
ED
Abbreviations
SD:standard deviation
PT
IQR: interquartile range
CE
GFR: Glomerular filtration rate
ACE-I: angiotensin coverting enzyme inhibitors
AC
ARBs: angiotensin receptor blockers; EF: ejection fraction
LVESV: left ventricular end-systolic volume (indexed) LVEDVi: left ventricular end-diastolic volume (indexed) LA vol i: left atrium volume (indexed) RA vol i: right atrium volume (indexed) RVd: right ventricle diameter RV-RA gr: right ventricular-right atrial gradient IVC-CI: inferior vena cava collapsibility index
ACCEPTED MANUSCRIPT Table 2: Prediction of readmission for HF and death from any cause at 100 days. Results of univariate and multivariate Cox analysis.
PT
Multivariate HR (95% CI) -
p value
SC RI
p value 0.116 0.009
0.034 0.028 0.692 0.502 0.0953
NU
1 2.44 (1.24-4.82) 0.98 2.14 1.04 (0.85 to 1.26) 1.05 (0.89 to 1.25)
MA
1 1.79 (0.90 to 3.55)
ED
1 1.01 (0.49 to 2.08) 0.98 (0.97 to 1.00)
PT
1 2.19 (1.12 to 4.29) 1.03 (0.98 to 1.08) 1 1.35 (0.62 to 2.97) 1.00 (0.99 to 1.00)
AC
CE
Age, years NYHA <3 ≥3 GFR GFR <50ml/min Hemoglobin Oxigen saturation Peripheral edema Absent Present Presence of crackles Absent Present EF EF categorical ≥40% <40% E/E’ E/E’ categorical ≤15 >15 LA vol i Pleural effusion Absent Present IVC-CI IVC-CI categorical ≥50% <50% JV turgescence Absent Present NT-proBNP NT-proBNP categorical <1500 pg/ml 1500-5000 pg/ml >5000 pg/ml Sonographic score
Univariate HR (95% CI) 1.04 (0.99 to 1.09)
Positive sectors bilaterally ≥1 ≥2
1 1.78 (0.91 to 3.49) 0.99 (0.97 to 1.00)
0.978
0.148 0.022
-
0.210 0.449
0.556 0.0925
0.133 0.540
1 1.24 (0.61 to 2.52) 0.360 1 1.51 (0.63 to 3.65) 1.18 (1.07 to 1.31) 1 3.92 (0.81 to 18.88) 9.95 (2.22 to 44.53) 1.19 (1.05 to 1.34)
3.62 (1.76 to 7.43) 2.27 (1.12 to 4.59)
0.001 0.005
0.005 0.022 0.0005 0.022
1 0.005 3.92 (0.81 to 18.88) 9.95 (2.22 to 44.53) -
ACCEPTED MANUSCRIPT 1.005 (1.002 to 1.008)
Total B-lines B-lines categorical ≤15 >15 B-lines categorical ≤30 >30
0.003 0.019
1 3.10 (1.20 to 8.02) 0.005
NU
SC RI
PT
1 2.85 (1.36 to 5.96)
Abbreviations
GFR: Glomerular filtration rate
ED
LA vol i: left atrium volume (indexed)
MA
CI: confidence interval
IVC-CI: inferior vena cava collapsibility index
AC
CE
PT
JV: jugular vein
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI
PT
Figure 1