Serial lung and IVC ultrasound in the assessment of congestive heart failure
© The Author(s) 2017
Received: 8 December 2016
Accepted: 1 March 2017
Published: 7 March 2017
Management of congestive heart failure (CHF) is dependent on clinical assessments of volume status, which are subjective and imprecise. Point-of-care ultrasound (POCUS) is useful in the diagnosis of CHF, but how POCUS findings correlate with therapy remains unknown. This study aimed to determine whether the changes in clinical evaluation of CHF with treatment are mirrored with changes in the number of B lines on lung ultrasound (LUS) and inferior vena cava (IVC) size. In this prospective observational study, investigators performed serial clinical and ultrasound assessments within 24 h of admission (T1), day 1 in hospital (T2) and within 24 h of discharge (T3). Clinical assessments included an evaluation of the jugular venous distension (JVD), hepatojugular reflux (HJR), pulmonary rales and a clinical congestion score was calculated. Ultrasound assessment included the IVC size and collapsibility, and the number of B lines in an 8-point scan.
Fifty consecutive patients were recruited with a mean age of 71.2 years (SD 12.7). Mean clinical congestion score on admission was 5.6 (SD 1.4) and declined significantly over time to 1.3 (0.91), as did the JVP, HJR and pulmonary rales. No significant changes were found in the IVC size between T1 [1.9 (0.65)] and T3 [2.0 (0.50)] or in the IVC collapsibility index [T1 0.3 (0.19) versus T3 0.25 (0.16)]. The mean number of B lines decreased from 11 (6.1) at T1 to 8.3 (5.5) at T3, although this decrease did not reach statistical significance. Spearman correlation between JVP and HJR versus IVC collapsibility and total B lines did not yield significant results.
Clinical exam findings correlate over time during the management of CHF, whereas LUS and IVC results did not. The number of B lines did decrease with therapy, but did not reach statistical significance likely because the sampled population was small and had only mild heart failure. Further studies are warranted to further explore the use of lung ultrasound in this patient population.
KeywordsVascular ultrasound Lung ultrasound Inferior vena cava B lines Heart failure
Cardiovascular disease is the leading cause of mortality, representing 29% of all deaths, and costs the healthcare system 20.9 billion dollars annually in North America . Congestive heart failure (CHF) is a complex clinical syndrome characterized by recurrent episodes of acute decompensation, requiring frequent hospitalizations and readmissions . The diagnosis and management of CHF remain a clinical challenge; decisions are based on clinical assessments of volume status, which are used to estimate cardiac filling pressures . Chest X-ray (CXR) has traditionally been used to suggest pulmonary congestion, but assessments can be subjective  and lack sensitivity and specificity . Recently, serial assessments of brain natriuretic peptide (BNP) have been shown to correlate with the degree of hemodynamic congestion  and may be a reliable tool to guide management in heart failure patients .
The gold standard for determining cardiac congestion is cardiac catheterization with determination of right atrial and left atrial pressures, as estimated by the pulmonary capillary wedge pressures. These tests are invasive, time consuming and carry risk, and are not typically performed in patients admitted with decompensated heart failure . As a result, physicians rely on clinical assessments to diagnose elevated cardiac filling pressures. Clinical evaluation, however, correlates poorly with invasive hemodynamics  and the physical exam lacks sensitivity and specificity for the diagnosis of heart failure, limiting its clinical utility .
Point of care ultrasound (POCUS) has been developed as an easy-to-learn, rapid, portable, and non-invasive adjunct to physical examination. Previous studies have shown that inferior vena cava size correlates with right atrial pressure and pulmonary capillary wedge pressure with good sensitivity and specificity [11, 12], and outperforms the physical exam in the detection of an elevated jugular venous pressure . However, IVC as a surrogate for volume status has been increasingly questioned, as the correlation to CVP has been variable . However, the American Society of Echo guidelines still recommend its use as a non-invasive measurement of CVP .
Similarly, lung ultrasound and B lines, an artifact representing extravascular lung water, similarly correlate significantly with pulmonary artery systolic pressure  and outperform classical signs and symptoms of congestion . Both lung [18, 19] and vascular ultrasound [20–22] have been shown to be helpful in the assessment of undifferentiated dyspnea and in the diagnosis of heart failure. It is not clear how POCUS findings change with treatment.
The primary objective of the current study is to compare serial lung and vascular ultrasound exam with clinical findings upon CHF treatment. The secondary objective was to explore the relationship between the severity of ultrasound-measured degree of congestion and length of hospital stay, hospital readmission and mortality.
Baseline patient demographics, including age, weight, sex, medical comorbidities, and etiology of heart failure were recorded from chart review. Laboratory values including hemoglobin, renal function and troponin values on admission were recorded from a chart review. Left ventricular ejection fraction (LVEF), diastolic function and valvulopathy were obtained from the most recent echocardiogram or during the current hospitalization. Chest X-ray (CXR) on admission was reviewed and graded by resident authors after they performed the clinical and ultrasound assessment as having interstitial edema or alveolar edema, where alveolar edema indicates more significant pulmonary congestion. Three months following discharge, a chart review was performed to assess for length of hospital stay, hospital readmissions and 90-day mortality. The hospital ethics committee approved the current study.
The Killip and clinical congestion score (CCS) were calculated for each patient at across the three time points. The Killip score (0–4)  was calculated to grade the severity of heart failure, with 0 indicating no heart failure and 4 indicating cardiogenic shock. Although this score has been designed specifically in the setting of acute myocardial infarction, it was used in this study as it is a widely known descriptor of the severity of heart failure.
The clinical congestion score (CCS)  was calculated based on pulmonary rales (scored 0–4), central venous pressure elevation (scored 0–4), S3 (presence 1, absence 0) and orthopnea (scored 0–4). The CCS has a maximum score of 13, where higher scores indicate more severe heart failure. The CCS was validated on an outpatient population with stable heart failure, and has been shown to correlate with echocardiographic findings of increased filling pressures .
Specifically, the following symptoms and physical exam findings were graded for the above scores as follows: the presence of dyspnea and PND was graded as present (1) or absent (0). Orthopnea was graded between 0 and 4, with a score of 0 indicating the need of one pillow during sleep, and 4 indicating at least one night spent sleeping in a sitting position. Central venous pressure elevation was graded as 0 if the crests of neither the internal nor the external jugular vein are visible above the clavicle, and 4 if the crests are visible at the earlobe with the patient assessed at 30–45°. The hepatojugular reflux was recorded as present (1) or absent (0). Rales/crackles were graded as being absent (0), present in less than 50% of the lung field (1), or more than 50% of the lung field (2). Peripheral edema was graded between 0 and 4 according to the depth of indentation at the ankle.
The POCUS ultrasound assessment was performed with the Zonare zone system (Zonare Medical Systems Inc, Mountain View, Ca, USA). Two authors (RS and MS) collected the ultrasound data after a 3-h training session and 10 supervised scans with an intensive care specialist with advanced cardiac echo and formal training in lung ultrasound prior to enrollment of patients. Brennan et al.  have shown that internal medicine residents with minimal training, including 4 h of formal didactic ultrasound teaching, and 20 sonographer supervised acquisitions were able to obtain adequate images of the IVC in 90% of patients. Each patient underwent serial scans at the time of the clinical assessments. Two authors (MS and JL) blinded to clinical information analyzed the stored images. A priori it was established that findings on vascular and LUS were not disclosed to the treating team unless they found a significant unexpected life-threatening diagnoses, such as venous thrombus, complex pleural effusion suggestive of empyema, pneumothorax or cardiac tamponade. As a result, no treatment decisions regarding the patient’s heart failure were made based on the ultrasound findings.
The IVC was analyzed in the long axis via the subcostal window using a 4-MHz phased-array ultrasound transducer. The diameter of the intra-hepatic IVC within 1–2 cm of the right atrium and the response of the IVC to an inspiratory maneuvre was measured and recorded. The size of the IVC on expiration was recorded, and the IVC collapsibility index was calculated as (IVC size on expiration − IVC size on inspiration)/IVC size on expiration.
All lung ultrasounds were performed using a 6-MHz curved array probe and a 3-s clip was recorded in each quadrant. A bilateral 8-point LUS examination was performed to assess for B lines following the International consensus guidelines . An oblique intercostal view was obtained to maximize the number of B lines, and the actual count was made at the moment the most B lines were visualized.
Clinical and ultrasound variables were analyzed using a paired t test. Pearson correlations were used to compare clinical and ultrasound variables. The impact of ultrasound and clinical assessments on hospital LOS and 90-day mortality was evaluated using univariate and multivariate analyses.
Baseline clinical characteristics at time of recruitment
Sex (% male)
Diastolic dysfunction (%)
Interstitial edema CXR (%)
Pulmonary edema CXR (%)
Mean or %
Pearson correlations between JVP and HJR showed significant correlations at T1 (0.58, p < 0.0001) and T2 (0.55, p = 0.0003). However, correlations between JVP or HJR and IVC size or collapsibility did not reach significance. Similarly, pulmonary rales and number of B lines across all time points did not show significant correlations. Univariate and multivariate analysis did not show any clinical or ultrasound variables that correlate with hospital length of stay and 90-day mortality rate.
Ninety-eight percent of the IVC scans were interpretable and read; 1 out of 50 could not be read because of poor image quality. All LUS scans were interpretable and read offline. The quality of the scans was determined by expert opinion, an Intensive care specialist with advanced cardiac echo and formal training in lung ultrasound. There were no significant differences in the ultrasound variables between the two authors. Twenty-eight percent of patients did not complete all three time points.
The present study showed no significant change in IVC size or collapsibility index despite significant improvement in clinical signs of congestion in patients treated for acute decompensated heart failure. The results did show a trend toward significance for number of B lines; there was a 38% decrease in the number of B lines with the treatment of heart failure.
There are several explanations for the current data. This study enrolled patients upon admission to the cardiac care unit, which occurred between hours and days after their presentation to the emergency department. Consequently, treatment for heart failure was already initiated in the emergency department prior to our first ultrasound assessment. Unfortunately, our hospital database does not track delay to admission for these patients in particular. With this delay, the degree of hemodynamic congestion was likely diminished, and any appreciable changes in IVC size and B lines subsequent to the initial findings were too small to detect. This may indicate that IVC size and B lines clear rapidly with treatment and the sensitivity of this technique to detect such changes is on the level of hours in the acute setting. Additionally, the patients in the current study had a median Killip score of 2, which indicates that this patient population had only mild heart failure.
In contrast to our results, Goonewardena et al.  demonstrated significant reduction in IVC size and increases in collapsibility index with heart failure treatment. This study found larger IVCs at admission (2.3 versus 1.9 cm in the current study). However, IVC size and collapsibility indices at discharge were identical between the two studies (2 cm in both studies). This may suggest that the greatest change in IVC size occurs early following initial therapy, and likely why the current study did not reach significance for this primary endpoint. However, IVC size as a surrogate for volume status has increasingly been questioned. Another explanation for the current data is that IVC size may not accurately represent hypervolemic states, and thus may not be useful in the management of decompensated heart failure.
Volpicelli et al.  conducted a study where number of B lines was calculated at admission and discharge, and found significant clearing of B lines after treatment. However, this population had more severe heart failure than the current study. This study used an arbitrary clinical scale out of ten points to grade the clinical severity of heart failure which included the presence (1) or absence (0) of lower extremity edema, pulmonary rales and wheezing, jugular venous distension, orthopnea, high respiratory rates (>25 breaths/min) and low pulse oximeter saturation; median score was 8/10 on admission whereas in the current study, the mean CCS was 5.6 out of 13. In another study, Gargani et al.  showed a significant reduction in B lines with treatment from 48 on admission to 20 at discharge, compared to the current study which showed a non-significant reduction of B lines from 11 to 8.3. Both these studies found a larger number of B lines on admission and sampled patients with more severe heart failure. This may indicate that LUS is a useful indicator of improvement in pulmonary congestion in patients with severe heart failure, but may fail to identify significant changes in pulmonary congestion in patients with milder CHF.
Of note, Volpicelli  analyzed the B lines using a different method. Eleven regions were scanned, compared to eight regions in the current study. Additionally, a positive region was defined as number of B lines over 3, and they analyzed the data based on the number of positive regions as opposed to absolute number of B lines. However, we did not find a significant change even when our data were analyzed in a similar manner. There was no prospective strategy in the study protocol to quantify confluent B lines, and as a result, they were counted as a single B line. Although there is a suggested method to quantify confluent B lines by calculating the proportion of visualized pleural line/length of the confluent B lines , we felt that there is insufficient evidence to apply this technique at the current time. It is possible that we underestimated the overall quantity of B lines using this approach; however, this method was consistently used across all time points and should not affect the overall results.
There are several limitations in the current study. This was a small study with only 50 patients enrolled at a single center. We did not formally estimate sample size as there were no reliable data to estimate effect size in this population. However, based on our results, a sample size of approximately 150 would be required to detect a statistically significant difference (α 0.05 and β 0.20) for B lines and a sample size of approximately 1000 would be required for change in IVC diameter.
Additionally, there was a high dropout rate with 28% of patients not completing all three time points. The clinical assessments were not blinded; resident authors performing these assessments knew at which time point each patient was being assessed, which could have introduced bias.
As mentioned above, since treatment was initiated early upon presentation and prior to recruitment, the ability to detect a potential decrease in IVC size and number of B lines may have be reduced. The clinical score would be less influenced as resolution of peripheral edema and crackles would lag, and the clinical history of orthopnea and PND is retrospective.
Assessment of volume status is challenging and point of care ultrasound may provide greater sensitivity to detect subclinical congestion before overt clinical congestion occurs. While this may be true for patients with more severe heart failure, the current study suggests that lung and vascular ultrasound in patients with less severe heart failure may not be sensitive enough to detect such small changes in hemodynamic congestion. As a result, serial POCUS may perform less well in patients with mild versus more severe heart failure. Based on the current results, IVC size or collapsibility index is not an accurate marker of improvement in hemodynamic congestion. B lines may be a better marker of congestion as these did decrease with the treatment of heart failure as expected. The lack of statistical significance for the number of B lines may be a result of a small sample size, and a sampled population with less severe heart failure than previously published, and an initial small number of B lines. Another interpretation of our data is that given the significant correlation over time with clinical scores and not with ultrasound, it may suggest that clinical exam is more sensitive than ultrasound at detecting congestive symptoms, although this is hypothesis generating given the small sample size.
The current study is the only study to the authors’ knowledge, which directly compared lung and IVC ultrasound in the dynamic clinical setting of heart failure management after hospital admission. Further studies are warranted to better understand the utility of bedside ultrasound in patients with mild heart failure, and which ultrasound modality provides the most accurate assessment of hemodynamic congestion for both diagnostic and prognostic purposes.
In conclusion, this study shows that vascular and lung ultrasound does not show significant changes with the management of heart failure in a hospitalized population with acute decompensated heart failure, although LUS shows a trend toward significance. This may indicate that vascular ultrasound parameters are less robust in patients with mild congestion and may show the greatest change during initial management.
Further larger scale studies are needed to explore how POCUS may be useful in the acute and sub-acute management of heart failure.
congestive heart failure
inferior vena cava
jugular venous pressure
clinical congestion score
Literature review, study design and implementation were performed by RS and JL. MAS and RS performed the clinical and ultrasound bedside assessments. MAS and JL performed the offline analysis of the ultrasound data. DJ performed all statistical analyses and assisted with the editing of the manuscript. JD and LR contributed to the study design and editing of the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
All above authors have no declarations.
Ethics approval and consent to participate
This study received ethics approval by the Research Ethics Office of the Jewish General Hospital. Inclusion criteria to enroll in the study required patients to sign a consent form, which was approved by the Research Ethics Office. The number assigned to this project is 14-052.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
- Writing Group Members, Lloyd-Jones D, Adams RJ et al (2010) Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 121(7):e46–e215. doi:10.1161/CIRCULATIONAHA.109.192667 View ArticleGoogle Scholar
- Jencks SF, Williams MV, Coleman EA (2009) Rehospitalizations among patients in the medicare fee-for-service program. N Engl J Med 360(14):1418–1428. doi:10.1056/NEJMsa0803563 View ArticlePubMedGoogle Scholar
- Gheorghiade M, Follath F, Ponikowski P et al (2010) Assessing and grading congestion in acute heart failure: a scientific statement from the acute heart failure committee of the heart failure association of the European Society of Cardiology and endorsed by the European Society of Intensive Care Medicine. Eur J Heart Fail 12(5):423–433. doi:10.1093/eurjhf/hfq045 View ArticlePubMedGoogle Scholar
- Mayhadoon H, Klein R, Eyler W et al (1989) Radiographic pulmonary congestion in end-stage congestive heart failure. Am J Cardiol 63:625–627View ArticleGoogle Scholar
- Mant J, Doust J, Roalfe A et al (2009) Systematic review and individual patient data meta-analysis of diagnosis of heart failure, with modelling of implications of different diagnostic strategies in primary care. Health Technol Assess 13(32):1–248. doi:10.3310/hta13320 View ArticlePubMedGoogle Scholar
- Md SNG, Md JEAB, Md AM et al (2010) Use of hand carried ultrasound, B-type natriuretic peptide, and clinical assessment in identifying abnormal left ventricular filling pressures in patients referred for right heart catheterization. J Card Fail 16(1):69–75. doi:10.1016/j.cardfail.2009.08.004 View ArticleGoogle Scholar
- Gackowski A, Isnard R, Golmard J et al (2004) Comparison of echocardiography and plasma B-type natriuretic peptide for monitoring the response to treatment in acute heart failure. Eur Heart J 25(20):1788–1796. doi:10.1016/j.ehj.2004.07.038 View ArticlePubMedGoogle Scholar
- International Liaison Committee on Lung Ultrasound (ILC-LUS) for the International Consensus Conference on Lung Ultrasound (ICC-LUS), Volpicelli G, Elbarbary M et al (2012) International evidence-based recommendations for point-of-care lung ultrasound. Intensive Care Med 38(4):577–591. doi:10.1007/s00134-012-2513-4 View ArticleGoogle Scholar
- Stevenson LW, Perloff JK (1989) The limited reliability of physical signs for estimating hemodynamics in chronic heart failure. JAMA 261:884–888View ArticlePubMedGoogle Scholar
- Cook DJ, Simel DL (1996) Does this patient have abnormal central venous pressure? JAMA 275(8):630–634. doi:10.1001/jama.1996.03530320054034 View ArticlePubMedGoogle Scholar
- Brennan JM, Blair JE, Goonewardena S et al (2007) Reappraisal of the use of inferior vena cava for estimating right atrial pressure. J Am Soc Echocardiogr 20(7):857–861. doi:10.1016/j.echo.2007.01.005 View ArticlePubMedGoogle Scholar
- Blair JE, Brennan JM, Goonewardena SN, Shah D, Vasaiwala S, Spencer KT (2009) Usefulness of hand-carried ultrasound to predict elevated left ventricular filling pressure. Am J Cardiol 103(2):246–247. doi:10.1016/j.amjcard.2008.08.061 View ArticlePubMedGoogle Scholar
- Goonewardena SN, Spencer KT (2010) Handcarried echocardiography to assess hemodynamics in acute decompensated heart failure. Curr Heart Fail Rep 7(4):219–227. doi:10.1007/s11897-010-0030-8 View ArticlePubMedGoogle Scholar
- Alavi-Moghaddam M, Kabir A, Shojaee M et al (2016) Ultrasonography of inferior vena cava to determine central venous pressure: a meta analysis and meta regression. Acta Radiol 18:1–5Google Scholar
- Rudski LG, Wyman WL, Afilalo J et al (2010) Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 23:685–713View ArticlePubMedGoogle Scholar
- Platz E, Lattanzi A, Agbo C et al (2012) Utility of lung ultrasound in predicting pulmonary and cardiac pressures. Eur J Heart Fail 14(11):1276–1284. doi:10.1093/eurjhf/hfs144 View ArticlePubMedGoogle Scholar
- Anderson KLI, Jenq KY, Panebianco NL, Dean AJ (2013) Diagnosing heart failure among acutely dyspneic patients with cardiac, inferior vena cava and lung ultrasonography. Am J Emerg Med 31(8):1208–1214View ArticlePubMedGoogle Scholar
- Vitturi N, Soattin M, Allemand E, Simoni F, Realdi G (2011) Thoracic ultrasonography: a new method for the work-up of patients with dyspnea. J Ultrasound 14(3):147–151. doi:10.1016/j.jus.2011.06.009 View ArticlePubMedPubMed CentralGoogle Scholar
- Miglioranza MH, Gargani L, Sant’Anna RT et al (2013) Lung ultrasound for the evaluation of pulmonary congestion in outpatients. JACC Cardiovasc Imaging 6(11):1141–1151. doi:10.1016/j.jcmg.2013.08.004 View ArticlePubMedGoogle Scholar
- Miller JB, Sen A, Strote SR et al (2012) Inferior vena cava assessment in the bedside diagnosis of acute heart failure. Am J Emerg Med 30(5):778–783. doi:10.1016/j.ajem.2011.04.008 View ArticlePubMedGoogle Scholar
- Pellicori P, Carubelli V, Zhang J et al (2013) IVC diameter in patients with chronic heart failure. JACC Cardiovasc Imaging 6(1):16–28. doi:10.1016/j.jcmg.2012.08.012 View ArticlePubMedGoogle Scholar
- Md DJB, Md ED, Md RG (2009) Identification of congestive heart failure via respiratory variation of inferior vena cava diameter. Am J Emerg Med 27(1):71–75. doi:10.1016/j.ajem.2008.01.002 View ArticleGoogle Scholar
- Killip T, Kimball JT (1967) Treatment of myocardial infarction in a coronary care unit: a 2 year experience with 250 patients. Am J Cardiol 20(4):457–464View ArticlePubMedGoogle Scholar
- Rhode LE, da-Silva Beck L, Goldraich L, Grazziotin TC, Palombini DV, Polanczyk CA, Claussell N (2004) Reliability and prognostic value of traditional signs and symptoms in outpatients with congestive heart failure. Can J Cardiol 20(7):697–702Google Scholar
- Brennan JM, Blair JE, Goonewardena S, Ronan A, Shah D, Vasaiwala S et al (2007) A comparison by medicine residents of physical examination versus hand-carried ultrasound for estimation of right atrial pressure. Am J Cardiol 99(11):1614–1616. doi:10.1016/j.amjcard.2007.01.037 View ArticlePubMedGoogle Scholar
- Goonewardena SN, Gemignani A, Ronan A et al (2008) Comparison of hand-carried ultrasound assessment of the inferior vena cava and N-terminal pro-brain natriuretic peptide for predicting readmission after hospitalization for acute decompensated heart failure. JACC Cardiovasc Imaging 1(5):595–601. doi:10.1016/j.jcmg.2008.06.005 View ArticlePubMedGoogle Scholar
- Gargani L, Pang PS, Frassi F et al (2015) Persistent pulmonary congestion before discharge predicts rehospitalization in heart failure: a lung ultrasound study. Cardiovasc Ultrasound 13:1–9. doi:10.1186/s12947-015-0033-4 View ArticleGoogle Scholar
- Volpicelli G, Caramello V, Cardinale L, Mussa A, Bar F, Frascisco MF (2008) Bedside ultrasound of the lung for the monitoring of acute decompensated heart failure. Am J Emerg Med 26(5):585–591. doi:10.1016/j.ajem.2007.09.014 View ArticlePubMedGoogle Scholar
- Anderson KL, Fields JM, Panebianco NL, Jenq KY, Marin J, Dean AJJ (2013) Inter-rater reliability of quantifying pleural B-lines using multiple counting methods. Ultrasound Med 32(1):115–120View ArticleGoogle Scholar