Diagnostic Accuracy of Point-of-Care Lung Ultrasonography and Chest Radiography in Adults With Symptoms Suggestive of Acute Decompensated Heart Failure: A Systematic Review and Meta-analysis.

Maw AM, Hassanin A, Ho PM, et al.

JAMA Netw Open. 2019 Mar 1;2(3):e190703. [Full Text]

Congestive heart failure (CHF) is a clinical syndrome with multiple etiologies. The history is to suspect CHF in a dyspneic patient but classic physical exam findings for CHF are more specific than sensitive [1].

Sensitivity (%)Specificity (%)
LE edema5078
Adapted from Wang CS et al. JAMA Rational Clinical Exam [1]

A chest x-ray (CXR) is routinely obtained in patients with dyspnea and it too is more specific than sensitive (congestion 54% sensitive, 96% specific; interstitial edema 34% sensitive, 97% specific) [1].

These limitation in diagnostic accuracy have led to searches for alternative diagnostic modalities and lung point of care ultrasound (LUS) has been studied in multiple small trials for this purpose. Maw et al. completed a systematic review and meta-analysis (SR/MA) of 6 such studies comparing LUS to CXR for the diagnosis of acute CHF.

Study Extraction and Practice Location

The SM/MA was conducted in compliance with the recommendations from the Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy [2] and reported using the PRISMA reporting guideline. Per those guidelines the search strategy used was reported in order to ensure reproducibility.

The search identified 1,377 titles that underwent screening, with 43 (3.1%) undergoing full-text review. After application of the exclusion criteria, 6 studies were identified representing a total of 1,827 patients. Four studies were conducted in emergency department cohorts and 2 studies in internal medicine ward patients, all with initially complaint of dyspnea.


Four of the studies utilized postero-anterior (PA) CXRs exclusively, and the other 2 used “mostly” PA CXRs. Within the studies,

The number of LUS operators ranged from 1 to 12. The experience of the operators was not always clear or defined. Many were noted as “experts” or “skilled”, some as novices, though in one study no comment on experience was made.

Three studies reported the Cohen’s kappa for interrater agreement (κ), and this ranged from κ 0.70 (moderate agreement) to 1.00 (almost perfect agreement). All studies based a positive LUS on established criteria from the international LUS guideline [3].

Sensitivity for LUS ranged from 58-97% and specificity ranged 69-94%. For CXR, sensitivity ranged 70-90% and specificity 61-98%. Pooled estimates for LUS, were sensitivity 0.88 (95% Cl, 0.75-0.95) and specificity 0.90 (95% Cl, 0.88-0.92) compared to CXR with pooled sensitivity 0.73 (95% Cl, 0.70-0.76) and specificity 0.90 (95% Cl, 0.75-0.97). When comparing relative performance, the relative sensitivity ratio of LUS, compared with CXR, was 1.2 (95% CI, 1.08-1.34; P < .001) but there was no difference in specificity between the two (relative specificity ratio, 1.0; 95% CI, 0.90-1.11; P = .96).


Based on this SR/MA of 6 studies, LUS is as specific as CXR, but is more sensitive. This potentially addresses the concern that exam and CXR findings tend to have low sensitivity, or better help rule out CHF. As the study authors note, “for every 100 patients presenting with dyspnea owing to cardiogenic pulmonary edema, LUS can diagnose 15 more cases than CXR without an increase in the number of false-positives”. It should be noted the sensitivity ranged as low as 58% in an individual study.

Other caveats include that results are based on a small number of studies and LUS was performed mostly by experienced providers. The few studies that reported agreement also noted variable strength of agreement (moderate to near perfect) which should emphasize that something as seemingly objective as LUS has an subjective and technical component that affects the results.

Regardless, these results are promising for increasing the diagnostic accuracy of CHF. LUS is non-invasive and is a non-ionizing imaging modality. The learning curve for LUS is also relatively low compared to other systems (i.e. cardiac). No single finding should be used for making a CHF diagnosis by LUS can be an additional tool to the clinician at the bedside.

  1. Wang CS, FitzGerald JM, Schulzer M, Mak E, Ayas NT. Does this dyspneic patient in the emergency department have congestive heart failure? JAMA. 2005 Oct 19;294(15):1944-56. doi: 10.1001/jama.294.15.1944.
  2. Deeks  JJBP, Gatsonis  C. Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy, Version 1.0.0. London, UK: The Cochrane Collaboration; 2013.
  3. Volpicelli G, Elbarbary M, Blaivas M, et al.; International Liaison Committee on Lung Ultrasound (ILC-LUS) for International Consensus Conference on Lung Ultrasound (ICC-LUS). International evidence-based recommendations for point-of-care lung ultrasound. Intensive Care Med. 2012 Apr;38(4):577-91.