Chest computed tomography features of heart failure: A prospective observational study in patients with acute dyspnea

Abstract

BACKGROUND: Pulmonary congestion is a key component of heart failure (HF) that chest computed tomography (CT) can detect. However, no guideline describes which of many anticipated CT signs are most associated with HF in patients with undifferentiated dyspnea.
METHODS: In a prospective observational single-center study, we included consecutive patients ≥ 50 years admitted with acute dyspnea to the emergency department. Patients underwent immediate clinical examination, blood sampling, echocardiography, and CT. Two radiologists independently evaluated all images. Acute HF (AHF) was adjudicated by an expert panel blinded to radiology images. LASSO and logistic regression identified the independent CT signs of AHF.
RESULTS: Among 232 patients, 102 (44%) had AHF. Of 18 examined CT signs, 5 were associated with AHF (multivariate odds ratio, 95% confidence interval): enlarged heart (20.38, 6.86–76.16), bilateral interlobular thickening (11.67, 1.78–230.99), bilateral pleural effusion (6.39, 1.98–22.85), and increased vascular diameter (4.49, 1.08–33.92). Bilateral ground-glass opacification (2.07, 0.95–4.52) was a consistent fifth essential sign, although it was only significant in univariate analysis. Eighty-eight (38%) patients had none of the five CT signs corresponding to a 68% specificity and 86% sensitivity for AHF, while two or more of the five CT signs occurred in 68 (29%) patients, corresponding to 97% specificity and 67% sensitivity. A weighted score based on these five CT signs had an 0.88 area under the curve to detect AHF.
CONCLUSIONS: Five CT signs seem sufficient to assess the risk of AHF in the acute setting. The absence of these signs indicates a low probability, one sign makes AHF highly probable, and two or more CT signs mean almost certain AHF.