The Impact of Clinically Significant Pleural Effusion on Survival of US Veterans With Cancer, Congestive Heart Failure, and Pneumonia: The Veterans Administration Lung Effusion Study.

Abstract

Background: Previous clinical data suggest that the presence of a pleural effusion is associated with poor survival. However, these studies were limited by either a small sample size or lack of an adequate control group.

Research questions: What is the impact of pleural effusion on survival in patients hospitalized with an admitting diagnosis of the 3 most common causes of pleural effusion: cancer, congestive heart failure, or pneumonia?

Study design and methods: This is a retrospective analysis of US veterans hospitalized between January 1, 2000 and December 31, 2020. International classification of disease codes were used to identify patients with an admitting diagnosis of congestive heart failure (CHF), pneumonia, or cancer. Patients were dichotomized as having a clinically significant pleural effusion (PE) when a pleural effusion drainage was performed or not (NO-PE). All-cause mortality was compared between the PE and NO-PE cohorts.

Results: We analyzed 34,707 patients in the PE group and 792,217 patients in the NO-PE group. Patients with PE had a significantly higher all-cause mortality compared with patients with NO-PE. The median survival time was significantly lower in PE group as compared with NO-PE group across all 3 diagnoses, CHF (PE, 1.51 years; 95% CI, 1.40-1.61 vs NO-PE, 3.23 years; 95% CI, 3.21-3.26), cancer (PE, 1.33 years; 95% CI, 1.27-1.39 vs NO-PE, 2.05 years; 95% CI, 2.02-2.08), and pneumonia (PE, 4.27 years; 95% CI, 3.94-4.61 vs NO-PE, 5.11 years; 95% CI, 5.06-5.15). The hazard ratio of all-cause mortality remained unchanged after adjusting for demographics and comorbidities.

Interpretation: The presence of a clinically significant pleural effusion was independently associated with higher all-cause mortality in patients with admitting diagnosis of CHF, cancer, and pneumonia. Clinicians and researchers should consider the association of CHF, cancer, and pneumonia with Pes when estimating the prognosis of individual patients and when assessing the survival of longitudinal cohorts.