Outcomes of Thoracentesis for Acute Heart Failure in Hospitals

Abstract Data on in-hospital outcomes for hospitalizations undergoing thoracentesis (THR) for any cause has been conflicting. For hospitalizations with acute heart failure (HF), however, to date, no study has evaluated the outcomes of THR. Accordingly, our current study addresses this knowledge gap. We analyzed data from the Nationwide Inpatient Sample (2005-14). The study population included all adults (>18 years) with the principal discharge diagnosis of HF and the presence of procedure code for THR. Hospitalizations with pneumonia, acute kidney injury and co-morbidities such as malignancy, lymphoma, liver disease, end-stage renal disease, metastatic disease and tuberculosis were excluded. Propensity matching was performed to identify a similar cohort of admissions that did not undergo THR. Primary outcome of interest was in-hospital mortality and length of hospitalization. During the study period, 2,251,927 hospitalizations for HF were found from the database; of which, 70,823 (3.14%) had THR. After propensity matching, a matched cohort of 70,785 hospitalizations for HF was identified. In-hospital mortality was higher for those who underwent THR (2.5% vs 1.6%; P