Is pericardial effusion a negative prognostic marker? Meta-analysis of outcomes of pericardial effusion

Background The prognostic relevance and the prevalence of pericardial effusion in several diseases are not well established. The aim of this meta-analysis is to summarize the available evidence related to pericardial effusion prevalence and outcomes according to the cause. Methods Articles investigating the prognosis of pericardial effusion were identified by literature search. Twenty-three studies were finally included (17 022 patients). All-cause mortality was the primary end-point. Secondary end-point was the prevalence of pericardial effusion in most common diseases related to this clinical condition. Results The pooled prevalence of pericardial effusion was 19.5% [95% confidence interval (CI): 14.3–26]. After a mean follow-up of 36±23 months, the risk of death was higher in pericardial effusion patients [hazard ratio (HR) 1.59, 95% CI 1.37–1.85, P < 0.0001]. Stratifying for the main disease, pericardial effusion is associated with unfavourable outcome in all available subgroups: pulmonary arterial hypertension HR 1.53 (95% CI: 1.22–1.92; P < 0.0001), chronic heart failure (CHF) HR 1.53 (95% CI: 1.22–1.92; P < 0.0001), myocardial infarction HR 2.65 (95% CI: 1.4–4.99; P=0.003) and malignancies HR 1.75 (95% CI: 1.09–2.81, P=0.021). The lack of data concerning the idiopathic pericardial effusion does not permit a secure risk assessment but the average incidence of mortality is 14.5% (95% CI: 7.7–25.6). Conclusion Pericardial effusion should be considered a marker of the severity of the underlying disease, whereas for idiopathic pericardial effusion the correlation with poor prognosis is less clear.