Model for end-stage liver disease predicts mortality after pericardiectomy for constrictive pericarditis

Prognosis after pericardiectomy for constrictive pericarditis (CP) is affected by the aetiology of the constriction as well as by concomitant cardiac and non-cardiac disease, including liver dysfunction. However, few data exist on the risk stratification that accounts for liver function in patients with CP. We evaluated the effectiveness of the model for end-stage liver disease (MELD) score, an established measure of liver dysfunction, in predicting long-term survival and identifying other risk factors for death.

A total of 79 patients who underwent pericardiectomy for CP at a single centre between 2009 and 2016 were analysed. The prognostic utility of the MELD score was evaluated in our cohort. Logistic regression and Cox proportional hazard regression analyses were performed to assess the association of various clinical variables with 1-year and overall mortality rates.

With multivariable analysis, only the MELD score was an independent predictor of 1-year mortality (P < 0.001); apart from the MELD (P = 0.003) score, post-surgical CP (P = 0.016), total bilirubin level (P = 0.042) and the European System for Cardiac Operative Risk Evaluation score II (P = 0.002) were independent predictors of overall mortality after pericardiectomy. Overall survival decreased as the MELD score increased. Scores ≤ 7.5, 7.51-15.50 and >15.5 were associated with overall survival rates of 92.9%, 69.8% and 8.3%, respectively.

In addition to the underlying aetiology, we demonstrated that assessment of liver dysfunction using the MELD score provides additional information about risk because it is associated with postoperative death in patients undergoing pericardiectomy for CP.