A 20-year experience with isolated pericardiectomy: Analysis of indications and outcomes

Objectives: Outcome after pericardiectomy depends on many factors, but no large study has provided clarity on the effects of patient variables or cause of pericarditis on patient survival. We report early and late results from a 20-year experience with isolated pericardiectomy. Methods: From January 1993 to December 2013, 938 patients underwent pericardiectomy at our institution. In order to establish a homogeneous population to analyze the impact of pericardiectomy, we excluded patients with prior chest radiation, malignancy, and concomitant valvular or coronary procedures. We identified a cohort of 521 who underwent isolated pericardiectomy; of these, 513 patients gave consent for research and comprise the cohort for this analysis; median age at operation was 57 years (range, 18-84 years) and 363 (71%) were men. Indications for pericardiectomy were effusive/chronic relapsing pericarditis in 158 (31%) and pericardial constriction in 355 (69%). Prior coronary artery bypass grafting had been performed in 84 patients (14%). Median preoperative left ventricular ejection fraction was 60% (range, 24%-80%), and 77% of patients were in New York Heart Association (NYHA) functional class III/IV. Results: Surgical approach was median sternotomy in 412 (80%), left thoracotomy in 71 (14%), and clamshell in 30 (5%). Extent of pericardial resection was radical in 414 (81%), subtotal in 71 (14%), and completion in 28 (5%). Cardiopulmonary bypass was used in 205 (40%). Overall mortality was 12/513 (2.3%); 3/158 (1.9%) for the effusive/chronic relapsing group versus 9/355 (2.5%) for the constriction group (P = .65). In the absence of multivariate predictors, which could not be identified, univariate predictors associated with increased risk of early death included lower left ventricular ejection fraction (hazard ratio [HR], 1.09; P = .03) and preoperative renal insufficiency (HR, 9.9; P < .001). Median duration of follow-up was 29 months (maximum 20.5 years) and overall 5-, 10-, and 15-year survival was 80%, 60%, and 38%, respectively. Overall survival according to surgical indication was higher in the effusive/chronic relapsing group when compared with the constriction cohort (P < .001). Independent predictors associated with increased risk of overall mortality identified on multivariate analysis included older age (HR, 1.05; 95% confidence interval [CI], [1.03, 1.07]; P < .001), congestive heart failure (HR, 1.49; 95% CI, [1.03, 2.2]; P = .02), diabetes (HR, 1.83; 95% CI, [1.2, 2.7]; P = .004), completion pericardiectomy (HR, 2.4; 95% CI, [1.2, 4.7]; P = .01), and chronic obstructive pulmonary disease (HR, 2.45; 95% CI, [1.5, 3.9]; P = .004). During the follow-up period, 80% of patients were free from NYHA functional class III/IV symptoms at 5 years and 78% at 10 years. Conclusions: Whereas early mortality after isolated pericardiectomy is low irrespective of the indication for surgery, late follow-up demonstrates better outcomes after pericardiectomy for effusive/chronic relapsing pericarditis compared with pericardial constriction. Importantly, the majority of patients were free from significant heart failure symptoms during follow-up.