Long-Term Outcome for the Surgical Treatment of Infective Endocarditis With a Focus on Intravenous Drug Users

Background. We reviewed our experience with surgical procedures for infective endocarditis (IE) in order to evaluate modern outcomes and objectively examine our institutional preferences, including the use of bioprostheses in intravenous drug users (IVDUs) regardless of age and prompt surgical intervention in patients with either septic cerebral emboli or active infection. Methods. Review of medical records was conducted from February 1999 to November 2010. The Social Security Death Index was used to determine death from any cause in the postoperative period. Hospital records were used to identify infectious complications, recurrent endocarditis, and reoperation. Results. Sixty-four patients were identified as IVDUs and 133 patients as non-IVDUs. Survival at 30 days, 1 year, 5 years, and 10 years for IVDUs and non-IVDUs was 91.2% versus 93.6%, 77.5% versus 83.0%, 46.7% versus 71.1%, and 41.1% versus 52.0%, respectively. Cox regression analysis identified intravenous drug use as an independent risk factor for diminished survival (p = 0.03), although not for reoperation (p = 0.95) despite 95.3% of IVDUs receiving bioprostheses versus 73.7% of non-IVDUs (p = 0.0002, Fisher's exact test). Forty-three patients were identified as having preoperative septic cerebral emboli; none had a perioperative hemorrhagic event. Active infection approached significance as an independent risk factor for the composite end point of recurrent IE and perioperative infection (odds ratio 2.8; 95% confidence interval, 0.777 to 10.9; p = 0.12, Fisher's exact test). Conclusions. Bioprostheses are reasonable for IVDUs undergoing valve replacement for IE regardless of age. Prompt surgical intervention in the setting of septic cerebral emboli is justified; in the setting of active infection it is less clear. (Ann Thorac Surg 2012;93:51-8) (C) 2012 by The Society of Thoracic Surgeons