Abstract
OBJECTIVE: Operative timing remains controversial for patients with left-sided infective endocarditis (LSIE) with preoperative stroke (PREOS). Operative guidelines are determined on the basis of postoperative radiologically confirmed strokes (RCS). We evaluated the impact of PREOS on surgical outcomes. METHODS: Over 15 years, 331 patients underwent valvular surgery for LSIE at our center. PREOS (n = 71, 21%) and non-PREOS (n = 260, 79%) cohorts were identified. Propensity score matching was performed. Logistic regression identified risk factors for postoperative clinical stroke (PCS, defined as any new postoperative neurologic deficit), RCS and mortality. RESULTS: Among patients with PREOS, 24 of 71 (34%) had a hemorrhagic component, 34 of 71 (48%) were within 2 weeks of surgery, 46 of 71 (65%) experienced residual deficits, and 2 of 71 (3%) experienced hemorrhagic conversion postoperatively. Operative mortality was 24 of 331 (7%) and did not significantly differ between groups (P = .083). Patients with PREOS had a greater incidence of PCS (P = .001), repeat imaging of the head (P < .001), new renal failure (P = .006), and nonhome discharges (P < .001). Propensity score matching upheld these trends. Logistic regression identified PREOS as a risk factor for PCS (odds ratio [OR], 8.76; P < .001). Intraoperative abscess (OR, 4.83; P = .013), cardiogenic shock (OR 8.51; P = .023), and tricuspid procedures (OR 5.03; P = .02) were RCS risk factors. PREOS (OR 3.12; P = .025), preoperative renal failure (OR 2.67; P = .043), immunosuppression (OR 7.09; P = .022), tricuspid regurgitation (OR 4.36; P = .011), and aortic valve procedures (OR 4.38; P = .033) were risk factors for mortality. CONCLUSIONS: Among patients with LSIE undergoing surgery, PREOS is a risk factor for PCS and new renal failure. Patients with PREOS may require greater level of care upon discharge and may benefit from more stringent preoperative evaluation.