Emergency valve surgery improves clinical results in patients with infective endocarditis complicated with acute cerebral infarction: analysis using propensity score matching†

Abstract

Abstract OBJECTIVES To date, the optimal timing for patients with infective endocarditis (IE) with acute cerebral infarction (CI) to undergo valve surgery is unknown. Although some previous studies have reported that early valve surgery for IE patients within 1 or 2 weeks after CI could be performed safely, an initial strategy has not been identified because of the unmatched cohorts in previous studies. This study aimed to assess the feasibility and safety of early surgery within a few days after cerebral infarction by using propensity score matching. METHODS Between 2009 and 2017, 585 patients underwent valve surgery for patients with active IE at 14 institutions. Among these, 152 had preoperative acute CI. Early surgery was defined as surgery within 3 days after the diagnosis of CI. Of these 152 patients, 67 underwent early valve surgery (early group), whereas 85 underwent delayed valve surgery (delayed group). Of the patients, 45 in each group were analysed using propensity score matching. The primary outcome was in-hospital death after valve surgery, and secondary outcomes included neurological complications. We compared the clinical results of these matched patients. RESULTS Hospital mortality was lower in the early group (2% vs 16%, P = 0.058). The rate of postoperative intracranial haemorrhage in the early and delayed groups was 4% in both groups. The postoperative modified Rankin scale was not significantly different [early group: 0 (0–2); delayed group: 0 (0–2)]. Incidence of neurological deterioration did not differ significantly between the groups. The survival rates after the first discharge at 1, 3 and 5 years after valve operation were 100%, 97% and 97% in the early group and 91%, 83% and 80% in the delayed group, respectively (P = 0.029). CONCLUSIONS Early valve surgery for patients with IE within 3 days after a CI measuring <2 cm in size improved clinical results without increasing the incidence of postoperative neurological complications.