Preoperative Intracranial Hemorrhage Affecting Surgical Decision Making in Endocarditis Patients: A Literature Review and a Single-Centre Retrospective Study

Objectives: Cardiac surgery in infectious endocarditis (IE) patients with preoperative intracranial hemorrhage (Pre-ICH) is considered a highly debatable issue, and guidelines are still not well defined. Recent studies suggest that the incidence of neurological exacerbations has decreased in recent years. In this study, we aimed to investigate the effect of cardiac surgery on clinical outcomes of IE patients accompanied by Pre-ICH.

Methods: Through using PubMed, EMBASE, and Cochrane databases, a systematic review of the recent publications (1995–August 2018) was conducted. Twenty-six articles were retrieved, but due to the lack of controlled studies, only seven articles provided the best evidence to answer our clinical question and thus were included. We also retrospectively analyzed data from our own patients who were operated for left-sided endocarditis between January 2007 and May 2018. A total of 525 patients were identified, but only 287 patients with preoperative brain imaging were included in further analysis.

Results: Literature review illustrated a total of 133 cases of IE patients with Pre-ICH who underwent cardiac surgery. A total of 112 patients (84.2%) reported no postoperative neurological deterioration regardless of the timing of the surgery; 12 patients (9.1%) had postoperative events (10 with new strokes as cerebral infarcts or ICH; 2 with ectopic asymptomatic hemorrhages). However, none of the observed events led to death. Nine patients (6.7%) died, and one article confirmed the cause of mortality in four of nine deaths as cardiac or multiorgan failure (MOF). As for our own retrospective study, we identified a total of 34 patients with Pre-ICH who were assigned to cardiac surgery. There was no significant difference in hospital mortality between patients with Pre-ICH and without Pre-ICH (29.4 vs. 28.9%; p = 1.0). The rate of postoperative bleeding was higher in patients with pre-ICH compared with those without pre-ICH, but the difference was not statistically significant (17.5 vs. 7.5%; p = 0.095). Out of six patients who presented with postoperative bleeding, three showed resorption of the ICH on follow-up brain imaging.

Conclusion: Recent studies report a low rate of neurological deterioration and good survival in operated IE patients with Pre-ICH. The results of our patient data and the literature review suggest that the perception of preoperative ICH may be worse than it actually is. These findings may facilitate decision making in endocarditis patients with intracranial hemorrhage.