Risk of aspirin continuation in spinal surgery: a systematic review and meta-analysis.

BACKGROUND CONTEXT

Aspirin is typically discontinued in spinal surgery because of increased risk of hemorrhagic complications. The risk of perioperative continuation of aspirin in neurosurgery needed to be evaluated.

PURPOSE

This study aimed to evaluate all available evidence about continuation of aspirin and to compare peri- and postoperative blood loss and complication rates between patients that continued aspirin and those who discontinued aspirin perioperatively in spinal surgery.

STUDY SETTING

Systematic review and meta-analysis were carried out.

METHOD

A meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies comparing aspirin continuation with discontinuation were included. Studies using a combination of anticlotting agents or non-spinal procedures were excluded. Operative outcomes (blood loss and operative length) and different complications (surgical site infection [SSI]), stroke, myocardial infarction within 30 days postoperatively) were extracted. Overall prevalence and means were calculated for the reported outcomes in fixed-effects models with heterogeneity (I-squared [I]) and effect modification (P-interaction) assessment.

RESULTS

Out of 1,339 studies, three case series were included in the meta-analysis. No significant differences in mean operating time were seen between the aspirin-continuing group (mean=201.8 minutes, 95% confidence interval [CI]=193.3; 210.3; I=95.4%; 170 patients) and the aspirin-discontinuing group (mean=178.4 minutes, 95% CI=119.1; 237.6; I=93.5%; 200 patients); (P-interaction=0.78). No significant differences in mean perioperative blood loss were seen between the aspirin-continuing group (mean=553.9 milliliters, 95% CI=468.0; 639.9; I=83.4%; 170 patients) and the aspirin-discontinuing group (mean=538.7 milliliters, 95% CI=427.6; 649.8; I=985.5%; 200 patients); (P-interaction=0.96). Similar non-significant differences between the two groups were found for cardiac events, stroke, and surgical site infections.

CONCLUSIONS

This meta-analysis showed an absence of significant differences in perioperative complications between aspirin continuation and discontinuation. Because of the paucity of included studies, further well-designed prospective trials are imperative to demonstrate potential benefit and safety.