A systematic review and meta-analysis of periprocedural bridging for patients with mechanical heart valves undergoing non-cardiac interventions.

OBJECTIVE

To evaluate the safety and efficacy of perioperative bridging in patients with mechanical heart valves undergoing non-cardiac interventions.

MATERIALS AND METHODS

A systematic research using Medline, EMBASE, and Google Scholar was implemented corresponding to the Preferred Reporting Items for Systematic reviews and Meta-analysis (PRISMA) statement. Data from the eligible studies were obtained and meta-analyzed. Primary endpoints included major bleeding and thromboembolism. Secondary endpoints included minor bleeding, overall mortality, and overall bleeding (major and minor bleeding). We conducted a comparative analysis between bridging and non-bridging along with a sensitivity analysis for patients undergoing major and minor operations.

RESULTS

Fifteen studies comprised of 2305 patients (2453 bridging episodes) were included. Pooled major bleeding and thromboembolism rates were 3.85 % (95 % CI: 2.12-5.98) (I = 69 %, p < 0.01) and 0.39 % (95 % CI: 0.00-1.41) (I = 64 %, p < 0.01). Bridging versus non-bridging major bleeding, thromboembolism, and overall bleeding risk ratios (RR) were RR 2.05 (95 % CI: 0.98-4.28) (I = 10 %, p = 0.34), RR 1.63 (95 % CI: 0.41-6.50) (I = 0 %, p = 0.63) and RR 1.79 (95 % CI: 1.17-2.72) (I = 55 %, p = 0.09) respectively. Subgroup analysis displayed major and minor operation thromboembolism and overall bleeding rates of 3.09 % (95 % CI: 0.78-6.43) (I = 0 %, p = 0.89) versus 0.14 % (95 % CI: 0.00-1.40) (I = 0 %, p = 0.93), test for subgroup differences (p < 0.01) and 17.37 % (95 % CI: 11.73-23.77) (I = 0 %, p = 0.61) versus 28.18 % (95 % CI: 22.80-33.88) (I = 0 %, p = 0.47), test for subgroup differences (p = 0.01) respectively.

CONCLUSION

Our analysis suggests that bridging may potentially put patients at an increased bleeding risk regarding overall bleeding rates, while failing to provide statistically significant benefits concerning thromboembolism and overall mortality compared to non-bridging. Limitations such as the mixed patient population don't allow for definite conclusions to be drawn warrantying further research through randomized controlled trials.