摘要

In terms of inconsistent conclusions across all relevant randomized controlled trials (RCTs) and available meta-analyses, we aimed to use a meta-analysis and trial sequential analysis (TSA) to evaluate whether clinical utility of a genotype-guided warfarin initiation dosing algorithm could be better than that of a standard therapy regimen, and whether currently relevant evidence could be reliable and conclusive. Overall, 11 eligible RCTs involving 2677 patients were included for further analyses. Compared with fixed dose or clinically adjusted warfarin initiation dosing regimens, genotype-guided algorithms significantly increased time in therapeutic range, shortened time to first therapeutic international normalized ratio (INR) and time to stable doses, but did not show any marked improvements in excessive anticoagulation, bleeding events, thromboembolism, or all-cause mortality. Subgroup analyses revealed that, genotype-guided algorithms showed better control in the outcomes of time in therapeutic range or excessive anticoagulation than fixed-dose regimens rather than clinically adjusted regimens. Except for excessive anticoagulation, currently available evidence of all other outcomes was unreliable and inconclusive as determined with TSA. Our findings suggest that genotype-guided warfarin initiation dosing algorithms have superiority in the improvement of surrogate quality markers for anticoagulation control, but that this does not translate into statistically significant differences in clinical outcomes, which is largely because of the insufficient sample size in the RCTs analyzed.