Prevention of Bleeding in Orthognathic Surgery—A Systematic Review and Meta-Analysis of Randomized Controlled Trials

PurposeThe objective of the present study was to evaluate the efficacy of hemostatic adjuncts on intraoperative blood loss (IOB) in orthognathic surgery (OS) detected by randomized controlled trials (RCTs) of the highest quality.Materials and MethodsA search of the Medline, Cochrane, Embase, and Web of Science databases was performed in January 2015, and the risk of bias was assessed using the Jadad and Delphi scales. The predictor variable was the hemostatic measures, and the main outcome variable was the total IOB volume. The secondary outcome variables were the hemoglobin and hematocrit and operating time. This review is registered at PROSPERO (CRD42014014840).ResultsEleven trials were included for review. The individual trials demonstrated the effects on IOB from hypotensive anesthetic regimens, the use of aprotinin, and the herbal medicine Yunnan Baiyao. Six studies of tranexamic acid (TXA), with 288 patients, were suitable for a meta-analysis of continuous data. TXA reduced IOB by an average of 171 mL (95% confidence interval [CI] −230 to −112; P < .00001). Its topical use yielded similarly significant results (mean difference −197, 95% CI −319 to −76; P < .001). A subgroup analysis showed a decreased operating time in the TXA groups by an average of 15 minutes (mean difference −14.78, 95% CI −22.21 to −7.35; P < .0001).ConclusionsEfficient hemostatic adjuncts exist for OS. Our meta-analysis showed that TXA significantly reduces IOB by an average of one third, regardless of whether it was given intravenously (IV) or applied topically. Additional RCTs are needed to confirm the effect of topical TXA in OS, and larger studies of intravenous administration are needed before any routine recommendations. No hemostatic effect of hypotensive anesthesia was found, mainly owing to imprecise descriptions of the blinding procedures. Transparent and uniform trial reporting is thus encouraged in future studies.