Introduction: The efficacy of massage to treat neonatal hyperbilirubinemia remains controversial. We conducted a systematic review and meta-analysis to explore the influence of massage on the neonatal hyperbilirubinemia. Methods: We search PubMed, Embase, Web of science, EBSCO, and Cochrane Library databases through November 2017 for randomized controlled trials (RCTs) assessing the effect of massage on neonatal hyperbilirubinemia. This meta-analysis is performed using the random-effect model. Results: Six RCTs involving 357 patients are included in the meta-analysis. Overall, compared with the control group in neonatal hyperbilirubinemia, massage therapy is associated with substantially reduced serum bilirubin level within 4 d (mean difference (MD) = -2.31; 95% CI = -2.92 to -1.70; p < .00001) and transcutaneous bilirubin level within 4 d for neonatal hyperbilirubinemia (MD = -1.97; 95% CI = -2.55 to -1.39; p < .00001), but results no remarkable impact on serum bilirubin level on 2 d (MD = -0.82; 95% CI = -2.16-0.52; p = .23), transcutaneous bilirubin level on 2 d (MD = -0.17; 95% CI = -1.34 to 1.00; p = .77), frequency of defecation daily on 2 d (MD = 0.57; 95% CI = -0.03 to 1.16; p = .06), and frequency of defecation daily within 4 d (MD = 0.83; 95% CI = -0.11 to 1.76; p = .08). Conclusions: Massage therapy can significantly reduce serum bilirubin level and transcutaneous bilirubin level within 4 d, but demonstrates no influence on serum bilirubin level and transcutaneous bilirubin level on 2 d, frequency of defecation daily on 2 and 4 d for neonatal hyperbilirubinemia.
Keywords: Bilirubin level; massage; meta-analysis; neonatal hyperbilirubinemia; randomized controlled trials.