RESUMEN
OBJECTIVES: The purpose of this study was to conduct a systematic review and meta-analysis in patients with local wound infection or infective risk, evaluating effects of topical gentamycin application on prophylaxis and treatment of wound infection. METHODS: Embase, the Cochrane Library, Pubmed, Medline (from Ovid) and three Chinese literature databases (CNKI, VIP and WANFANG) were searched. Randomised controlled studies (RCTs) and observational studies (OSs) that assessed the efficacy of topical gentamycin application on prophylaxis and treatment of local wound infection were included. The primary outcome was clinical efficacy. Secondary outcomes included duration of recovery time and length of hospital stay. RESULTS: Fifteen studies (1781 patients) met inclusion criteria. Twelve studies were RCTs and other three studies were OSs. Compared with non-gentamycin group, topical gentamycin application had significantly higher rates of clinical efficacy (OR = 3.57, 95% CI 2.52-5.07). In terms of duration of wound healing, it's taken shorter time in gentamycin group than non-gentamycin group (OR = -4.94, 95% CI -8.37 to -1.51). However, the length of hospital stay had no significantly difference between the two groups (OR = -3.40, 95% CI -8.42 to 1.63). Subgroup analyses were conducted according to study design (RCTs or OSs), purpose and administration type. And the results showed that there were no significant difference of clinical efficacy in study design (P = 0.21, I2 = 35.4%), purpose (P = 0.32, I2 = 0%) and administration type subgroup (P = 0.74, I2 = 0%). However, topical gentamycin application had significantly shorter duration of wound healing in randomly controlled trials compared with observational studies, but had no difference in terms of administration type(P = 0.20, I2 = 38.6%). CONCLUSIONS: Studies to date show that topical gentamycin application significantly increases the rate of clinical efficacy and decreases the duration of wound healing in patients with local wound infection or infective risk.