Incidence of sharps injuries in surgical units, a meta-analysis and meta-regression.

Verbeek, Jos; Basnet, Prativa

BACKGROUND Sharps injuries occur often among surgical staff, but they vary considerably. METHODS We searched PubMed and Embase for studies assessing the incidence of sharps injuries. We combined the incidence rates of similar studies in a random effects meta-analysis and explored heterogeneity with meta-regression. RESULTS We located 45 studies of which 11 were randomized control trials, 15 were follow-up studies, and 19 were cross-sectional studies. We categorized injuries as self-reported, glove perforations, or administrative injuries. We calculated the population at risk as person-years and as person-operations (po). Meta-analysis of the incidence rate based on the best outcome measure resulted in 13.2 injuries per 100 time-units (95% confidence interval [CI], 4.7-37.1; I2 = 100%). Per 100 person-years, the injury rate was 88.2 (95% CI, 61.3-126.9; 21 studies) for self-reported injuries, 40.0 for perforations (95% CI, 19.2-83.5; 15 studies), and 5.8 for administrative injuries (95% CI, 2.7-12.2; 5 studies). Per 100 po, the respective figures were 2.1 (95% CI, 0.8-5.0; 4 studies), 11.1 (95% CI, 6.6-18.9, 15 studies), and 0.1 (95% CI, 0.05-0.21). I2 values were all above 90%. Meta-regression indicated lower incidence rates in studies that used perforations per po. CONCLUSIONS A surgeon will have a sharps injury in about 1 in 10 operations . Reporting of sharps injuries in surgical staff should be standardized per 100 po and be assessed in prospective follow-up studies.