Vitreous loss during phacoemulsification learning curve performed by third-year residents.

OBJECTIVES To assess the resident's phacoemulsification learning curve as a risk factor for vitreous loss and to determine the incidence of vitreous loss among the residents performing phacoemulsification. DESIGN Retrospective matched case-control study MATERIAL AND METHOD A case-control study comparing all consecutive cases of attempted phacoemulsification with intraocular lens (IOL) implantation from January 1st, 1998 to December 31st, 1999. The surgeon variable will be categorized into two groups, the third year ophthalmology residents, representing resident's phacoemulsification learning curve, and faculty staffs. The study group consisted of eyes that had had intraoperative complication of vitreous loss. The control group consisted of eyes that had not had vitreous loss. Continuous variables were compared with the 2-sided unpaired t-test. Categorical variables were compared between groups using analytical matched case-control study with relative risk or odd ratio, Mc Nemar's (Marginal) o 2 test and 95% confident interval of relative risk. RESULTS The odds that the eyes in the resident group would have an intraoperative complication of vitreous loss were 4 times the odds that the eyes in the faculty staff group would have such complication (P = 0.0052, 95% confidential interval (CI) of relative risk (RR) = 1.516-10.556). The incidence of vitreous loss among residents was 6.93% (28/404) and 2.06% (28/1358) among the faculty staffs. The overall incidence of vitreous loss was 3.18% (56/1762). CONCLUSION The incidence of intraoperative complication of vitreous loss, the relative risk of such complication performed by the learning curve surgeon in the present study serve as benchmarks for residents-in-training, beginning and surgeon-in-practice converting to phacoemulsification.