Comparison of the efficacy of topical insulin with autologous serum eye drops in persistent epithelial defects of the cornea.

PURPOSE: To investigate the effect of topical insulin on epithelization in persistent epithelial defects (PED) refractory to usual treatment compared to autologous serum. DESIGN: Retrospective, consecutive case-control series. METHODS: The charts of 61 consecutive patients with PED treated with topical insulin (case group) and 23 treated with autologous serum (control group) were reviewed. Primary efficacy end points were the percentage of patients in which epithelization was achieved, as well as the rate and time until epithelization. Secondary efficacy point was need for amniotic membrane transplantation (AMT) or other surgeries. RESULTS: Mean time between PED diagnosis and start of topical insulin was 22.7 ± 18.5 days (range 13-115) and the mean area was 14.8 ± 16.2 mm2 (range 1.1-70.6). In the control group, mean time was 27.9 ± 16.8 days, mean epithelial defect area being 18.6 ± 15.0 mm2 (range 1.7-52.9). No differences in baseline characteristics were found between groups (p > 0.05). Epithelization was achieved in 51 patients (84%) on insulin and 11 patients (48%) on autologous serum (p = 0.002). In those patients, mean time until reepithelization was 32.6 ± 28.3 days (range 4-124) in the insulin group and 82.6 ± 82.4 days (range 13-231) in the autologous serum group (p = 0.011). The need for AMT was significantly lower in the insulin group (p = 0.005). PED recurrence was higher in patients treated on autologous serum (43%) compared with insulin (11%) (p = 0.002). CONCLUSIONS: Topical insulin is an effective treatment and safely promotes healing of PED. In our series, topical insulin presented better epithelization outcomes than autologous serum and could thus be considered as a first-line treatment.