In response to the letter from Amin (1), we provide further details on the risk factors for progression and regression of retinopathy in our natural history study (2). Of 136 patients with retinopathy at baseline, 72 progressed or persisted compared with 50 patients (37%) who regressed to no retinopathy and 14 who regressed to a lower grade of retinopathy after a median follow-up of 3.1 years in both groups. Those who progressed or persisted had longer duration (7.8 vs. 5.9 years, P = 0.0086) and higher HbA1c (9.1 vs. 8.5%, P = 0.034) at baseline and were older at follow-up (18.1 vs. 17.4 years, P = 0.037). In multivariate logistic regression analysis of baseline factors, glycemic control (P = 0.0057) and duration of diabetes (P = 0.017) were significant predictors of progression/persistence versus regression of retinopathy to normal; these data are consistent with the Diabetes Control and Complications Trial. Neither high baseline BMI nor blood pressure percentiles were significant contributors to the incidence or progression/regression of retinopathy.
We agree that our data support a relationship between puberty and microvascular complications (using age ≥11 years as a surrogate marker for puberty). In this cohort, however, we did not find a relationship between pubertal staging and progression/regression of retinopathy, but the small number (n = 13) of prepubertal patients with retinopathy prevented us from answering this question. In addition, the permissive effect of puberty, growth hormone, and IGF-1 may be more pronounced in the pathogenesis of microalbuminuria than retinopathy. When we studied an incident cohort after 6 years’ duration, we found that higher pubertal stage (Tanner stage 4–5 vs. 1–3) had a larger effect on elevation of albumin excretion than retinopathy (odd ratios 5.2 vs. 3.4). The effect on albumin excretion rate was independent of HbA1c (3).