This study was undertaken to examine the possible relationships between muscle capillary basement membrane width (CBMW) and glycemic control, bone age, chronologic age, and duration of diabetes in young patients with insulin-dependent diabetes mellitus (IDDM) during different stages of pubertal development. We studied 49 males and 43 females (age, 7-20 yr) with IDDM for up to 16 yr for whom bone age and glycosylated hemoglobin (HbA(1c)) data were available at the time of right quadriceps muscle biopsy. Based on pubic hair Tanner stage, subjects were assigned to prepubertal (Tanner I), pubertal (Tanner II and III), and postpubertal (Tanner IV and V) groups. In 30 pubertal and prepubertal subjects, none of the variables studied was significantly correlated with CBMW. This is attributable in part to the small number of subjects in each group. In 62 postpubertal subjects, CBMW was correlated with age (r = .27, P = .03), bone age (r = .43, P = .0005), and postpubertal duration of diabetes (r = .38, P = .003) but not total duration of diabetes. In the postpubertal subjects, CBMW was correlated with HbA(1c) at the time of biopsy (r = .31, P = .01) but correlated more strongly with the mean of HbA(1c) values obtained during the 1- and 2-yr periods before biopsy (r = .37, P = .01, and r = .54, P = .03, respectively). An analysis of covariance revealed that the slopes for the regression of log(e) CBMW on HbA(1c) differed significantly (P = .02) among the three groups. This difference is due primarily to the negative relationship between log(e) CBMW and HbA(1c) in the prepubertal group in contrast to the positive relationship in the other two groups. Multiple regression analyses performed on data from the postpubertal group showed that bone age and glycemic control were independently associated with CBMW. These observations suggest that in children with IDDM, CBMW is influenced after puberty by the factors responsible for skeletal maturation (i.e., sex steroids) and also by the degree of hyperglycemia.