We investigated familial clustering of diabetic retinopathy and nephropathy in the families of 372 subjects from the Diabetes Control and Complications Trial (DCCT). These subjects had 467 first-degree relatives with IDDM or NIDDM. Family sizes ranged from two to six. A complete data set was obtained from 241 relatives of 217 DCCT subjects. Among the DCCT subjects, 53% were in the intensive treatment group and 47% were in the conventional group; 44% were from the primary prevention cohort (no retinopathy or microalbuminuria at the DCCT baseline) and 56% were from the secondary intervention cohort (mild-to-moderate nonproliferative retinopathy and <200 mg/24 h albumin excretion rate [AER] at baseline). Retinopathy and nephropathy were assessed with seven-field stereo fundus photography and timed urinary AER measurements. Retinopathy was assessed using the Early Treatment Diabetic Retinopathy Study (ETDRS) scale. ETDRS scores and AERs were adjusted for the DCCT treatment group and for significant covariates from among sex, age, diabetes duration, HbA(lc) value, and body weight. Familial associations were assessed by comparing the prevalence of retinopathy and nephropathy in diabetic relatives of the respective positive versus negative DCCT subjects. To determine family clustering of the severity of retinopathy or nephropathy, the intraclass (familial) correlation was computed from the log-adjusted retinopathy and nephropathy scores of DCCT subjects and their relatives for all family members and sib-sib relationships. For parent-offspring, mother-child, and father-child relationships, the pairwise estimate of the correlations was computed. A correlation of 0.2 was considered to be biologically meaningful a priori. Among families of patients in the intensive and conventional groups combined, there was an increased risk of severe retinopathy (an ETDRS score greater than or equal to 47, clinically significant macular edema, or laser treatment in either eye) among relatives of retinopathy-positive vs. retinopathy-negative DCCT subjects in the secondary intervention cohort (odds ratio [OR], 3.1; 95% CI 1.2-7.8; P < 0.05). There was no increase in the risk of retinopathy of any severity (microaneurysms or morsel in the relatives of retinopathy-positive vs. retinopathy-negative DCCT subjects of the primary prevention cohort. There was an increased risk of nephropathy (AER >40 mg/24 h) in relatives of nephropathy-positive versus nephropathy-negative DCCT subjects of the secondary intervention cohort (OR, 5.4; 95% CI 2.2-13.7; P < 0.001). The risk of severe retinopathy in the relatives of positive versus negative subjects from the conventional treatment group alone (OR, 4.3; 95% CI 1.01-18.6; P < 0.05) was statistically significant and somewhat greater than that among relatives of the subjects in intensive treatment group (OR, 2.4; 95% CI 0.7-8.1), which was not significant. Correlations for the severity of retinopathy were 0.187 (all family members), 0.327 (parent-offspring), 0.249 (father-child), 0.391 (mother-child), and 0.060 (sib-sib), using the combined treatment group families. All these correlations were statistically significant (P < 0.05), except for sib-sib. The results showed similar trends when the families from the conventional and intensive treatment groups were analyzed separately. Correlations for nephropathy in the combined treatment group families were 0.063 (all family members), 0.138 (parent-offspring), 0.170 (father-child), 0.103 (mother-child), and 0.107 (sib-sib). None of these correlations is statistically significant. The lack of significant correlation for the severity of nephropathy may reflect the relatively short duration of diabetes in the offspring of these families and the known high intrasubject variability of AERs. These data provide the first available evidence that the severity of diabetic retinopathy is influenced by familial (possibly genetic) factors and confirmatory evidence that such factors influence the development of diabetic nephropathy.
