To investigate the effects of normoglycaemia on diabetic retinopathy, we evaluated 18 uremic diabetic patients before and after successful pancreas kidney transplantation. In all, 12 uremic diabetic patients who submitted to kidney transplantation alone served as the control group; 4 of these subjects received a kidney transplantation alone, whereas 8 underwent a double kidney-pancreas transplantation but lost the pancreas graft within the first few weeks post-surgery. The mean age and the mean duration of both diabetes and dialysis were comparable in the two groups. All patients were studied prior to and at 6 and 9 months after surgery, then at annual intervals. Subjects were divided into three groups according to follow-up: less than 1 year, between 1 and 3 years and greater than 3 years. At each control visit, a complete clinical examination was performed by two independent examinators; retinal fluorescein angiography was carried out as well. The following parameters were evaluated: visual acuity, capillary closure, macular oedema, neovascularization at the disk and elsewhere and vitreous haemorrhage. A score ranging from -2 to +2 was assigned to each parameter for quantification of the variation between baseline values and those obtained at the end of the follow-up. This score was assigned by two different ophthalmologists. Eyes that were affected at baseline by end-stage diabetic retinopathy (secondary retinal detachment, neovascular glaucoma) were not entered in the study. A total of 18 eyes were lost to follow-up in the 2 groups because of laser treatment, cataract extraction, anterior ischaemic optic neuropathy and cytomegalovirus retinitis.(ABSTRACT TRUNCATED AT 250 WORDS)