Ethnic differences in the clinical and laboratory associations with retinopathy in adult onset diabetes: studies in patients of African, European and Indian origins.

OBJECTIVE: To evaluate the prevalence of diabetic retinopathy (DR) and its associations in adult onset diabetic patients of African, European and Indian origins. DESIGN: The prevalence of retinopathy was determined by 60 degrees retinal photography in 507 consecutive out-patients. Clinical and laboratory associations were evaluated. SETTING: Diabetes clinic in a large community hospital. MAIN OUTCOME MEASURES: The associations between clinical and laboratory measurements with retinopathy. RESULTS: African patients (A) had shorter duration of diabetes (P < 0.001), higher HbA1 levels (P < 0.01) compared to those of Europeans (E) and Indian (I) extraction. A also had lower C-peptide levels (median 0.57 nmol L-1; vs. E. 0.81 nmol L-1 and I, 0.93 nmol L-1) (P < 0.001). The prevalences of retinopathy at diagnosis (21-25%) and overall were similar (A 37%, E 41%, I 37%). Severe DR was more frequent in the Africans (52%, P < 0.0001) and Indians (41%, P = 0.03) compared to the Europeans (26%). In Africans DR was significantly associated only with duration of diabetes (P < 0.0001) and macro-albuminuria (P = 0.01); in I it was also associated with systolic BP (P = 0.03); in E also with lower C-peptide levels (P = 0.0002), worse glycaemic control and greater use of insulin (P < 0.0001). In patients with DR insulin was used less frequently in A (35%) than in E patients (62%) (P = 0.001). CONCLUSIONS: In South Africa, the African population with adult onset diabetes has the highest prevalence of severe retinopathy, probably the result of very poor glycaemic control attributable to more severe insulinopenia and infrequent insulin treatment. Visual loss from diabetic retinopathy is likely to be considerable in Africans.

Screening for diabetic retinopathy in South Africa with 60 degrees retinal colour photography.

OBJECTIVES: Comparison of 60 degrees mydriatic retinal photography, in screening for diabetic retinopathy, with diabetes clinic doctors, formal ophthalmological assessment, and with one or two 45 degrees fields. DESIGN: Consecutive subjects screened by clinicians and photography, and selected eyes evaluated by an ophthalmologist. Randomized photographs assessed through one or two 45 degrees fields (by masking the slides), and at 60 degrees. SETTING: The first 663 patients attending for routine clinic visits and screened for retinopathy. MAIN OUTCOME MEASURES: The relative diagnostic sensitivity of screening methods, the utility of screening one eye only, and the costs of photographic screening. RESULTS: Compared to an ophthalmologist's assessment, retinal photography had a sensitivity of 93% and a specificity of 89% for any retinopathy, and 100 and 75%, respectively, for severe retinopathy. Photography detected 28% more retinopathy (16% severe) than the clinicians. Compared to a 60 degrees field, one 45 degrees field missed 31%, and 2 x 45 degrees fields 11% of retinopathy. Of 57 patients with retinopathy meeting referral criteria, 31 pairs of eyes had substantially discordant scores. The cost of diagnosis in a patient requiring referral to ophthalmologist was about US $37.00. CONCLUSIONS: 60 degrees retinal photography compares well with an ophthalmologists screening, and is better than clinical and one to two 45 degrees field assessments. Both retinae should be screened. This method is cost-effective in our hands.