Diabetic retinopathy and insulin therapy in a rural diabetic population

Mayombian ethnic, vegetables low intake, insulin treatment, diabetic nephropathy and severe diabetic retinopathy are determinants of blindness in diabetic Africans

AIM

To determine the frequency and causes of blindness in diabetic Africans.

METHODS

The study was a cross-sectional survey carried out among known black diabetics consecutively admitted at the Teaching Hospital, University of Kinshasa, between 2005 and 2007. Examination methods included interviewer-administered structured questionnaire, eye examinations (visual acuity, tonometry, funduscopy), and fasting plasma glycaemia test.

RESULTS

Of the 227 patients examined, 15.9% had blindness. Univariate analyses showed significant association between female, severity of diabetic retinopathy, Mayombian ethnic group, use of insulin treatment, low intake of vegetables, diabetic nephropathy, open angle glaucoma and blindness in all diabetics. After logistic regression, only diabetic nephropathy, use of insulin treatment, macular oedema, Mayombian ethnic group and vegetables low intake were the independent risk factors of blindness in all diabetics. However, after logistic regression in the sub-group with diabetic retinopathy, only open angle glaucoma and proliferative diabetic retinopathy were the independent determinants of blindness.

CONCLUSION

The majority of the causes of blindness in these diabetic Africans are avoidable. It is recommended that appropriate diabetes care, nutrition education, periodic eye examination and laser photocoagulation facilities should be provided for treating diabetics in sub-Saharan Africa.

Diabetic eye disease: a natural history study

In previous studies on diabetic retinopathy it has not been possible to relate risk factors to reduced vision because of the influence of vision-preserving treatment. Demographic data, cardiovascular risk factors and ocular features from the diabetic population of the Seychelles are described. Diabetic retinopathy in this population had not been modified by laser treatment. The population described consists of entirely type 2, maturity onset diabetics. Using a multivariate logistic regression model, the risk factors were deduced for three outcome variables: (1) reduced vision, defined as 6/36 or worse in both eyes; (2) the presence of diabetic retinopathy; and (3) the presence of maculopathy, preproliferative and proliferative retinopathy, grouped as severe retinopathy. Insulin treatment was associated with all outcome variables, duration from diagnosis of diabetes with retinopathy of all forms, and increasing age with reduced vision and severe retinopathy. Hypertensive diabetic patients were twice as likely to have reduced vision as compared with non-hypertensive diabetic individuals.

Determinants of end-stage renal disease in Pima Indians with type 2 (non-insulin-dependent) diabetes mellitus and proteinuria

To identify factors related to the development of end-stage renal disease after the onset of proteinuria, its incidence was determined in 364 Pima Indians aged 35 years or older with Type 2 (non-insulin-dependent) diabetes mellitus and proteinuria (protein-to-creatinine ratio > or = 0.5 g/g). Of these 364 subjects, 95 (36 men, 59 women) developed end-stage renal disease. The cumulative incidence was 40% 10 years after and 61% 15 years after the onset of proteinuria. The incidence of end-stage renal disease was significantly related to the duration of diabetes, the duration of proteinuria, higher 2-h plasma glucose concentration, type of diabetes treatment, and the presence of retinopathy at the time of recognition of the proteinuria, but not to age, sex, or blood pressure. Duration of proteinuria influenced the risk of end-stage renal disease, contingent, however, upon the duration of diabetes at the onset of proteinuria. The higher cumulative incidence of end-stage renal disease 15 years after the onset of proteinuria in Pima Indians (61%) than in Caucasians from Rochester, Minnesota (17%) may be attributable, in part, to the younger age of onset of Type 2 diabetes in Pima Indians than in Caucasians, to ethnic differences in susceptibility to renal disease, or to lower death rates among the Pima Indians from competing causes of death, such as coronary heart disease.

The prevalence of diabetic retinopathy in northwest Spain. An epidemiological study of diabetic retinopathy in Galicia. I

The prevalence of retinopathy was studied in a group of 1179 diabetic patients attending 11 diabetologic centres in Gallica (Northwest of Spain). The age of the patients was 61.7 +/- 13.8 years, 43.8% were males and 56.2% females. The patients had had diabetes from 8.4 +/- 7.5 years, 30.4% were being treated with insulin, and 69.6% with oral antihyperglycaemics. 16.1% were aged under 40 years at onset as against 83.9% aged over 40 years. The prevalence of retinopathy was 44.7, in the whole group. Stepwise multiple regression analysis identified treatment time since onset and age at onset as the factors affecting prevalence, which was 61.8% in the insulin-treated subgroup vs 37% among those treated with oral antihyperglycaemics (p < 0.0001). The prevalence of proliferative diabetic retinopathy was 5.8% in the whole group.

The epidemiology and prevalence of diabetic retinopathy in the Veneto region of north east Italy. Veneto Group for Diabetic Retinopathy

The prevalence of diabetic retinopathy and its relationship to a number of risk factors were examined in a population-based study in the Veneto region of North East Italy. Of 1321 diabetic patients selected, 98% attended for examination. Prevalence of diabetic retinopathy was 26.2% (24.4% background and 1.8% proliferative). The prevalence of retinopathy was significantly related (p less than 0.01) to the duration of diabetes (17.3% for less than 5 years; 60.8% for greater than 20 years). Proliferative retinopathy was much more prevalent after 20 years of diabetes. After 10 years most proliferative retinopathy was found in Type 1 diabetic patients, but before 10 years from diagnosis it was most prevalent in Type 2 diabetes. The prevalence of retinopathy was significantly related (p less than 0.001) to the type of diabetes and was found predominantly in Type 1 (46.2%) and insulin-treated Type 2 (45.9%) subjects and to a lesser degree in non-insulin-treated patients (24.6%). The prevalence of retinopathy was significantly related to both fasting and post-prandial blood glucose levels (p less than 0.001), blood urea nitrogen (p less than 0.05), and systolic (p less than 0.001) and diastolic (p less than 0.01) blood pressure. No significant differences were found in the prevalence of total or proliferative retinopathy between males and females. No significant relationships were found with family history of diabetes, alcohol intake, smoking habits, cholesterol, triglycerides, and serum uric acid.

Non-stereo fundus photography as a screening procedure for diabetic retinopathy among patients with type II diabetes. Compared with 60D enhanced slit-lamp examination

The spectrum of diabetic retinopathy, the need for fundus screening and the evaluation of two presumptive screening methods, was investigated in a population based study among patients with type II diabetes. Retinal evaluation was performed in 86.9% of the known diabetic population. Background diabetic retinopathy was detected in 37.8%, pre-proliferative in 1.1% and proliferative retinopathy in 3.8%. Diabetic maculopathy was found in 24.3% of the patients. Laser therapy was considered in 11.4% of the patients due to diabetic retinopathy, and in 14.6% when venous occlusive diseases were included. Two methods, a slit-lamp observation enhanced by a 60D lens and reading from two non-stereo photographs of the posterior pole, were evaluated among 154 patients willing and mentally capable of being examined by either method. The sensitivity of the photographic method was 87/97% (right eye/left eye) when detecting background retinopathy and 81/80% for maculopathy versus 69/61% and 79/63%, respectively, with the slit-lamp method. The photographic method could be applied in 93% of the patients mentally capable of cooperation. Only 5 or 6 patients could be examined per hour with the 60D slit-lamp compared with 30-35 examined by reading retinal photographs. The photographic method is recommended as an easy and reliable screening device for diabetic retinopathy among patients with type II diabetes.

Subclinical left ventricular abnormalities in young diabetics

Twenty young asymptomatic diabetic patients were evaluated for left ventricular dysfunction by determining the radionuclide ejection fraction response to supine bicycle ergometry. The double product at peak exercise (28,743 +/- 3,314 vs 29,007 +/- 3,625, p greater than .05) was not significantly different between the two groups. Seven of 20 diabetics had either no change or a drop in their ejection fraction during exercise while 1 of 20 control subjects had no change in ejection fraction. There was no correlation between the FBS (r = .26) and HbA1c (r = .32) and ejection fraction change during exercise, although those diabetics with LV dysfunction tended to have a higher HbA1c level as compared to diabetics with a normal response (16.8 +/- 3.1 percent vs 12.5 +/- 3.8 percent respectively, p greater than .05). The LV systolic dysfunction in young asymptomatic diabetic subjects does not appear to correlate with the degree of acute or chronic hyperglycemia, and therefore, is not a direct function of the dynamic metabolic state of diabetes.

Prevalence of retinopathy in diabetes treated with oral antihyperglycaemic agents

In a random sample of all diabetics treated with oral antihyperglycaemic agents on the island of Gotland, Sweden, retinopathy (Rp) was present in 24 (17%) of 140 patients examined. Eighteen patients had background Rp. 7 of these with exudative lesions. Four patients had pre-proliferative and 2 proliferative Rp. On multiple logistic regression analysis a significant correlation was found between Rp on the one hand and a longer duration and a younger age at onset of the diabetes, on the other. Rp was also significantly correlated to higher levels of glycosylated haemoglobin (HbA 1C). In an additional analysis restricted to the group with Rp, higher levels of HbA 1C were found in persons with Rp of a more severe degree. On a simple logistic regression analysis, a significant correlation (P less than 0.05) was found between a high body mass index and a low prevalence of Rp. on multiple regression analysis this correlation became weaker (P = 0.06).