Microvascular disease: what does the UKPDS tell us about diabetic retinopathy?

Background The UK Prospective Diabetes Study (UKPDS) was a randomised controlled clinical study which looked at the effect of improved blood glucose and blood pressure control on macro- and microvascular complications in Type 2 diabetes. Retinopathy was the commonest microvascular outcome and this paper looks at this complication. Methods Newly diagnosed diabetic patients were randomised to intensive or conventional glycaemic control and a subset of hypertensive patients to tight or less tight blood pressure control. Patients were seen every 3 months in study clinics and retinopathy was assessed by adjudicated grading of triennial colour retinal photographs. Photocoagulation treatment, vitreous haemorrhage and cataract extraction were predefined UKPDS endpoints. Observational analyses of the data were used to examine the relationship of updated mean glycated haemoglobin (HbA(1c)) and mean blood pressure levels to retinopathy outcomes. Results The UKPDS showed that both improved glucose control and improved blood pressure control reduced the risk of retinopathy, with a linear relationship between the log hazard ratio for retinopathy and both updated mean HbA(1c) and updated mean blood pressure. A 1% decrement in HbA(1c) equated to a 31% reduction in retinopathy and a 10 mmHg decrement in systolic blood pressure equated to an 11% reduction in photocoagulation or vitreous haemorrhage. Evidence of retinopathy at diagnosis, including the presence of microaneurysms only, increased significantly the risk of progression to photocoagulation. Conclusions The UKPDS stands out as a landmark study in Type 2 diabetes, emphasising the crucial importance of controlling both blood glucose and blood pressure in order to minimise the risk of developing sight-threatening retinopathy.

The problems of retinal blood flow in diabetes.

Three methods are available for measurement of retinal blood flow: cinefluroescein angiography, measurement of mean transit time, and estimation of volume flow by laser-Dopler velocimetry. Cineangiography requires intra-arterial injection for volume flow measurement, and this greatly inhibits its use. Difficulties also arise in estimating the relation of the axial to the mean flow, since it is now known that the retinal flow is plug flow. Transit time measurement using fluroescein has recently been automated and thus allows for more accurate measurement and better estimation of volume flow. The problems of estimating vascular volume and allowing for leakage have not been overcome. Laser-Doppler velocimetry measures mean red cell velocity but its relation to volume flow has not been established. Using the mean-transit-time method, several authors have found reduced transit time and increased volume flow in patients with mild and no retinopathy, compared to normals and those with more severe lesions, This most probably is due to autoregulatory adaptation and, possibly, to hypoxia. In more severe retinopathy, flow is reduced and there are large areas of nonperfusion. These areas are the result of abnormality of the endothelial cells and also of the blood within the vessels, which probably coagulates more easily than in normals. These areas are of importance because they stimulate the growth of new vessels.

Insulin resistance and insulin deficiency in diabetic retinopathy of non-insulin-dependent diabetes.

To assess the role of insulin resistance and insulin deficiency in the pathogenesis of diabetic retinopathy in non-insulin-dependent diabetes mellitus, 13 patients with and 12 patients without retinopathy were studied. The glucose clamp technique was used to measure insulin resistance and insulin response to glucose. During the euglycemic clamp, at comparable steady-state levels of glucose and insulin, the mean glucose infusion rate, which indicates the rate of glucose utilization, was lower in the retinopathy group than in the nonretinopathy group (6.1 +/- 0.5 versus 8.1 +/- 0.7 mg . kg-1 . min-1, P less than 0.02). Growth hormone (GH) concentrations were higher in the retinopathy group 8.4 +/- 2.5 versus 2.5 +/- 0.7 microIU/ml, P less than 0.05), but they did not correlate significantly with insulin resistance, expressed as mean glucose turnover. During the hyperglycemic clamp (+7 mmol/L above the fasting plasma glucose), the insulin response in the two groups of diabetics was similar. Increased insulin resistance represents an additional factor, which together with other factors, may be important in the pathogenesis of diabetic microvascular complications.

Role of blood flow and impaired autoregulation in the pathogenesis of diabetic retinopathy.

Several mechanisms are implicated in the pathogenesis of diabetic retinopathy. They include biochemical, hemodynamic, and hormonal factors, all of which have an important role in the development of diabetic retinopathy. These factors are not independent of each other, but rather they interact and together are responsible for the well-known lesions of vascular occlusion, microaneurysms, hemorrhages' hard exudates, and eventually new vessel formation.

Comparison of stereofundus photographs in patients with insulin-dependent diabetes during conventional insulin treatment or continuous subcutaneous insulin infusion.

Sixty-five patients with mild to moderate nonproliferative diabetic retinopathy who enrolled in a prospective controlled clinical trial had stereofundus photographs assessed for change over an 8-mo period. The entire study group showed a worsening of retinopathy with time (P less than 0.001). The worsening was greater in the pump-treated group (15/32) than in the conventionally treated group (9/33). The significance of this difference ranged from P = 0.67, if changes in mean retinopathy level for each patient were compared, to P = 0.177 if a grading system keyed to the worse eye was compared. The difference in rates of change between treatment groups was found to be related to the baseline mean retinopathy level (P = 0.031), but less significantly so if baseline retinopathy keyed to the worse eye was used as a covariate (P = 0.08). Worsening occurred more frequently in those patients starting with the lower retinopathy levels. Progression was associated with the appearance of retinal infarcts (cotton-wool spots, soft exudates) and/or intraretinal microvascular abnormalities, with the pump patients showing a significant increase in these individual retinal lesions compared with the conventionally treated patients over 8 mo (P less than 0.025).

Conference on insulin pump therapy in diabetes. Multicenter study of effect on microvascular disease. Assessment of fluorescein angiograms.

Quantitation of the earliest changes in abnormal retinal morphology using fluorescein angiography is potentially superior to retinal photography. However, the critical importance of flawless technique, limitations in the size of the field available for detailed study, and observer variability constitute major disadvantages. A protocol describing standards of photography and of injection was developed. Methodology for counting microaneurysms (Ma) was developed at a central laboratory and applied in suitable photographs obtained at 0 (baseline), 4, and 8 mo in 68 patients. Counts of "definite" or "possible" Ma were made on films projected under standard conditions by two observers known to achieve consistently reproducible results. Semiquantitative assessment of diffusibility of fluorescein reflecting capillary leakage was performed in 61 patients. Leakage was graded according to three degrees of severity permitting study of observer variation, concordance of change in pairs of eyes, and treatment effects. The average number of Ma at baseline was slightly higher in the continuous subcutaneous insulin infusion (CSII) group than in the conventional insulin treatment (CIT) group, but the difference was not statistically significant at the 0.05 level. At both 4 and 8 mo, definite Ma were more prevalent in the CSII group, and the difference was statistically significant at the 0.05 level using both parametric and nonparametric tests. In 27 CSII and 23 CIT patients having complete sets of 0, 4, and 8-mo photographs, Ma counts increased during the 0-4-mo interval. During the 4-8 mo interval, a further increase occurred in the CSII group but contrasted with a decrement observed in the CIT group, which showed no net change from baseline.(ABSTRACT TRUNCATED AT 250 WORDS)

Florid diabetic retinopathy and its response to treatment by photocoagulation or pituitary ablation.

A group of 34 patients with florid diabetic retinopathy are reported. Nine of these had at least one eye untreated and of these only two maintained vision at one year-the others were blind. Ten patients had 11 florid eyes treated by photocoagulation. At one year six were blind and five had good vision. At two years only three could still see. Of 20 pituitary-ablated patients with 29 florid eyes, only three were blind at one year. Even at five years, 12 of 17 eyes could see and only two patients were blind. The one- and two-year visual acuities were significantly better in the pituitary ablated eyes than in the untreated and photocoagulated ones (p = 0.01-0.03). It is suggested that for this rare form of retinopathy pituitary ablation remains the treatment of choice if vision is to be maintained.

Production of endothelin 1 by cultured bovine retinal endothelial cells and presence of endothelin receptors on associated pericytes.

Endothelinlike immunoreactivity was detected by radioimmunoassay in medium conditioned by cultured endothelial cells obtained from bovine retinal microvessels (9.2 +/- 6.5 pM, n = 4). Sephadex G-25 column chromatography and fast-protein liquid chromatography revealed that most of the endothelinlike immunoreactivity was eluted in an identical position to synthetic endothelin 1. Retinal capillary pericyte-conditioned medium contained 2.9 pM endothelinlike immunoreactivity. In contrast to endothelial cells, retinal pericytes were found to bind endothelin. The dissociation constant and binding capacity were 0.14 nM and 1.5 x 10(5) sites/cell (n = 3), respectively. These findings suggest that endothelin produced by the retinal endothelial cells binds to the pericytes, adding support to the suggestion that pericytes in the retina may have a musclelike function.

The effect of aminoguanidine and tolrestat on glucose toxicity in bovine retinal capillary pericytes.

Cultured bovine retinal capillary pericytes (BRP) were used to investigate the effect of an aldose reductase inhibitor, tolrestat, and an inhibitor of advanced glycation end products (AGE) formation, aminoguanidine, on glucose toxicity. Glucose at high concentration reduced the replicative activity of pericytes in a dose-dependent manner. Tolrestat completely inhibited the production of sorbitol in cells exposed to a high concentration of glucose but failed to protect the cells from glucose toxicity. These results suggest that sorbitol accumulation in cells is probably not the major mechanism for glucose toxicity. In contrast, the addition of aminoguanidine at 10 mM concentration to the culture media protected pericytes from glucose toxicity. The degree of protected pericytes from glucose toxicity. The degree of protection was dose-dependent and evident at aminoguanidine concentration as low as 1 mM. The drug was only slightly toxic to BRP but induced morphological changes in pericytes with the loss of cellular processes and decreased cell spreading. This may suggest some action of aminoguanidine on the pericyte cytoskeleton. High concentration of glucose significantly increased the level of early glycation but not fluorescent AGE formation on BRP proteins. This was inhibited by the addition of aminoguanidine suggesting that glycation of cellular/membrane proteins and other mechanisms may play an important role in the toxic action of high glucose levels in cultured pericytes.

Activity of the glycosylating enzyme, core 2 GlcNAc (beta1,6) transferase, is higher in polymorphonuclear leukocytes from diabetic patients compared with age-matched control subjects: relevance to capillary occlusion in diabetic retinopathy.

The exact mechanism for capillary occlusion in diabetic retinopathy is still unclear, but increased leukocyte-endothelial cell adhesion has been implicated. We examined the possibility that posttranslational modification of surface O-glycans by increased activity of core 2 transferase (UDP-Glc:Galbeta1-3GalNAcalphaRbeta-N-acetylglucoaminyltr ansferase) is responsible for increased adhesion of leukocytes to vascular endothelium in diabetes. The mean activity of core 2 transferase in polymorphonuclear leukocytes isolated from type 1 and type 2 diabetic patients was higher compared with age-matched control subjects (1,638 +/- 91 [n = 42] vs. 249 +/- 35 pmol x h(-1) x mg(-1) protein [n = 24], P = 0.00013; 1,459 +/- 194 [n = 58] vs. 334 +/- 86 [n = 11], P = 0.01). As a group, diabetic patients with retinopathy had significantly higher mean activity of core 2 transferase compared with individuals with no retinopathy. There was a significant association between enzyme activity and severity of retinopathy in type 1 and type 2 diabetic patients. There was a strong correlation between activity of core 2 transferase and extent of leukocyte adhesion to cultured retinal capillary endothelial cells for diabetic patients but not for age-matched control subjects. Results from transfection experiments using human myelocytic cell line (U937) demonstrated a direct relationship between increased activity of core 2 transferase and increased binding to cultured endothelial cells. There was no relationship between activity of core 2 transferase and HbA(1c) (P = 0.8314), serum advanced glycation end product levels (P = 0.4159), age of the patient (P = 0.7896), and duration of diabetes (P = 0.3307). On the basis that branched O-glycans formed by the action of core 2 transferase participate in leukocyte adhesion, the present data suggest the involvement of this enzyme in increased leukocyte-endothelial cell adhesion and the pathogenesis of capillary occlusion in diabetic retinopathy.

Quantitative evaluation of fluorescein angiograms: microaneurysm counts.

This study describes a method for quantifying microaneurysms (MA) from fluorescein angiograms. The method was validated by the reproducibility of the number of MA in 30 angiograms read twice each by two independent observers; and by the absolute difference in MA counts between two readings by the same observer, and difference in numbers counted by two different observers. The precise location of each MA on two readings was also studied and the reproducibility of location varied from 60 to 71%, depending on the quality of the angiograms. Clinicians and technicians working in the same or in different centers obtained similar results. The coefficient of correlation between observers and between readings was satisfactory, r greater than 0.9. The method is easy to learn and the reproducibility allows for its use in clinical trials.

Genetic and immunologic factors in microvascular disease in type I insulin-dependent diabetes.

A detailed study of 133 subjects with insulin-dependent (type I) diabetes with severe microvascular disease has failed to substantiate the hypothesis that HLA factors influence the predisposition to this type of complication. A significant association between proliferative retinopathy and raised levels of circulating immune complexes was found. The distribution of insulin-binding levels in serum was similar to that in patients without complications. There was no correlation between insulin binding and the presence of immune complexes and no evidence was found that these complexes contained anti-insulin, anti-nuclear, or organ-specific antibodies. The distribution of insulin-binding levels in these subjects with diabetes of long duration was similar to that observed in 270 subjects with juvenile-onset short-duration type I diabetes. When the data were combined, significant associations between HLA-B8 and low/absent insulin binding levels were observed. HLA-BW62 was not associated with either high or low insulin-binding capacity. It is concluded that HLA genetic factors, insulin-binding capacity, and autoimmunity are unrelated to the pathogenesis of microvascular disease. Raised levels of circulating immune complexes may well be secondary to widespread tissue damage in diabetes of long duration.

Successful treatment of immune-mediated insulin resistance by human insulin (recombinant DNA).

An 82-yr-old woman with type II diabetes developed antibody-mediated insulin resistance while on mixed pork-beef insulin concomitantly with a non-Hodgkin lymphoma. Insulin resistance was initially treated with highly purified pork insulin, but this was unsuccessful. Treatment with human insulin (recombinant DNA) was associated with marked decrease of both insulin requirement and high-affinity antibodies, increase of free insulin levels, and improvement of diabetic control. This patient's case shows that human insulin can be considered as an alternative treatment for immune-mediated insulin resistance.

The relationship of hand abnormalities to diabetes and diabetic retinopathy.

The hands of 299 diabetic patients with and 161 without retinopathy were examined for abnormalities. Almost all abnormalities were finger joint contractures resulting in limited joint mobility (LJM) and/or Dupuytren's contractures (DC). Both LJM and DC occurred not only in insulin-dependent diabetes (IDDM) but also in non-insulin-dependent diabetes (NIDDM). In retinopathy patients LJM and DC occurred in 48% and 36% of patients, respectively, compared with 24% and 16% in those without retinopathy. These differences were statistically significant (P less than 0.001). The higher prevalence of LJM in the retinopathy group affected mainly those with severe retinopathy, there being no difference between background and nonretinopathy patients. DC was less clearly related to severe retinopathy. LJM was more severe in those with than without retinopathy. LJM and DC were also related to age and duration of known diabetes. Subgroups matched for age and duration of known diabetes showed that the main relationship of hand abnormalities was to retinopathy in IDDM, but more to age and duration of known diabetes in NIDDM.

Changes in insulin resistance with long-term insulin therapy.

Thirteen newly diagnosed diabetic subjects, 5 with insulin-dependent diabetes mellitus (IDDM) and 8 with non-insulin-dependent diabetes mellitus, mean age 37.1 yr (range 25-64 yr), underwent glucose-clamp studies at diagnosis of diabetes at plasma glucose 200 mg/dl. Each subject was then treated twice daily with insulin for 6 mo with improvement in glycemic control, and the glucose-clamp studies repeated. Changes in glucose uptake at an insulin infusion rate of 1.0 mU X kg-1 X min-1 varied greatly from diagnosis to 6 mo. There were significant negative correlations between change in glucose uptake and diabetes type (r = -.78, P less than .002), C-peptide secretion (r = -.66, P less than .05), and age (r = -.62, P less than .05). At an insulin infusion rate of 10 mU X kg-1 X min-1 there was improvement in glucose uptake from diagnosis to 6 mo that did not reach statistical significance. During the steady-state periods of the glucose-clamp studies at diagnosis, growth hormone (GH) rose above basal, which reached statistical significance at the higher insulin infusion rate. This increase in GH was not apparent at the time of the glucose-clamp studies after insulin therapy. Our results indicate that in the clinical situation, only patients with IDDM can expect an improvement in their sensitivity to physiologic insulin levels with long-term insulin therapy. In all subjects, improvement in glycemic control leads to abolition of GH secretion in the presence of hyperglycemia.

Effect of meal on retinal blood flow in IDDM patients.

OBJECTIVE: The effect of postprandial hyperglycemia on retinal blood flow was examined in 11 adults with insulin-dependent diabetes mellitus (IDDM) and 9 nondiabetic subjects. RESEARCH DESIGN AND METHODS: Retinal blood flow was measured with bidirectional laser Doppler velocimetry before and 60 and 120 min after a 50-g carbohydrate breakfast. Diabetic subjects were studied twice, being randomized to take their usual insulin (isoglycemic study) or to omit their morning short-acting insulin (hyperglycemic study). Nondiabetic subjects were studied once and received no insulin. RESULTS: Plasma glucose rose significantly (from 11.5 +/- 5.2 to 22.1 +/- 5.5 mM, P less than 0.001) during the hyperglycemic study but not during the isoglycemic or nondiabetic studies. There were no significant changes in blood pressure or retinal blood flow in any of the groups studied. CONCLUSIONS: Short-term hyperglycemia does not increase retinal blood flow.

Vitreous fluorophotometry in patients with no or minimal diabetic retinopathy.

Vitreous fluorophotometry was used to estimate fluorescein leakage into the posterior vitreous of 20 insulin-dependent diabetic patients with no or minimal diabetic retinopathy 60 min after intravenous administration of 14 mg X kg-1 fluorescein. The permeability coefficient (p), a measure of fluorescein penetration through the blood-retinal barrier (BRB) into the vitreous, and the diffusion coefficient (D), a measure of fluorescein dispersion within the vitreous, were obtained by fitting a mathematical model to the vitreous fluorescence scan and plasma free fluorescence curve. A permeability index (PI) was also derived by dividing the area under the fluorescence scan by the area under the plasma free fluorescence time curve. The fluorescence concentrations at discrete distances from the retina were also noted. The mean +/- SD for p, D, and PI were 1.95 +/- 1.03 cm X s-1 X 10(-7), 1.74 +/- 1.53 cm2 X s-1 X 10(-5), and 2.14 +/- 1.21 cm X s-1 X 10(-7), respectively, and were not significantly different from values determined in normal subjects. Diabetic patients and normal subjects also had similar fluorescence measurements at corresponding distances from the retina. Quantitative indices of fluorescein leakage did not correlate with either the microaneurysm counts on fluorescein angiograms or the duration of diabetes. Vitreous fluorophotometry did not detect any abnormality of the BRB in diabetic patients with no or minimal retinopathy on fluorescein angiography.

Risk factors associated with severe proliferative retinopathy in insulin-dependent diabetes mellitus.

The natural history of disease and suspected risk factors for bad prognosis were investigated in 40 subjects with insulin-dependent diabetes mellitus who had severe retinopathy and in 22 patients with a similar duration of diabetes without evidence of complications. The retinopathy group showed a marked excess of men (ratio 2:1). Examination of the data in the literature also showed a striking excess of men, 61% (P less than 0.001) in patients with insulin-dependent diabetes and severe microvascular disease. In addition, proliferative retinopathy was found to have significant associations with current poor diabetic control, hypertension, and previous treatment with once-daily insulin regimens, particularly with protamine zinc insulin.

Continuous subcutaneous insulin infusion does not provoke significant acute-phase response.

C-reactive protein (CRP), the classical acute-phase reactant, and serum amyloid A protein (SAA), the putative precursor of AA-type amyloid fibrils, were measured in 62 diabetic patients. They were all attending their regular clinic appointments and had been asymptomatic during the 2 wk preceding sampling. CRP and SAA levels were similar in 18 patients on continuous subcutaneous insulin infusion (CSII), 27 patients treated by conventional insulin therapy (CIT), nine treated by diet only, and eight treated by diet and oral hypoglycemic agents, and were almost entirely within the normal range. It is concluded that CSII does not provoke an acute-phase reaction in diabetic patients and, while caution should always be exercised with a new form of treatment, it does not seem likely that CSII will predispose to the development of reactive systemic amyloidosis.

Markers of insulin resistance are strong risk factors for retinopathy incidence in type 1 diabetes.

OBJECTIVE: To determine the incidence of retinopathy and the relative importance of its risk factors in type 1 diabetes. RESEARCH DESIGN AND METHODS: This is a 7.3-year follow-up of 764 of 1,215 (63%) people with type 1 diabetes across Europe, aged 15-60 years at baseline with no retinopathy (the EURODIAB Prospective Complications Study). Retinal photographs were taken at baseline and follow-up and risk factors were assessed to a standard protocol. RESULTS: Retinopathy incidence was 56% (429/764, 95% CI 52-59%). Key risk factors included diabetes duration and glycemic control. We found no evidence of a threshold effect for HbA1c on retinopathy incidence. Univariate associations were observed between incidence and albumin excretion rate, cholesterol, triglyceride, fibrinogen, von Willebrand factor, gamma-glutamyltransferase, waist-to-hip ratio, and insulin dose. No associations were observed for blood pressure, cardiovascular disease, or smoking. Independent risk factors, as assessed by standardized regression effects, were HbA1C (1.93, P = 0.0001), duration (1.32, P = 0.008), waist-to-hip ratio (1.32, P = 0.01), and fasting triglyceride (1.24, P = 0.04). CONCLUSIONS: Retinopathy incidence in type 1 diabetes remains high. Key risk factors include diabetes duration and glycemic control, with no evidence of a threshold for the latter. Other independent risk factors, such as waist-to-hip ratio and triglyceride levels, both markers of insulin resistance, were strongly related to incidence.