I WHAT CAN BE DONE TO PREVENT ONSET AND PROGRESSION OF DIABETIC RETINOPATHY?
Medical treatment can not only prevent diabetic retinopathy but can ameliorate its progression, if retinopathy develops. To be successful, medical treatment requires a well-informed and empowered patient. This is because not only is good compliance with any medication required, but also compliance with life style issues such as diet and exercise.
1 Tight Blood Pressure Control
Hypertension has been shown in many studies to be more prevalent in patients with diabetic retinopathy and to be a risk factor for its development and progression. Early studies showed that in type 1 diabetes mellitus systolic blood pressure is associated with diabetic retinopathy whereas in type 2 diabetes mellitus it was believed that perhaps that only diastolic blood pressure was associated. The United Kingdom Prospective Diabetes Study reported recently, however, that the most important measure to prevent the progression of microvascular complications, and in particular retinopathy, in patients with type 2 diabetes mellitus was tight systolic and diastolic blood pressure control. Indeed many of the features of hypertensive retinopathy- flame haemorrhages, cotton wool spots – cannot be differentiated from diabetic retinopathy. In young patients, with type 1 diabetes mellitus, blood pressure readings <141/91mmHg but greater than the 90th centile for age have also been found to be significantly associated with both an increased incidence and progression of retinopathy suggesting patients with type 1 diabetes mellitus also need more stringent blood pressure targets than the general population.
2 Good Control of Blood Sugars
1993 saw the publication of the definitive evidence that tight metabolic control reduced the onset and progression of retinopathy, nephropathy and neuropathy in patients with type 1 diabetes mellitus10
Evidence that metabolic control was important first came to light in 1977 with the publication of Pirart’s cohort study of four thousand diabetic patients followed for up to twenty-five years. The prevalence of microvascular complications was much higher in patients with “poor” control. Around the same time it was shown that microaneurysms appeared more slowly in type 1 diabetes mellitus patients treated with several daily insulin injections than in those on one or two daily injections.
Prior to the advent of home blood glucose monitoring and measurement of HBA1C in the late 1970’s it had not been possible for patients or doctors to achieve or define easily, good control. Before then home monitoring relied on urine testing and clinic monitoring was dependent on random blood glucose measurements. Urine testing was very inaccurate as the amount of glucose in the urine (glycosuria) depended on the renal threshold for glucose, which varies greatly between individuals. Furthermore, the amount of glycosuria had no correlation with the blood sugar at that moment in time. The development of home blood glucose monitoring and measurement of HBA1C in the late 1970’s meant that the measurement of fluctuations in day to day blood glucose levels and average blood glucose levels was now possible. Prior to the development of home blood glucose monitoring patients with type 1 diabetes mellitus tried to achieve good metabolic control by increasing their insulin dosage until they developed hypoglycaemic symptoms. This was rather hazardous. These new developments meant that intense metabolic control was more feasible and led to several small trials looking at the impact of good glycaemic control on the complications of diabetes. All showed that good glycaemic control had a beneficial impact and this was confirmed in a meta-analysis by Wang, which was published at the same time as the Diabetes Control and Complications Trial in 1993.
The Diabetes Control and Complications Trial employed almost two investigators for every patient recruited. The Primary Prevention cohort consisted of 725 patients with type 1 diabetes mellitus who had no retinopathy at recruitment. Their average duration of diabetes was 2.6 years. The Secondary Prevention cohort consisted of 715 patients with mild non-proliferative diabetic retinopathy and an average duration of diabetes of 8.7 years. Patients were randomly allocated to three or more insulin injections per day (intensive control) or one or two injections per day (conventional control). Day to day adjustments were made to the insulin regimen of patients in the intensive group by telephone. Patients were followed for an average of 6.5 years with 99% completing the study. HBA1C was 7% on average in the intensive group and 9% in the conventional group. Less than 5% of the intensive group maintained the target of a normal HBA1C (6.05%) throughout the study. As stated earlier the results were very convincing with the risk of diabetic retinopathy development or progression decreased by 76% in the primary prevention cohort treated with intensive therapy and 63% in the secondary prevention cohort treated similarly. It is important to note that there was no absolute HBA1C level below which no retinopathy occurred or progressed. Furthermore, a reduction in HBA1C by, for example 1% was of greater benefit the higher the initial level suggesting that in those with very poor control a modest improvement would reap proportionally large dividends. These benefits, however, were obtained at a cost, with the incidence of severe hypoglycaemia increasing more than threefold in the intensively treated cohort. Not all patients will be able to achieve the level of control of the intensively treated cohort and insulin regimens will have to be tailored according to the abilities and expectations of individual patients.
In 1998 the main outcomes of the United Kingdom Prospective Diabetes Study were published. This twenty year study, with twenty three participating centres, looked at the role of intensive blood-glucose control with sulphonylureas or insulin compared with conventional therapy in patients with type 2 diabetes mellitus. 5102 patients were recruited into this study. They were initially managed with a three month dietary run-in and were then stratified by body weight. Non-overweight patients were then randomly assigned to intensive treatment with insulin (30%), intensive treatment with sulphonylurea (40%) or conventional treatment with diet (30%). Overweight patients were randomly assigned treatment with the additional possibility of metformin: intensive treatment with insulin (24%), intensive treatment with sulphonylurea (32%), intensive treatment with metformin (20%) or conventional treatment with diet (24%). Over ten years the intensive treatment group had a 25% reduction in microvascular complications, most of which was due to fewer patients requiring photocoagulation. Concerns expressed by earlier studies that cardiovascular mortality might be increased by sulphonylureas or that hyperinsulinaemia might accelerate atherosclerosis were also importantly allayed.
3 Warning: Rapid Tightening of Control May Cause Worsening of Diabetic Retinopathy
Early evidence that rapid tightening of metabolic control might cause worsening of diabetic retinopathy was noted in the following studies:
- Steno Study Group
- KROC Collaborative Study Group
- Oslo Study Group
The Diabetes Control and Complications Trial Research Group showed conclusively that this was a real phenomenon. Indeed in the first two years of the trial patients with mild or moderate non-proliferative retinopathy treated intensively appeared to be deteriorating more rapidly compared to those treated with conventional “poorer” control. At this point the trial was almost halted. Thankfully this did not happen and in the subsequent years the intensively treated group did far better than the conventionally treated group.
In 1998 the DCCT group reported formally on “early worsening of retinopathy”. For this study worsening was defined as:
- progression of 3 steps or more on the Early Treatment Diabetic Retinopathy Study final scale
- the development of soft exudates and/or intraretinal microvascular abnormalities
- the development of clinically important retinopathy
- any of the above
It was considered “early” if it occurred between the baseline and 12-month follow-up visits.
Early worsening occurred twice as frequently in patients assigned to intensive treatment. 50% of patients recovered from early worsening of their retinopathy. Twice as many patients with early worsening continued to progress after the 18-month visit.
However, the overall outcome was still better in intensively treated patients who experienced early worsening than in conventionally treated patients who did not.
The most important risk factors for early worsening were:
- the degree of inadequate control at baseline
- the fall in HBA1C
Longer duration of diabetes was also a risk factor. More severe retinopathy was a risk factor only when early worsening was defined as development of cotton-wool patches or intra-retinal microvascular abnormalities.
Most of the early worsening did not reach a clinically important level (defined as proliferative or severe non-proliferative retinopathy or clinically significant macular oedema). Among intensively treated patients, worsening of this degree was observed in none of 378 patients with no retinopathy at baseline, in 2 of 249 who had microaneurysms only, in 6 of 82 who had mild non-proliferative diabetic retinopathy, and in 6 of 32 who had moderate non-proliferative diabetic retinopathy. High-risk proliferative diabetic retinopathy developed in 3 eyes, all of which responded well to photocoagulation, and macular oedema accompanied by a small decrease in visual acuity occurred in 6 eyes, all of which recovered to an acuity of 20/30 or better.
It is important to remember that the DCCT did not have patients with worse than moderate non-proliferative diabetic retinopathy. It is difficult therefore to generalise the results to patients with more severe levels of retinopathy. Nevertheless based on looking at all the evidence the DCCT investigators concluded that their results were consistent with previous reports of the occurrence of sight-threatening worsening when intensive treatment is initiated in patients with long-standing poor glycaemic control, particularly if retinopathy is at or past the moderate non-proliferative stage.
They recommended monitoring before initiation of intensive treatment and at 3-month intervals for 6 to 12 months. In patients whose retinopathy is already approaching the high-risk stage (very severe non-proliferative diabetic retinopathy or early proliferative diabetic retinopathy), they felt it may be prudent to delay the initiation of intensive treatment until photocoagulation can be completed, particularly if HBA1C level is high and a large reduction is anticipated.
4 Treatment of Hyperlipidaemia
The Wisconsin trial showed that there is an association between serum cholesterol and diabetic retinopathy. Furthermore broad spectrum lipid-modulating drugs, such as clofibrate, have been shown to reduce hard exudate formation in patients with diabetic retinopathy, but without any effect on visual outcome. Perhaps this is asking too much as even the aim of laser treatment in such cases is merely to maintain vision and in practice it is only the rare patient with laser treatment that gains an improvement in vision. At the time anxiety over the risks of lipid-lowering therapy led to the abandonment of such lines of treatment. More recent data suggests that cholesterol lowering agents may again have a role in exudative macular oedema. To date such research has not been translated into clinical practice and this may reflect the fact that laser therapy is most efficacious in this form of macular oedema.
5 Correction of Anaemia
More recently, anaemia (reduced haemoglobin levels leading to a reduced oxygen carrying capacity) has been recognised as a risk factor. This is presumably relates to the subsequent hyperdynamic state leading to increased shearing stress on blood vessels. It may also contribute to tissue hypoxia at a local level.
6 Smoking
Smoking has been shown to be associated with the onset of diabetic retinopathy but whether it has any effect on its progression is unclear. There is no doubt it is a significant risk factor for the macrovascular complications and as ischaemic heart disease is the main cause of mortality in both men and women with diabetes, it should be strongly discouraged.
7 Other Pharmacological Interventions
Although the role of many drugs has been looked at with regard to diabetic retinopathy none have been shown to make a significant difference. The ETDRS looked at the role of.htmirin in preventing the development of proliferative diabetic retinopathy in patients with type 1 diabetes mellitus. Although.htmirin had no beneficial effect the authors reported that.htmirin had no harmful effects and did not cause more severe or longer lasting vitreous haemorrhage in patients with proliferative diabetic retinopathy. This latter finding is extremely important in view of the crucial role that.htmirin has in the modern management of ischaemic heart disease and cerebrovascular disease- common complications in diabetes. .htmirin and ticlopidine have been shown to reduce the rate of microaneurysm formation in type 1 diabetes mellitus although the effect was minimal. Aldose reductase inhibitors have not been shown to have any benefit despite claims of efficacy in peripheral neuropathy.
Recently results from the EUCLID trial (EURODIAB controlled trial of lisinopril in diabetic retinopathy) have suggested that the ACE inhibitor lisinopril appears to delay the progression of diabetic retinopathy in normotensive, normoalbuminuric patients. 354 patients with type 1 diabetes mellitus were studied. Retinopathy was graded using two 45( photographs according to the EURODIAB method. 80% of the placebo group and 79% of the lisinopril group had either no retinopathy or mild non-proliferative diabetic retinopathy. During the trial period 23% of the placebo group and 13% of the lisinopril group progressed by at least one stage of retinopathy (p= 0.02) reflecting a reduction in the risk of progression by 50% in those patients on lisinopril. This effect was claimed to be independent of blood pressure. The main criticism of this paper lies in the fact that patients on lisinopril had significantly lower baseline HBA1c compared to controls (6.9% c.f. 7.3%; p = 0.05), which may have explained the greater progression of retinopathy in the control group. Until recently the role of blood pressure lowering agents in diabetic retinopathy was very unclear. The UKPDS has shown that at least for type 2 diabetes, tight blood pressure control is very important, regardless of whether an ACE inhibitor or a Beta blocker is used, in preventing the progression of diabetic retinopathy. In the same way that tight metabolic control has been shown to be beneficial to patients with both type 1 and type 2 diabetes mellitus, it is likely that future studies will confirm that tight blood pressure control will also be beneficial to both types of diabetes.
One of the reasons that published pharmacological interventions have not found much success must lie in the slow progression of diabetic retinopathy in its early stages. To detect change perhaps requires more sensitive techniques than the current Airlie-House method or extremely large numbers such as the 1441 patients followed in the Diabetes Control and Complications Trial for ten years.
8 Special Circumstances
PREGNANCY
There is an increased risk of progression of retinopathy during pregnancy. The presence of retinopathy alone is not associated with a poorer pregnancy prognosis for the foetus unless concurrent nephropathy is present. Poor glycaemic control in the first trimester and pregnancy-induced or chronic hypertension are independently associated with progression of retinopathy. Work by Temple et al 2001 suggests that the risk of progression is probably lower than previously documented. They attribute this to better glycaemic control afforded by such measures as HBA1C resulting in fewer women having any retinopathy at the time of pregnancy. Progression of retinopathy in pregnancy was uncommon (5.0% pregnancies) but was significantly more common in women with duration of diabetes > 10 years and in women with moderate to severe retinopathy at baseline. Laser therapy was needed in 2.2% pregnancies, which is much lower than that reported in earlier studies.
The exact reason for progression of retinopathy is not known but the following factors probably play a role:
- Hyperdynamic circulation
- Early worsening of retinopathy associated with tight glycaemic control
- Placental hormones
- Anaemia
The SIGN guidelines recommend:
- Fundal examination prior to conception and during each trimester.
- More frequent assessment may be required in those with poor glycaemic control or hypertension
- Early referral of pregnant women with moderate or worse retinopathy to an ophthalmologist is recommended due to the potential for rapid development of neovascularisation.
DIABETIC NEPHROPATHY
The presence of diabetic nephropathy appears to be a risk factor for the development of diabetic retinopathy. This may just reflect the fact that both are microvascular complications with similar aetiologies. Nephropathy can, however, lead to further contributing risk factors such as:
- Raised blood pressure
- Anaemia