Author Affiliation: Jaeb Center for Health Research, Tampa, Florida.
With the recent increases and future projected increases in the incidence of type 2 diabetes mellitus and with the incidence increasing in teenagers and young adults, the already substantial public health effect of diabetes and diabetic retinopathy will become greater in years to come. Despite the strength of the evidence that optimizing control of glucose, blood pressure, and lipid levels will reduce the incidence and progression of diabetic retinopathy, metabolic control remains suboptimal for many patients with diabetes. In addition, many patients do not follow recommended guidelines for regular eye examinations, which is unfortunate because there is good evidence that with regular follow-up and intervention with photocoagulation as indicated, severe vision loss from diabetic retinopathy is uncommon. Yet, diabetic retinopathy is a leading cause of severe vision loss in adults. The current health care system too often fails to adequately manage diabetes and is lacking in providing proper education and motivation for patients to optimize their metabolic control. In addition to treating retinopathy, ophthalmologists can play an important role in educating and motivating patients to achieve better metabolic control, which, if successful, potentially could do more to reduce the progression of retinopathy than any of the ocular treatments currently in the armamentarium of the ophthalmologist.
The increase in obesity has been considerable and, as a consequence, there has been a rise in the number of cases of diabetes in the United States during the past 10 to 20 years. The prevalence of diabetes, approximately 95% of which represents type 2 diabetes mellitus (DM) and approximately 5% type 1 DM, is estimated to be more than 23.6 million people in the United States (approximately 8% of the population), with 5.7 million of these people representing undiagnosed cases.1 Another 57 million people are considered to have prediabetes, meaning that their glucose levels are not normal but are not sufficiently abnormal for a diagnosis of type 2 DM. It is projected that the prevalence of diabetes will exceed 30 million in the United States by 2030.2
The prevalence of type 2 DM varies according to race and ethnicity, being highest in black individuals and Native Americans, lowest in Asian Americans, and midrange in Hispanic and non-Hispanic white persons.1 Of particular concern is the fact that type 2 DM is developing with increasing frequency at younger ages, attributable to an increase in obesity in youth.3 The Centers for Disease Control and Prevention has estimated the prevalence to be approximately 0.2%.1 Although this sounds small, it represents approximately 186 000 cases and has grown considerably in the past 20 years, particularly in minority groups.
As a consequence of the increasing incidence of diabetes, a substantial increase in diabetic retinopathy and other complications of diabetes can be expected. It has been estimated that the number of Americans older than 40 years with diabetic retinopathy will increase from a 2005 estimate of 6.7 million to 19.4 million by 2050.4 With the incidence of type 2 DM in those younger than 40 years increasing at a rapid rate, undoubtedly ophthalmologists will begin seeing more and more patients with diabetic retinopathy at younger and younger ages.
The ophthalmologist is in a position to be able to influence a patient's diabetes self-management. Thus, it behooves the ophthalmologist to be well informed about diabetes management and the impact of glycemic control, blood pressure control, and lipid levels on the risk and progression of retinopathy. I review the evidence in the literature that supports an aggressive approach to metabolic control as a means to reduce the incidence and progression of diabetic retinopathy.
Unequivocally, the development and progression of diabetic retinopathy are strongly related to the degree of glucose control. Glycemic control can be assessed by measuring the glycated hemoglobin (HbA1c) level in the blood, which provides an indication of the degree of glycemic control during the previous 2 to 3 months; in people without DM, the level is less than 6.0%. The association between a reduction in the HbA1c level and a reduced risk of retinopathy development or progression has been demonstrated conclusively for type 1 DM in the Diabetes Control and Complications Trial (DCCT)5,6 and for type 2 DM in the United Kingdom Prospective Diabetes Study (UKPDS).7,8 Much also has been learned from other studies, particularly the Wisconsin Epidemiology Study of Diabetic Retinopathy.9,10 The effect of glycemic control on kidney disease, the other major type of microvascular disease, parallels the effect on retinopathy. Recently, there has been controversy about a possible increase in macrovascular disease (eg, myocardial infarction and stroke) with intensive glucose lowering in type 2 DM.11,12 However, Ray et al13 in a meta-analysis concluded that intensive glycemic control does not increase the risk of cardiac events, stroke, or death. Ray et al13 did note that the optimal approach to achieving good glycemic control might be different in differing populations.
The DCCT,5 funded by the National Institutes of Health, included 1441 patients with type 1 DM who ranged in age from 13 to 39 years. Approximately half these patients had preexisting retinopathy and half had no retinopathy. Patients were randomized to receive either 1 or 2 daily injections of insulin (conventional care at the time) or intensive therapy involving use of an insulin pump or multiple daily injections of insulin. The mean HbA1c level was approximately 8.8% at baseline and averaged 7.2% in the intensive treatment group and 9.1% in the control group during mean follow-up of 6.5 years. This improvement in glycemic control in the DCCT intensive treatment group had a substantial effect on the risk of retinopathy. In the patients who did not have retinopathy at baseline, intensive therapy reduced the risk of retinopathy by 76% from a 3-step change in retinopathy severity of 4.7 per 100 person-years of follow-up to 1.2. In patients with retinopathy at baseline, intensive therapy compared with conventional therapy substantially reduced the progression of retinopathy (3.7 vs 7.8 events per 100 patient-years), the development of proliferative or severe nonproliferative disease (1.1 vs 2.4 events per 100 patient-years), and the development of macular edema (2.0 vs 3.0 events per 100 patient-years). Across both treatment groups, there was a strong association between HbA1c levels and the rate of progression of retinopathy and nephropathy. For each 10% relative decrease in HbA1c (eg, from 9.0% to 8.1%), there was approximately a 40% reduction in the progression of retinopathy.6
When the DCCT results were unmasked, the patients in the control group were offered intensive therapy, and the study continued in follow-up with a new study name: the Epidemiology of Diabetes Interventions and Complications.14 What has been remarkable is that although both treatment groups from that point forward were being treated the same and had HbA1c levels that were similar (a mean of approximately 8.0%), the treatment group differences in the incidence and progression of retinopathy have persisted. In the most recent report from the DCCT/Epidemiology of Diabetes Interventions and Complications,14 after mean follow-up of 18.5 years, there were still substantial treatment group differences in the development of proliferative retinopathy, the development of clinically significant macular edema, and the receipt of scatter photocoagulation. In addition, across both treatment groups, there were few patients with substantial visual acuity loss, a marked contrast with earlier studies. Only 5 of 1441 patients lost visual acuity to worse than 20/100 in either eye, and in only 2 of the 5 was the loss attributable to diabetic retinopathy.15
There was a downside to the tighter glycemic control in the DCCT. The rate of developing severe hypoglycemic events increased 3-fold from 19 events per 100 person-years in the control group to 62 events in the intensive treatment group.5 With newer insulins and advancements in home glucose monitoring, including better blood glucose meters and the advent of continuous glucose monitoring, the rate of severe hypoglycemia at low HbA1c levels is substantially lower than it was in the DCCT. A recent clinical trial16 evaluating continuous glucose monitoring reported a rate of severe hypoglycemia of less than 10 events per 100 person-years in adults with type 1 DM compared with 62 events per 100 person-years in the DCCT; the lower rate occurred at a mean HbA1c level of 6.9%, which was even lower than that in the DCCT intensive treatment group.
One interesting sidelight is that approximately 10% of the patients in the DCCT intensive treatment group had an initial transient worsening of retinopathy, although it did not affect visual acuity.17 This worsening consisted of cotton-wool patches and intraretinal microvascular abnormalities. It was more likely to occur when retinopathy was already present, occurring in approximately 1% of patients with no retinopathy and in 48% with retinopathy. This finding is not unique to the DCCT. Another study18 reported this phenomenon as well. The importance of this phenomenon for the physician is that in patients who have severe nonproliferative or proliferative retinopathy, it may be wise to treat this with scatter photocoagulation before beginning intensive therapy and to monitor closely when intensive therapy is initiated.
The UKPDS7,8 included 4209 participants with type 2 DM enrolled between 1977 and 1991 at 23 sites. The patients (median age, 54 years) were randomized to medical treatment with oral agents or insulin or to diet. The mean HbA1c level at baseline was 7.1%. In the medical group, the mean HbA1c level initially decreased to 6.2% and then gradually increased, whereas in the control group it stayed at approximately 7% for 3 years and then increased. During follow-up, HbA1c averaged 7.0% in the medical treatment group and 7.9% in the control group.
After 12 years, there was a 21% relative reduction in a 2-step progression of retinopathy (49% vs 39%). A risk reduction was seen in those with and without retinopathy at study entry. Similar to the DCCT, a strong association was seen between HbA1c level and risk of retinopathy development or progression. This was true for eyes with and without retinopathy at baseline.
This association between retinopathy and HbA1c level also has been shown in population-based studies such as the Wisconsin Epidemiology Study of Diabetic Retinopathy.9 This long-term study of 2990 patients with type 1 or type 2 DM began in 1980. The mean baseline HbA1c level was substantially higher (approximately 10.0%) than that in the DCCT or the UKPDS.
Although the evidence of a beneficial effect of blood pressure control is not as strong as it is for glycemic control, a variety of studies have demonstrated that elevated blood pressure augments the risk of retinopathy in patients with diabetes.10,19,20 The strongest evidence comes from the UKPDS. The UKPDS included a substudy that evaluated blood pressure reduction with either the angiotensin-converting enzyme inhibitor captopril or the β-blocker atenolol compared with a control group.21,22 There was a mean systolic blood pressure difference of approximately 10 mm Hg between the treated and control groups. Both drugs were associated with a lower retinopathy progression rate (development/progression of diabetic retinopathy: 51% in the control group, 31% with captopril, and 37% with atenolol). The reduction in risk was greater in those who at baseline had no retinopathy than in those who already had retinopathy.
Angiotensin-converting enzyme inhibitors or blockers have been shown to have a beneficial effect on retinopathy aside from lowering blood pressure, although results across studies have not been fully consistent.23- 27 These drugs often are prescribed as prophylaxis against renal disease, but they seem to have a beneficial effect on retinopathy as well.
The effect of lipid levels on diabetic retinopathy has not been studied as extensively as has glycemic control and blood pressure control. Studies28- 31 that have evaluated the association have produced mixed results. The strongest evidence of an association comes from the Early Treatment Diabetic Retinopathy Study, which measured lipid levels in 2709 patients.32 Participants with low-density lipoprotein cholesterol levels of at least 160 mg/dL (to convert to millimoles per liter, multiply by 0.0259) were 2 times more likely to develop hard exudates than were those with low-density lipoprotein cholesterol levels less than 130 mg/dL.
Data from the National Health and Nutrition Examination Surveys indicate that the answer is not often enough. Goals for HbA1c (<7.0%), blood pressure (systolic <130 mm Hg and diastolic <80 mm Hg), and cholesterol (total cholesterol <200 mg/dL) levels were achieved by only 37%, 36%, and 48% of patients, respectively, and only 7.3% attained the recommended levels for all three.33
Why are these results so poor when the evidence is so convincing that better metabolic control has a substantial beneficial effect on retinopathy? I think that there are 2 overarching reasons. First, diabetes can be a difficult disease to control, particularly type 1 DM. It is not as simple as taking a pill or an injection of insulin. In type 1 DM, good control requires frequent glucose monitoring and continual adjustments in insulin, diet, and exercise; and in type 2 DM, it takes a complete lifestyle change to optimize control, usually including weight loss, which is difficult for many patients to achieve. So, that is part of it. But the other part is that patients are just not well enough informed and not sufficiently self-motivated to optimize their control. In a study of patients with diabetes attending a retina clinic, only approximately half knew what HbA1c was, and only approximately a third knew their HbA1c level.34
Diabetes is a disease in which patient education is a critical component of management. However, patient education is not well reimbursed for providers and is given short shrift in many situations. Too many patients and physicians too readily accept blood glucose levels in the range of 200 to 400 mg/dL (to convert to millimoles per liter, multiply by 0.0555).
What should be the role of the ophthalmologist in this systemic disease in which one of the most important complications is ocular? Although it is easy to say that patient education should be the responsibility of the primary care provider or endocrinologist, the eye care provider is in a position to potentially influence a patient's diabetes management. Loss of vision is a feared complication for many patients with diabetes. So, the ophthalmologist is in a position to potentially empower a patient to achieve better control.
Because of the potential effect that ophthalmologists could have, the Diabetic Retinopathy Clinical Research Network will be conducting a study to assess whether an educational intervention in the retina specialist's office can improve HbA1c control. Patients with diabetes will be randomly assigned to receive either an educational intervention or usual care. The educational intervention will include measurement of HbA1c levels while the patient is in the office, a personalized risk assessment to show the patient his or her risk of vision loss from retinopathy, and educational materials about managing diabetes. The outcomes will be HbA1c levels after 1 and 2 years and progression of retinopathy.
There have been important changes in the management of diabetes during the past 30 years. Despite this, many patients still spend substantial portions of the day with elevated glucose levels, and the incidence of retinopathy is unacceptably high. Although diabetic retinopathy remains a leading cause of severe vision loss, data from the DCCT showing a very low risk of severe vision loss, and similar data from the Joslin Diabetes Center (Lloyd P. Aiello, MD, PhD, oral communication, October 2009), indicate that patients who regularly have eye examinations generally will do well, although many will develop retinopathy requiring treatment. Unequivocally, the better the glycemic control, the lower the risk of retinopathy and nephropathy, and the more controlled hypertension is, the lower the risk. Control of lipid levels is less clear-cut but probably also is beneficial. Despite the strength of the evidence that optimizing metabolic control reduces the complications of diabetes and has long-term benefit for patients and for overall public health by reducing health care costs, the current health care system too often does not adequately manage diabetes and is lacking in properly educating and motivating patients to optimize their metabolic control.
With the marked increase in the incidence of type 2 DM in the past decade with no end in sight and with the incidence increasing in teenagers and young adults, the public health impact of diabetes, which is already great, will become even greater in years to come. Beyond retinopathy treatment alone, ophthalmologists can play an important role in educating and motivating patients to achieve better metabolic control. Achieving better metabolic control, particularly if initiated soon after diagnosis of diabetes, potentially could do more to reduce the progression of retinopathy than any of the ocular treatments currently in the armamentarium of the ophthalmologist. Ophthalmologists, after all, are physicians, and diabetes, more than any other disease, provides the ophthalmologist with the opportunity to affect a patient's overall health.
Correspondence: Roy W. Beck, MD, PhD, Jaeb Center for Health Research, 15310 Amberly Dr, Ste 350, Tampa, FL 33647 (firstname.lastname@example.org).
Submitted for Publication: December 10, 2009; final revision received January 26, 2010; accepted January 27, 2010.
Financial Disclosure: None reported.
Previous Presentation: This paper was presented as the Alfred W. Bressler Prize in Vision Science Lecture at the Bressler Symposium; October 17, 2009; New York, New York.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Ophthalmology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 3
Customize your page view by dragging & repositioning the boxes below.
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
Diabetes, Foot Ulcer
The Rational Clinical Examination: Evidence-Based Clinical Diagnosis
Diabetic Peripheral Neuropathy
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.