Diabetic retinopathy is the most prevalent cause of legal blindness and visual loss among working-age Americans. A variety of lifestyle and dietary factors are contributing to an increased incidence of type 2 diabetes, which 90% of the overall population of people with diabetes have. The Centers for Disease Control and Prevention (CDC) recently reported that at least one-fourth of adult Americans are known to have prediabetes,1 a condition defined as having impaired fasting glucose, impaired glucose tolerance, or both. Persons with prediabetes are at increased risk for developing diabetes. The CDC estimates that 10% of Americans have diabetes.1
ETDRS STANDARD OF CARE
The development of diabetic retinopathy (DR) in type-2 diabetes is related to the duration of diabetes, severity of hyperglycemia, and degree of hypertension. Although proliferative diabetic retinopathy (PDR) develops in up to 25% of individuals who have had type 2 diabetes for at least 25 years, diabetic macular edema (DME) is the primary cause of visual loss for most people with diabetes.2 The pathogenesis of DME is multifactorial, involving the microangiopathic effects of hyperglycemia and hypertension upon the retinal vasculature. The net effect is retinal edema manifested as increased retinal thickness and visual loss. Fluorescein angiography demonstrates retinal microvascular leakage and nonperfusion as the primary drivers of retinal edema. More recently, optical coherence tomography (OCT) has better defined the role and importance of vitreomacular traction in some cases of DME. OCT also improves the diagnosis and follow-up of DME.3 An improved understanding of molecular mechanisms of DME provides additional information beyond the anatomic correlates of imaging studies and has led to new treatments.
Since the initial report of the Early Treatment Diabetic Retinopathy Study (ETDRS) in 1985, focal photocoagulation has remained the primary treatment for clinically significant macular edema (CSME).4 The ETDRS demonstrated that focal and/or grid photocoagulation of CSME reduced the 3-year risk of losing three or more lines of visual acuity from 30% in the control group to 15% in the laser-treated group. The ETDRS laser technique has withstood the test of time. A recent report from the Diabetic Retinopathy Clinical Research Network (DRCR.net) demonstrated that a modified ETDRS-style laser technique was superior to grid photocoagulation. The study group concluded that "Modified ETDRS focal photocoagulation should continue to be a standard approach for treating diabetic macular edema."5
NEW THERAPIES BEING TESTED
Despite the benefits of ETDRS laser for DME, a substantial number of patients fail to respond. Also, it is important to recognize that the primary endpoint of the ETDRS was prevention of visual loss and not visual improvement. Therefore, it is not surprising that a variety of other treatments have been attempted. Vitrectomy with or without internal limiting membrane (ILM) removal has been reported to be effective, particularly in the presence of vitreomacular traction.6 Although there is a strong rationale for vitrectomy, there are no definitive data from clinical trials to establish which patients benefit.
More recently, several pharmacologic therapies have been introduced for DME, including oral protein kinase C inhibitors, intravitreal antivascular endothelial growth factor inhibitors, and intravitreal injection of steroids including fluocinolone, dexamethasone and triamcinolone acetonide.7 Each of these treatments is based upon further elucidation of the molecular pathways operative in DME. Pharmacologic therapy for DME holds great promise. Some of the results reported to date, however, are disappointing. For example, the oral protein kinase C inhibitor ruboxistaurin failed to obtain US Food and Drug Administration approval.7 The DRCR.net recently reported that over a 2-year period, focal/grid photocoagulation is more effective and has fewer side effects than 1 mg or 4 mg doses of preservative-free intravitreal triamcinolone for most patients with DME. The study group concluded that focal/grid photocoagulation should remain the benchmark against which other treatments are compared in clinical trials of DME.8
AN INCREASING BURDEN
A recent study by Saaddine and colleagues estimated the number of Americans 40 years or older with DR or vision-threatening diabetic retinopathy (VTDR) for the years 2005 to 2050.9 They concluded that the number of Americans 40 years of age or older with DR and VTDR will increase from 5.5 million in 2005 to 16.0 million by 2050 and from 1.2 million in 2005 to 3.4 million by 2050, respectively. Not surprisingly, the increases among those 65 years of age or older will be even greater, with those with DR and VTDR increasing from 2.5 million and 0.5 million in 2005 to 9.9 million and 1.9 million in 2050 respectively. These staggering demographic projections illuminate the urgent need for improved, evidence-based therapies of measurable value.
The increasing number of therapeutic options and potential combination therapies for DME raise significant questions about the costs of treating DME. These costs are related to the expense of treatment, including physicians' services, imaging, and drugs. Although these direct medical costs are substantial, they must be considered in relation to the economic impact of DME in terms of the non–treatment-related direct costs of DME and the lost productivity of individuals with DME.
Rein and colleagues estimated that the total direct medical cost of diabetic retinopathy among adult Americans for 2004 was $493 million.10 This number is based on analysis of claims data from 2000-2001. This study did not examine subgroups of diabetic retinopathy. Lee and colleagues analyzed the direct and indirect costs of diabetic retinopathy by analyzing claims data from employees of 17 large companies.11 They compared the annual costs of diabetic employees with DR to those of diabetic employees without DR. Subgroup analyses included employees with DME and proliferative DR, and employees receiving photocoagulation or vitrectomy. The study concluded that employees with DR had significantly higher costs than controls and that indirect costs accounted for about 20% of the total. Costs varied between the subgroups. For patients with diabetes and DME compared with patients with diabetes alone, the difference was $28,606 vs $16,363. For PDR compared with no PDR, the difference was $30,135 vs $13,445. For DR with or without photocoagulation, the difference was $34,539 vs $16,041. The difference was greatest for employees receiving vitrectomy—$63,933 vs $17,239. This study did not distinguish between the indications for photocoagulation or laser.
COST STUDIES
A recent analysis of Medicare beneficiaries from 2000 to 2004 compared the resource use and costs associated with DME compared with those for patients with diabetes alone.12 The presence of DME was associated with 31% higher 1-year costs and 29% higher 3-year costs. Controlling for demographic variables and baseline comorbidities, new-onset DME was a significant independent predictor of total medical costs at 1 and 3 years. When the control group was corrected for nonocular complications, the effects on inpatient and outpatient costs were negated. However, professional service costs remained high, suggesting that increases in professional costs are attributable to services for the diagnosis and treatment of DME. This study also identified some interesting trends in practice patterns. For example, in 2000, OCT was used in less than 3% of patients with incident DME, but by 2004 the use of OCT had increased 15-fold to more than 40%. The use of intravitreal injection increased from less than 1% in 2000 to more than 13% in 2004. This time period incorporates the increased use of intravitreal triamcinolone, still an unproven therapy. Interestingly, during the same time period the use of photocoagulation decreased by 30%, despite the fact that laser remains the only therapy for DME supported by large randomized controlled trials. These data are confirmed by the Practice and Trends survey of the American Society of Retina Specialists for 2008. For a patient with diffuse DME after one unsuccessful laser treatment, only 12% of respondents would repeat laser, while 60% would use an intravitreal injection alone and the remainder would use an intravitreal injection in combination with laser (www.asrs.org).
COST-EFFECTIVENESS DIFFICULT TO ASSESS IN DME
The above cost studies, in combination with the impending demographic explosion of diabetes and DME, present a sobering picture of increasing cost and visual loss attributable to DME. This picture is well understood by health care policy makers and payers and will necessitate a transition to value-based screening and treatment of DME. For a condition as complex and multifactorial as DME, this transition will be difficult. In the laser era, the cost-effectiveness of detecting and treating diabetic retinopathy was well established.13 In fact, laser is a very cost-effective treatment for DME.14 Using a patient-based cost-utility analysis, the cost per quality-adjusted life year (QALY) was $3,101. Generally, a QALY of $20,000 or less is considered cost-effective. However, it is difficult with current methodology to compare the cost-effectiveness of multiple therapies due to variations in utility valuation methods and levels of evidence. As we enter the era of pharmacotherapeutics for DME, the cost of the respective drugs and the likely need for long-term treatment will be significant factors in determining the cost-effectiveness of treatment. In a time of increasing pressure on health care resources, an evidence-based and cost-effective approach to DME will become a necessity.
George A. Williams, MD, is Chair of the Department of Ophthalmology at Williams Beaumont Hospital and Director at Beaumont Eye Institute in Royal Oak, MI. He is also the alternate delegate for the American Academy of Ophthalmology (AAO) to the Amercian Medical Association's Specialty Society Relative Value Scale Update Committee (RUC) and a member of the AAO's Health Policy Committee. Dr. Williams has no financial interest in the information contained in this article. He can be reached via e-mail at GWilliams@beaumont.edu.
_1773249222.png?auto=compress,format&w=75)
