The Wisconsin epidemiologic study of diabetic retinopathy: what can we learn at 14 years?

Abstract

In the lead article in this issue of Ophthalmology, Klein et al1Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XVII. The fourteen-year incidence and progression of diabetic retinopathy and associated risk factors in type I diabetes.Ophthalmology. 1998; 105: 1801-1815Abstract Full Text Full Text PDF PubMed Scopus (613) Google Scholar report on the retinopathy status of a population-based cohort of 634 individuals with type I diabetes followed up 14 years in the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR). Information on progression to retinopathy endpoints (first appearance of proliferative diabetic retinopathy [PDR], first appearance of macular edema, progression of retinopathy) is provided, stratified by gender, age, and duration of diabetes. Although previous studies on cohorts of individuals with diabetes have been published, few have had such lengthy follow-up and such consistent use of standardized methods to describe retinopathy status.Significantly increased risk of PDR after 14 years’ follow-up was associated with longer duration of diabetes, older age, more severe retinopathy at baseline, hypertension, proteinuria, and higher baseline levels of glycosylated hemoglobin (Hgb A1). After multivariate adjustment, baseline retinopathy, hypertension, baseline Hgb A1, and change in Hgb A1 from baseline to 4 years were significantly associated with increased risk of PDR at 14 years. Similarly, after multivariate adjustment, the risk of macular edema was significantly higher for those with more severe retinopathy at baseline, proteinuria, higher Hgb A1 levels at baseline, and change in Hgb A1 from baseline to 4 years. These results are consistent with earlier reports from the WESDR, which followed cohorts of individuals with diabetes using standardized methods at baseline and at 4 and 10 years.2Klein R. Klein B.E.K. Moss S.E. et al.The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IX. Four-year incidence and progression of diabetic retinopathy when age at diagnosis is less than 30 years.Arch Ophthalmol. 1989; 107: 237-243Crossref PubMed Scopus (390) Google Scholar, 3Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XIV. Ten-year incidence and progression of diabetic retinopathy.Arch Ophthalmol. 1994; 112: 1217-1228Crossref PubMed Scopus (429) Google Scholar In all the reports on retinopathy over the years from this cohort, regardless of whether progression, incidence, PDR, or macular edema is considered, higher Hgb A1 level has been identified as a significant risk factor, in agreement with results from many other epidemiologic studies. However, demonstrating an association between a risk factor and an outcome in an observational study does not provide adequate evidence that abating the risk factor will decrease the chance of the outcome. The critical epidemiologic step of moving from a demonstrated association to evaluating the efficacy of an intervention was supplied by the Diabetes Control and Complications Trial,4The DCCT Research GroupThe effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus.New Engl J Med. 1993; 329: 977-986Crossref PubMed Scopus (22519) Google Scholar which confirmed the value of better control of glycemic levels in reducing the risk of retinopathy and visual loss. The WESDR results indicate that the beneficial effect of lower Hgb A1 may extend to a more general population, and one in which therapy was not standardized.Vision outcomes at 14 years are reported for a small subset of the WESDR cohort in this report. (Moss et al5Moss S.E. Klein R. Klein B.E.K. The 14-year incidence of visual loss in a diabetic population.Ophthalmology. 1998; 105: 998-1003Abstract Full Text Full Text PDF PubMed Scopus (389) Google Scholar reported on the incidence of visual impairment and blindness for the entire cohort at 14 years, although these endpoints were not stated to be due solely to diabetic retinopathy.) In this report, of those with PDR at baseline, 8% lost vision in at least one eye by the 14-year follow-up. Of those with high-risk characteristics (Diabetic Retinopathy Study [DRS]-HRC),6Diabetic Retinopathy Study Research GroupIndications for photocoagulation treatment of diabetic retinopathy.Int Ophthalmol Clin. 1987; 27 (DRS Report No. 14): 239-253Crossref PubMed Scopus (145) Google Scholar nearly 10% lost vision in both eyes and nearly 40% lost vision in one eye. The mechanism of visual loss is not stated; presumably most was due to retinopathy.The progression of retinopathy needs to be interpreted with care. Progression was defined as a two-step increase on an ordinal scale of retinopathy. However, a two-step increase at the lower end of the scale could mean a change from minimal retinopathy to early nonproliferative changes (e.g., hard exudates), whereas a two-step increase at the higher end of the scale could mean moving from preproliferative retinopathy to advanced PDR. This lack of linearity in the scale is discussed by the authors as a possible reason for the observation that those with less severe retinopathy were more likely to progress.What information is here for the ophthalmologist who manages patients with diabetes? As stated by the authors, these results confirm current guidelines7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar: patients who are age 12 or older with type I diabetes for more than 5 years should continue to receive a dilated examination by an ophthalmologist yearly, because their risk of serious retinopathy is high. Patients with hypertension, proteinuria, and more severe retinopathy need to be followed up more closely, because it has been shown, in this study and in others, that these patients are at higher risk for developing PDR and/or macular edema. These data also show that, regardless of duration of diabetes, Hgb A1 level, or stage of retinopathy, improved glycemic control was associated with a decreased risk of PDR and macular edema. In this group of participants at 14 years, a 1% decrease in Hgb A1 over the first 4-year period was associated with a 30% decrease in the risk of PDR or macular edema.It may be presumed that participants in a large epidemiologic study would obtain information on their retinopathy status from the study examiner. In addition, it is likely that referrals to community ophthalmologists were made for individuals with retinopathy requiring treatment. Thus the reported frequency of examination and treatment by “nonstudy” physicians may be greater for this cohort than for the general population of persons with diabetes. Consequently, it is surprising that, of those individuals with DRS-HRC at the baseline examination, fewer than 40% were found to have been treated 14 years later. One explanation for this finding may be poor compliance with monitoring recommendations.7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar Indeed, at the 10-year follow up, only 64% of this cohort reported having had a dilated eye examination in the past year.8Moss S.E. Klein R. Klein B.E.K. Factors associated with having eye examinations in persons with diabetes.Arch Fam Med. 1995; 4: 529-534Crossref PubMed Scopus (115) Google Scholar Additionally, it is known that many individuals with type I diabetes do not receive the recommended frequency of ophthalmologic examinations.9Javitt J.C. Aiello L.P. Bassi L.J. et al.Detecting and treating retinopathy in patients with type I diabetes mellitus. Savings associated with improved implementation of current guidelines.Ophthalmology. 1991; 98: 1565-1574Abstract Full Text PDF PubMed Scopus (120) Google Scholar Methods to improve compliance with examinations, whether through patient education, outreach programs, or improved technology for remote evaluations, are a necessary first step to combat progression of diabetic retinopathy.In addition to compliance, one needs to focus on the treatment of retinopathy. Recommendations on laser treatment are available to guide the ophthalmologist.7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar However, it must be realized that although laser photocoagulation has been shown to be beneficial for DRS-HRC and for clinically significant macular edema, it is not a cure. Approximately 14% of those treated with laser photocoagulation for diabetic macular edema in the Early Treatment Diabetic Retinopathy Study lost vision.10Early Treatment Diabetic Retinopathy Study Research GroupPhotocoagulation for diabetic macular edema.Arch Ophthalmol. 1985; 103 (ETDRS Report No. 1): 1796-1806Crossref PubMed Scopus (2492) Google Scholar In addition, treatment may entail significant expense. In the future, additional modalities may be available to treat or prevent diabetic retinopathy. Indeed, some pharmacologic treatments have shown promise in treating diabetic retinopathy.11Aiello L.P. Vascular endothelial growth factor. 20th-century mechanisms, 21st-century therapies.Invest Ophthalmol Vis Sci. 1997; 38: 1647-1652PubMed Google ScholarKlein et al1Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XVII. The fourteen-year incidence and progression of diabetic retinopathy and associated risk factors in type I diabetes.Ophthalmology. 1998; 105: 1801-1815Abstract Full Text Full Text PDF PubMed Scopus (613) Google Scholar have conducted a high-quality study showing the 14-year rates of serious, vision-threatening retinopathy in a cohort of nonclinic-based individuals with diabetes. Such information challenges ophthalmologists and researchers to increase efforts to monitor the ophthalmic status of patients with diabetes effectively, to apply appropriate treatment, and to explore new methods for treating and preventing retinopathy. In the lead article in this issue of Ophthalmology, Klein et al1Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XVII. The fourteen-year incidence and progression of diabetic retinopathy and associated risk factors in type I diabetes.Ophthalmology. 1998; 105: 1801-1815Abstract Full Text Full Text PDF PubMed Scopus (613) Google Scholar report on the retinopathy status of a population-based cohort of 634 individuals with type I diabetes followed up 14 years in the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR). Information on progression to retinopathy endpoints (first appearance of proliferative diabetic retinopathy [PDR], first appearance of macular edema, progression of retinopathy) is provided, stratified by gender, age, and duration of diabetes. Although previous studies on cohorts of individuals with diabetes have been published, few have had such lengthy follow-up and such consistent use of standardized methods to describe retinopathy status. Significantly increased risk of PDR after 14 years’ follow-up was associated with longer duration of diabetes, older age, more severe retinopathy at baseline, hypertension, proteinuria, and higher baseline levels of glycosylated hemoglobin (Hgb A1). After multivariate adjustment, baseline retinopathy, hypertension, baseline Hgb A1, and change in Hgb A1 from baseline to 4 years were significantly associated with increased risk of PDR at 14 years. Similarly, after multivariate adjustment, the risk of macular edema was significantly higher for those with more severe retinopathy at baseline, proteinuria, higher Hgb A1 levels at baseline, and change in Hgb A1 from baseline to 4 years. These results are consistent with earlier reports from the WESDR, which followed cohorts of individuals with diabetes using standardized methods at baseline and at 4 and 10 years.2Klein R. Klein B.E.K. Moss S.E. et al.The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IX. Four-year incidence and progression of diabetic retinopathy when age at diagnosis is less than 30 years.Arch Ophthalmol. 1989; 107: 237-243Crossref PubMed Scopus (390) Google Scholar, 3Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XIV. Ten-year incidence and progression of diabetic retinopathy.Arch Ophthalmol. 1994; 112: 1217-1228Crossref PubMed Scopus (429) Google Scholar In all the reports on retinopathy over the years from this cohort, regardless of whether progression, incidence, PDR, or macular edema is considered, higher Hgb A1 level has been identified as a significant risk factor, in agreement with results from many other epidemiologic studies. However, demonstrating an association between a risk factor and an outcome in an observational study does not provide adequate evidence that abating the risk factor will decrease the chance of the outcome. The critical epidemiologic step of moving from a demonstrated association to evaluating the efficacy of an intervention was supplied by the Diabetes Control and Complications Trial,4The DCCT Research GroupThe effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus.New Engl J Med. 1993; 329: 977-986Crossref PubMed Scopus (22519) Google Scholar which confirmed the value of better control of glycemic levels in reducing the risk of retinopathy and visual loss. The WESDR results indicate that the beneficial effect of lower Hgb A1 may extend to a more general population, and one in which therapy was not standardized. Vision outcomes at 14 years are reported for a small subset of the WESDR cohort in this report. (Moss et al5Moss S.E. Klein R. Klein B.E.K. The 14-year incidence of visual loss in a diabetic population.Ophthalmology. 1998; 105: 998-1003Abstract Full Text Full Text PDF PubMed Scopus (389) Google Scholar reported on the incidence of visual impairment and blindness for the entire cohort at 14 years, although these endpoints were not stated to be due solely to diabetic retinopathy.) In this report, of those with PDR at baseline, 8% lost vision in at least one eye by the 14-year follow-up. Of those with high-risk characteristics (Diabetic Retinopathy Study [DRS]-HRC),6Diabetic Retinopathy Study Research GroupIndications for photocoagulation treatment of diabetic retinopathy.Int Ophthalmol Clin. 1987; 27 (DRS Report No. 14): 239-253Crossref PubMed Scopus (145) Google Scholar nearly 10% lost vision in both eyes and nearly 40% lost vision in one eye. The mechanism of visual loss is not stated; presumably most was due to retinopathy. The progression of retinopathy needs to be interpreted with care. Progression was defined as a two-step increase on an ordinal scale of retinopathy. However, a two-step increase at the lower end of the scale could mean a change from minimal retinopathy to early nonproliferative changes (e.g., hard exudates), whereas a two-step increase at the higher end of the scale could mean moving from preproliferative retinopathy to advanced PDR. This lack of linearity in the scale is discussed by the authors as a possible reason for the observation that those with less severe retinopathy were more likely to progress. What information is here for the ophthalmologist who manages patients with diabetes? As stated by the authors, these results confirm current guidelines7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar: patients who are age 12 or older with type I diabetes for more than 5 years should continue to receive a dilated examination by an ophthalmologist yearly, because their risk of serious retinopathy is high. Patients with hypertension, proteinuria, and more severe retinopathy need to be followed up more closely, because it has been shown, in this study and in others, that these patients are at higher risk for developing PDR and/or macular edema. These data also show that, regardless of duration of diabetes, Hgb A1 level, or stage of retinopathy, improved glycemic control was associated with a decreased risk of PDR and macular edema. In this group of participants at 14 years, a 1% decrease in Hgb A1 over the first 4-year period was associated with a 30% decrease in the risk of PDR or macular edema. It may be presumed that participants in a large epidemiologic study would obtain information on their retinopathy status from the study examiner. In addition, it is likely that referrals to community ophthalmologists were made for individuals with retinopathy requiring treatment. Thus the reported frequency of examination and treatment by “nonstudy” physicians may be greater for this cohort than for the general population of persons with diabetes. Consequently, it is surprising that, of those individuals with DRS-HRC at the baseline examination, fewer than 40% were found to have been treated 14 years later. One explanation for this finding may be poor compliance with monitoring recommendations.7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar Indeed, at the 10-year follow up, only 64% of this cohort reported having had a dilated eye examination in the past year.8Moss S.E. Klein R. Klein B.E.K. Factors associated with having eye examinations in persons with diabetes.Arch Fam Med. 1995; 4: 529-534Crossref PubMed Scopus (115) Google Scholar Additionally, it is known that many individuals with type I diabetes do not receive the recommended frequency of ophthalmologic examinations.9Javitt J.C. Aiello L.P. Bassi L.J. et al.Detecting and treating retinopathy in patients with type I diabetes mellitus. Savings associated with improved implementation of current guidelines.Ophthalmology. 1991; 98: 1565-1574Abstract Full Text PDF PubMed Scopus (120) Google Scholar Methods to improve compliance with examinations, whether through patient education, outreach programs, or improved technology for remote evaluations, are a necessary first step to combat progression of diabetic retinopathy. In addition to compliance, one needs to focus on the treatment of retinopathy. Recommendations on laser treatment are available to guide the ophthalmologist.7Preferred Practice Pattern. Diabetic Retinopathy. The American Academy of Ophthalmology, San Francisco1991Google Scholar However, it must be realized that although laser photocoagulation has been shown to be beneficial for DRS-HRC and for clinically significant macular edema, it is not a cure. Approximately 14% of those treated with laser photocoagulation for diabetic macular edema in the Early Treatment Diabetic Retinopathy Study lost vision.10Early Treatment Diabetic Retinopathy Study Research GroupPhotocoagulation for diabetic macular edema.Arch Ophthalmol. 1985; 103 (ETDRS Report No. 1): 1796-1806Crossref PubMed Scopus (2492) Google Scholar In addition, treatment may entail significant expense. In the future, additional modalities may be available to treat or prevent diabetic retinopathy. Indeed, some pharmacologic treatments have shown promise in treating diabetic retinopathy.11Aiello L.P. Vascular endothelial growth factor. 20th-century mechanisms, 21st-century therapies.Invest Ophthalmol Vis Sci. 1997; 38: 1647-1652PubMed Google Scholar Klein et al1Klein R. Klein B.E.K. Moss S.E. Cruickshanks K.J. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XVII. The fourteen-year incidence and progression of diabetic retinopathy and associated risk factors in type I diabetes.Ophthalmology. 1998; 105: 1801-1815Abstract Full Text Full Text PDF PubMed Scopus (613) Google Scholar have conducted a high-quality study showing the 14-year rates of serious, vision-threatening retinopathy in a cohort of nonclinic-based individuals with diabetes. Such information challenges ophthalmologists and researchers to increase efforts to monitor the ophthalmic status of patients with diabetes effectively, to apply appropriate treatment, and to explore new methods for treating and preventing retinopathy.