Abstract
Based on self-report in the GRADE diabetes study, cumulative incidence of retinopathy was low over 5 years (3.7%; 184 of 4098 participants) and did not differ among the 4 treatment groups (glargine 4.0%, glimepiride 3.2%, liraglutide 3.7%, sitagliptin 3.8%). There were no differences in retinopathy with specific therapies in GRADE.
Keywords: Diabetes Mellitus Type 2, Diabetic Retinopathy, Patient Reported Outcome Measures, Glycemic Control, Clinical Trial
Graphical Abstract
1.0. INTRODUCTION
Diabetic retinopathy (DR) is an important microvascular complication of type 2 diabetes mellitus (T2DM) with an estimated global prevalence of 35% (1). The development and progression of DR is strongly associated with T2DM duration, severity of hyperglycemia and other risk factors such as nephropathy, hypertension and dyslipidemia (2). In this brief report, we report the retinopathy outcomes in GRADE, a comparative effectiveness trial of four classes of glucose-lowering medications added to metformin in patients with T2DM of < 10 years duration.
2.0. METHODS
The GRADE study is a multi-center trial funded by the National Institutes of Diabetes, Digestive and Kidney Diseases to compare four different medications: insulin glargine, glimepiride (sulfonylurea), liraglutide (glucagon-like peptide 1 receptor agonist), and sitagliptin (dipeptidyl peptidase-4 inhibitor) for glycemic control in participants with a baseline HbA1c of 6.8 to 8.5% (51–69 mmol/mol) when added to metformin. The study randomly assigned 5047 adults with T2DM at 36 centers and 9 additional subsites across the U.S. to one of the four medications. Participants were followed for a mean of 5 years. The rationale and details of the study design are described elsewhere (3). All participants provided written informed consent, and the study was approved by each center’s institutional review board. There were 49 participants with prevalent retinopathy at baseline who were not included in these analyses.
The primary glycemic outcomes as well as prespecified secondary outcomes with respect to objectively measured microvascular and cardiovascular disease have been reported previously (4, 5). In this report, we specifically examine the incidence of self-reported retinopathy and compare rates by treatment group. Self-reported retinopathy was recorded on a standardized form at screening, baseline, and quarterly and included an occurrence of diagnosed retinopathy, treatment for retinopathy, or both. Treatment was defined as laser therapy (excluding laser to treat implanted lens after cataract surgery or laser surgery to correct vision e.g. LASIK), intravitreal injections into the eye, or vitrectomy.
2.1. Statistical Analyses
In order to assess baseline characteristics associated with DR, the incidence rate of DR was presented within categories (strata) of a qualitative covariate and within strata for quantitative variables divided into categories (discretized). Differences in risk among strata were tested by the logrank test. Cumulative incidence within the treatment groups was estimated by Kaplan-Meier curves and group differences tested using logrank tests. Nominal significance was reported at p ≤ 0.05 with no adjustment for multiple tests.
3.0. RESULTS
Overall, 184 participants reported incident retinopathy during follow-up. There were a few characteristics with differences among strata in the incidence of retinopathy (Table 1). The diabetes duration, baseline HbA1c, and blood pressure were all similar among those with incident retinopathy. Participants with incident retinopathy were more likely to be male than female (8.51 versus 5.95 per 1000 person years (PY), nominal p = 0.023), to have a higher median albumin to creatinine ratio (ACR) (8.1 vs 6.3 mg/g Cr), and to have moderate albuminuria (i.e. defined as ACR 30–300 mg/g Cr) at baseline with an incidence of 11.84 versus 6.88 per 1000 PY (p = 0.002) among those with moderate albuminuria compared to those without. There were very few participants with severe albuminuria (N=6).
Table 1.
Incidence of Retinopathy Based on Baseline Characteristics*
| N | Incidence Per 1000 PY | Log-rank p-value+ | |
|---|---|---|---|
| Age (years) | 0.525 | ||
| <45 years | 23 | 7.74 | |
| 45–59 years | 79 | 6.94 | |
| 60+ years | 82 | 8.21 | |
| Sex | 0.023 | ||
| Male | 130 | 8.51 | |
| Female | 54 | 5.95 | |
| Race | 0.106 | ||
| Black | 32 | 6.52 | |
| White | 115 | 7.25 | |
| All others | 37 | 10.36 | |
| Ethnicity | 0.537 | ||
| Non-Hispanic | 147 | 7.45 | |
| Hispanic | 37 | 8.35 | |
| Education | 0.854 | ||
| Less than high school | 12 | 7.04 | |
| High school/GED | 39 | 7.77 | |
| Some college | 48 | 6.86 | |
| College | 55 | 8.50 | |
| Graduate School | 30 | 7.22 | |
| Systolic ≥130 mmgHg | 0.752 | ||
| Yes | 78 | 7.34 | |
| No | 106 | 7.73 | |
| Diastolic ≥80 mmgHg | 0.815 | ||
| Yes | 184 | 7.56 | |
| No | 0 | 0.00 | |
| History of hypertension | 0.113 | ||
| Yes | 132 | 8.19 | |
| No | 52 | 6.32 | |
| BMI (kg/m2) | 0.564 | ||
| 18.2–30.7 | 66 | 8.22 | |
| 30.8–36.2 | 63 | 7.68 | |
| 36.3–74.3 | 55 | 6.81 | |
| Duration of diabetes (years) | 0.837 | ||
| 0–2.3 | 54 | 7.28 | |
| 2.4–5.2 | 65 | 7.53 | |
| 5.3–10.9 | 65 | 7.85 | |
| HbA1c (%, mmol/mol) | 0.220 | ||
| 6.8–7.2, or 50.8–55.2 | 67 | 7.11 | |
| 7.3–7.7, or 55.3–60.7 | 52 | 6.78 | |
| 7.8–8.5, or 60.8–69.4 | 65 | 8.97 | |
| eGFR < 60 mL/min/1.73m2 | 0.505 | ||
| Yes | 3 | 5.1 | |
| No | 181 | 7.62 | |
| Moderate albuminuria (ACR 30–300 mg/g Cr) | 0.002 | ||
| Yes | 40 | 11.84 | |
| No | 144 | 6.88 | |
| Severe albuminuria (ACR >300 mg/g Cr) | 0.057 | ||
| Yes | 6 | 16.05 | |
| No | 178 | 7.44 |
PY= Patient Year
Within categories (strata) for each qualitative characteristic, and strata of discretized quantitative characteristic, the number of retinopathy cases (N) is shown, along with the incidence rate computed as the number of events divided by the total time at risk among the participants in each stratum and presented per 1000 patient years. Participants with baseline retinopathy excluded.
The p-value is from a logrank test of homogeneity (no difference) across strata. All p-values are nominal without correction for multiple tests.
3.1. Incident Retinopathy by Treatment Group
Incident retinopathy over a mean follow-up of 5 years was not significantly different by treatment group. Moreover, each component of the retinopathy definition was not different among the 4 groups, recognizing that one participant could experience multiple components of the outcome as shown in Table 2. Finally, the analysis of time to first reported retinopathy showed that the cumulative incidence of retinopathy was not statistically different by treatment group (Figure 1).
Table 2.
Comparison of Incident Retinopathy During Follow-up by Treatment Group
| Glargine (N=1253) | Glimepiride (N=1241) | Liraglutide (N=1251) | Sitagliptin (N=1253) | P Value | |
|---|---|---|---|---|---|
| Any Retinopathy Outcome | 50 (4.0%) | 40 (3.2%) | 46 (3.7%) | 48 (3.8%) | 0.766 |
| Self-Reported Retinopathy Diagnosis | 42 (3.4%) | 36 (2.9%) | 31 (2.5%) | 42 (3.4%) | 0.521 |
| Laser | 9 (0.7%) | 8 (0.6%) | 13 (1.0%) | 9 (0.7%) | 0.680 |
| Injections | 18 (1.4%) | 9 (0.7%) | 15 (1.2%) | 9 (0.7%) | 0.186 |
| Vitrectomy | 5 (0.4%) | 2 (0.2%) | 5 (0.4%) | 3 (0.2%) | 0.629 |
The p-values are from logrank tests comparing the treatment groups, with no adjustment for multiple comparisons
Participants with baseline retinopathy (n=49) excluded from analyses
Figure 1.
Cumulative Incidence of Retinopathy by Treatment Group Over Follow-up
4.0. DISCUSSION
In this analysis, 184 new cases of retinopathy were reported over a mean follow up of 5 years resulting in an overall incidence rate < 4%. The GRADE study had several strengths including the large and racially diverse cohort with a relatively short baseline duration of T2DM of <10 years and HbA1c of 6.8–8.5% (51–69 mmol/mol) while using metformin monotherapy. Treatments such as laser, intravitreal injections, and vitrectomy are highly memorable and reliable participant reported outcomes related to retinopathy. However, while measurements of microvascular complications in GRADE included objective measures of renal function, albuminuria and formal neuropathy testing; eye examinations were not a component of the study, and thus measurements of retinopathy were restricted to self-report leading to possible ascertainment bias. In addition, participants who have not seen an eye doctor recently may have underreported retinopathy; and reporting of retinopathy may not be sensitive, as this diagnosis is not always clearly communicated to patients by their eye doctor or understood by patients. Of the participants who met the retinopathy definition in the study, some were diagnosed during the study but not treated, some received treatment but had no diagnosis during GRADE (they may have had the condition prior to randomization), and some had both a diagnosis and treatment at some point during the study. In addition, there was no medical record review or adjudication of events, and the retinopathy cases were not confirmed. Despite these limitations, the relatively low rate of retinopathy outcomes is unsurprising given the relatively short duration of T2DM (< 10 years) and relatively good glycemic control at baseline (mean HbA1c of 7.5%). Importantly, the incidence of retinopathy did not differ by treatment groups in GRADE. In keeping with recent analyses demonstrating the safety of glucagon-like peptide 1 receptor agonists for retinopathy (6, 7), the GRADE study did not reveal any retinopathy concerns among the 4 treatment groups (including liraglutide) in the setting of intensification of glycemic control.
In conclusion, our report adds to the literature by describing the low prevalence and incidence of diabetes-related retinopathy in people with T2DM of relatively short duration irrespective of study treatment provided over a mean follow up of 5 years.
Supplementary Material
ACKNOWLEDGEMENTS
Funding:
The GRADE Study was supported by a grant from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health under Award Number U01DK098246. The planning of GRADE was supported by a U34 planning grant from the NIDDK (U34-DK-088043). The American Diabetes Association supported the initial planning meeting for the U34 proposal. The National Heart, Lung, and Blood Institute and the Centers for Disease Control and Prevention also provided funding support. The Department of Veterans Affairs provided resources and facilities. Additional support was provided by grant numbers P30 DK017047, P30 DK020541-44, P30 DK020572, P30 DK072476, P30 DK079626, P30 DK092926, U54 GM104940, UL1 TR000439, UL1 TR000445, UL1 TR001108, UL1 TR001409, 2UL1TR001425, UL1 TR001449, UL1 TR002243, UL1 TR002345, UL1 TR002378, UL1 TR002489, UL1 TR002529, UL1 TR002535, UL1 TR002537, 2UL1 TR001425 and UL1 TR002548. Educational materials have been provided by the National Diabetes Education Program. Material support in the form of donated medications and supplies has been provided by Becton, Dickinson and Company, Bristol-Myers Squibb, Merck & Co., Inc., Novo Nordisk, Roche Diagnostics, and Sanofi. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The GRADE Study Research Group is deeply grateful to our participants whose loyal dedication made GRADE possible.
Footnotes
Clinicaltrials.gov identifier: NCT01794143
Ethics Approval: GRADE is a multi-center RCT, approved by over 30 institutional review boards, so it is not possible to cite all this information including approval numbers in the main text. We have indicated the primary review board submission information from the most recent annual renewal below:
The George Washington University, Office of Human Research - Institutional Review Board; IRB Number: 071245; Last Approved: 7/25/2023; Expires: 8/23/2024.
Conflicts of Interest: DH, LSK, HKS and NY have nothing to disclose.
Data Availability Statement:
The Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study (GRADE) is funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). This manuscript is based on follow-up data and outcome assessments from the 5047 participants enrolled into the study. This database will be available in the NIDDK Central Repository by 2024.
REFERENCES
- 1.Yau JW, Rogers SL, Kawasaki R, Lamoureux EL, Kowalski JW, Bek T, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35(3):556–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Solomon SD, Chew E, Duh EJ, Sobrin L, Sun JK, VanderBeek BL, et al. Diabetic Retinopathy: A Position Statement by the American Diabetes Association. Diabetes Care 2017;40(3):412–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nathan DM, Buse JB, Kahn SE, Krause-Steinrauf H, Larkin ME, Staten M, et al. Rationale and design of the glycemia reduction approaches in diabetes: a comparative effectiveness study (GRADE). Diabetes Care. 2013;36(8):2254–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.The GRADE Study Research Group, Nathan DM, Lachin JM, Balasubramanyam A, Burch HB, Buse JB, et al. Glycemia Reduction in Type 2 Diabetes - Glycemic Outcomes. N Engl J Med 2022;387(12):1063–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.The GRADE Study Research Group, Nathan DM, Lachin JM, Bebu I, Burch HB, Buse JB, et al. Glycemia Reduction in Type 2 Diabetes - Microvascular and Cardiovascular Outcomes. N Engl J Med. 2022;387(12):1075–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bethel MA, Diaz R, Castellana N, Bhattacharya I, Gerstein HC, Lakshmanan MC. HbA(1c) change and diabetic retinopathy during GLP-1 receptor agonist cardiovascular outcome trials: a meta-analysis and meta-regression. Diabetes Care 2021;44(1):290–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ueda P, Pasternak B, Eliasson B, Svensson AM, Franzen S, Gudbjornsdottir S, et al. Glucagon-like peptide 1 receptor agonists and risk of diabetic retinopathy complications: cohort study in nationwide registers from two countries. Diabetes Care 2019;42(6):e92–e4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study (GRADE) is funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). This manuscript is based on follow-up data and outcome assessments from the 5047 participants enrolled into the study. This database will be available in the NIDDK Central Repository by 2024.