Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Characterization of the Systemic Findings of Patients Undergoing Initiation of Anti-Vascular Endothelial Growth Factor Therapy for Diabetic Macular Edema in Routine Clinical Practice

Siraj Haq, BS; Waseem H. Ansari, MD; Michael M. Han, BS; Thais F. Conti, MD; Felipe F. Conti, MD; Fabiana Q. Silva, MD; Rishi P. Singh, MD

Abstract

BACKGROUND AND OBJECTIVES:

Previous studies have validated that baseline visual acuity (VA) can predict a variance response to anti-vascular endothelial growth factor (VEGF) treatment. However, little is known about the initial systemic presentation of diabetic macular edema (DME) in clinical practice. The aim of this study is to report the baseline systemic findings of patients presenting with DME who received anti-VEGF in clinical practice.

PATIENTS AND METHODS:

A retrospective chart review of patients with DME presenting between April 2012 and December 2016 was performed.

RESULTS:

Data from 638 patients were retrieved. The average patient age was 63.1 years (±11.6 years), and 53% were male. There were 95.6% type II diabetics with an average HgA1c of 8.1% (range: 5.1% to 14.5%). Insulin use was present in 67%, biguanides in 43%, sulfonylureas in 32.8%, DDP4 inhibitors in 11.8%, thiazolidinediones in 3.9%, and D-phenylalanine derivatives in 0.94%. Hypertension was present in 78.4% of patients, cardiac comorbidities in 29.3%, peripheral vascular disease in 16.5%, and renal insufficiency in 22.6%. Patients were then split into two different cohorts based on VA (ETDRS < 70 and ETDRS ≥ 70), and variables were compared between groups.

CONCLUSION:

It was shown that older age, hypertension, elevated creatinine, elevated high-density lipoprotein cholesterol, and decreased biguanide use were positively associated with worse presenting VA.

[Ophthalmic Surg Lasers Imaging Retina. 2019;50:16–24.]

Abstract

BACKGROUND AND OBJECTIVES:

Previous studies have validated that baseline visual acuity (VA) can predict a variance response to anti-vascular endothelial growth factor (VEGF) treatment. However, little is known about the initial systemic presentation of diabetic macular edema (DME) in clinical practice. The aim of this study is to report the baseline systemic findings of patients presenting with DME who received anti-VEGF in clinical practice.

PATIENTS AND METHODS:

A retrospective chart review of patients with DME presenting between April 2012 and December 2016 was performed.

RESULTS:

Data from 638 patients were retrieved. The average patient age was 63.1 years (±11.6 years), and 53% were male. There were 95.6% type II diabetics with an average HgA1c of 8.1% (range: 5.1% to 14.5%). Insulin use was present in 67%, biguanides in 43%, sulfonylureas in 32.8%, DDP4 inhibitors in 11.8%, thiazolidinediones in 3.9%, and D-phenylalanine derivatives in 0.94%. Hypertension was present in 78.4% of patients, cardiac comorbidities in 29.3%, peripheral vascular disease in 16.5%, and renal insufficiency in 22.6%. Patients were then split into two different cohorts based on VA (ETDRS < 70 and ETDRS ≥ 70), and variables were compared between groups.

CONCLUSION:

It was shown that older age, hypertension, elevated creatinine, elevated high-density lipoprotein cholesterol, and decreased biguanide use were positively associated with worse presenting VA.

[Ophthalmic Surg Lasers Imaging Retina. 2019;50:16–24.]

Introduction

Diabetic macular edema (DME) is the most common cause for visual impairment in patients with diabetes. It has been shown to be caused by a breakdown of the blood retinal barrier influenced by increasing levels of vascular endothelial growth factor (VEGF) in a response to microvascular ischemia.1 Multiple phase 3 studies have validated that anti-VEGF treatment results in improved visual and anatomic outcomes.2–4 The Diabetic Retinopathy Clinical Research Network (DRCRnet) Protocol T showed that treatment with intravitreal bevacizumab (Avastin; Genentech, South San Francisco, CA), ranibizumab (Lucentis; Genentech, South San Francisco, CA), and aflibercept (Eylea; Regeneron, Tarrytown, NY) resulted in improvements in visual and anatomic outcomes over baseline. Results at 1 year showed that the mean change in visual acuity (VA) primary outcome varied based on initial presenting baseline VA. Intravitreal aflibercept injection (IAI) demonstrated superiority in patients with 20/50 and worse vision, whereas no clinically significant difference was seen across treatment groups when the initial vision was 20/40 and better.5

Routine clinical practice does not always replicate the findings found in clinical research trials, as these studies enroll patients with strict inclusion and exclusion criteria that do not mimic the breadth of patients seen in routine clinical practice. A comparison between a few baseline systemic characteristics reported in recent DME trials and routine clinical practice is shown in Table 1. However, little is known about the initial presentation of DME in routine clinical practice with regard to patient systemic and ocular comorbidities, their presenting clinical findings, and initial anatomic and visual data. The most recently referenced paper from a Protocol T editorial references data for presenting VA in patients with DME from 1995 and shows that the majority of patients (more than 70%) had 20/40 or better acuity at presentation.6 Many of the DME trials included patients with 20/32 or worse VA; hence, updated information on presenting baseline characteristic information for patients with DME from routine clinical practice is lacking.

Comparison Between DME Studies

Table 1:

Comparison Between DME Studies

The purpose of this study is to report the systemic characteristics of patients presenting with DME at the time of initiation of treatment with intravitreal anti-VEGF in routine clinical practice. The systemic variables were also compared when stratifying patients into two VA cohorts (ETDRS <70 [20/50 or worse] and ETDRS ≥70 [20/40 or better]).

Patients and Methods

This study was performed at Cole Eye Institute, Cleveland, Ohio, after receiving approval from the Cleveland Clinic Investigational Review Board. All study related procedures were performed in accordance with good clinical practice (International Conference on Harmonization of Technical Requirements of Pharmaceuticals for Human Use [ICH] E6), applicable U.S. Food and Drug Administration regulations, and the Health Insurance Portability and Accountability Act. A retrospective chart review was performed to identify patients who were anti-VEGF treatment-naïve and diagnosed with DME at a single institution from April 2012 to December 2016.

Patients ages 18 years and older were included if they presented with Type 1 or 2 diabetes, defined by current or historical use of insulin or oral hypoglycemic, or a documented HbA1C greater than 6.5% at the time of diagnosis. Inclusion required foveal-involving retinal edema secondary to DME as documented by clinical exam and OCT, and initiation of treatment for DME with intravitreal anti-VEGF therapy based on investigator discretion as documented in medical record. Exclusion criteria included any history of macular edema not related to diabetes, including pathologic myopia (spherical equivalent of −6 diopters or more negative, or axial length of 25 mm or more), ocular histoplasmosis syndrome, angioid streaks, choroidal rupture, choroidal neovascularization, macular edema following retinal vein occlusion, age-related macular degeneration, polypoidal choroidal vasculopathy, and any history of intraocular surgery within 3 months of encounter. Epiretinal membranes were not considered an exclusion criterion. Only one eye per subject was enrolled in the study. For patients who met eligibility criteria in both eyes, one was randomly chosen.

Study variables collected from the medical records included best-corrected VA (BCVA); demographics such as age, gender, and race; type and duration of diabetes; medications; history of systemic conditions; systolic / diastolic blood pressure and relevant lab values, when available, including glycated hemoglobin (HbA1c), creatinine, blood urea nitrogen (BUN), estimated glomerular filtration rate (eGFR), cholesterol levels (total, low-density lipoprotein [LDL-C], high-density lipoprotein [HDL-C], triglycerides), urine protein and urine microalbumin within 3 months of initiation of anti-VEGF treatment.

Categorical variables were described using frequencies and percentages, whereas continuous variables were described using means and standard deviations, medians and quartiles, and/or ranges. Relationships between categorical variables and ETDRS groups were assessed using Pearson Chi-square tests, Fisher's exact tests or Kruskal-Wallis tests (for ordered variables), while relationships between continuous variables and ETDRS group were assessed using analysis of variance, t-tests (for normally distributed variables), or Kruskal-Wallis tests (for non-normally distributed variables). Analyses were performed using SAS software (version 9.4; Cary, NC).

Results

Patients were split into two different cohorts based on presenting VA of ETDRS less than 70 (20/50 or worse; n = 326) and ETDRS 70 or greater (20/40 or better; n = 312) like in Protocol T, and variables were compared between groups.

The average age of patients with DME requiring anti-VEGF treatment in the overall cohort was 63.1 years ± 11.6 years. Patients in the ETDRS less than 70 cohort were older, with an average age of 64.2 years ± 12.1 years versus those in the ETDRS greater than 70 cohort, who had an average age of 61.9 years ± 11.0 years (P = .016). In the overall cohort, 53% of patients were male; in the ETDRS less than 70 cohort, 53.7% were male, and in the ETDRS greater than 70 cohort, 52.2% were male (P = .72). In the overall cohort, Caucasians made up 60.7% of the group, African-Americans comprised 30.7% of the group, and Asians comprised 0.78% of the group. Ethnic groups of lower population expression in this study (Latinos, Indians, etc.) were grouped as “other,” which represented 7.8% of patients. In the ETDRS less than 70 cohort, Caucasians made up 58.9% of the group, African-American made up 31.6%, and Asians made up 0.31%, with 9.2% patients identified as “other.” In the ETDRS greater than 70 cohort, Caucasians made up 62.5% of the group, African-American made up 29.8%, and Asians made up 1.3%, with 6.4% patients identified as “other.” No statistically significant differences were found between cohorts in gender or race distributions.

When evaluating patients for history of systemic comorbidities (Table 2), 78.4% of patients had history of hypertension (ETDRS < 70 = 81.8% vs. ETDRS >70 = 74.8%), and 29.3% had history of cardiac comorbidities (ETDRS < 70 = 31.3% vs. ETDRS > 70 = 27.1%). History of peripheral vascular disease was noted in 16.5% of patients (ETDRS < 70 = 16.3% vs. ETDRS > 70 = 16.6%), and 22.6% had history of renal insufficiency (ETDRS < 70 = 25.7% vs. ETDRS >70 = 19.3%). Within all systemic comorbidities, only history of hypertension between VA groups was significantly different (P = .034). However, there was a trend of significance in history of renal insufficiency (P = .058).

Comorbidities Summaries and Comparisons by ETDRS Group

Table 2:

Comorbidities Summaries and Comparisons by ETDRS Group

Patients were diagnosed with diabetes for an average of 177.5 months ± 107.6 months (ETDRS < 70 = 179.8 ± 112 vs. ETDRS > 70 = 175.1 ± 103; P = .62) at the time of diagnosis of DME. More than 95% of the population consisted of type II diabetics (ETDRS < 70 = 95.7% vs. ETDRS >70 = 95.5%; P = .91). When evaluating which medications were used for treatment of diabetes (Table 3), 67% of patients were using insulin (ETDRS < 70 = 67.8% vs. ETDRS > 70 = 67%), 43% were using biguanides (ETDRS < 70 = 37.7% vs. ETDRS > 70 = 48.7%), 32.8% were using sulfonylureas (ETDRS < 70 = 30.7% vs. ETDRS > 70 = 34.9%), 11.8% were using DDP4 inhibitors (ETDRS < 70 = 0.61% vs. ETDRS > 70 = 1.3%), 3.9% were using thiazolidinediones (ETDRS < 70 = 3.7% vs. ETDRS > 70 = 4.2%), 11.8% were using D-phenylalanine derivatives (ETDRS < 70 = 12.6% vs. ETDRS > 70 = 10.9%), and 0.31% were using bile acid sequestrants (ETDRS < 70 = 0.31% vs. ETDRS > 70 = 0.32%), with 5% reporting being controlled by diet alone (ETDRS < 70 = 3.1% vs. ETDRS > 70 = 6.3%). No patients reported using alpha glucosidade inhibitors. Only the use of biguanides was found to be significantly different between groups (P = .005). The number of patients using metiglinides in the whole cohort was too low (n = 5) for the statistical assessment to be relevant.

Medications Comparisons by ETDRS Group

Table 3:

Medications Comparisons by ETDRS Group

The average HgA1c in the whole cohort was found to be 8.1% ± 1.8% (ETDRS < 70 = 8.1% ± 1.8% vs. ETDRS >70 = 8% ± 1.9%). Renal function labs included an average creatinine of 1.6 mg/dL ± 1.4 mg/dL (ETDRS < 70 = 1.7 mg/dL ± 1.6 mg/dL vs. ETDRS > 70 = 1.4 mg/dL ± 1 mg/dL); BUN of 26.7 mg/dL ± 16.6 mg/dL (ETDRS < 70 = 27.9 mg/dL ± 17.2 mg/dL vs. ETDRS > 70 = 25.5 mg/dL ± 15.7 mg/dL); eGFR of 41.6 mL/min/1.73m2 ± 15.9 mL/min/1.73m2 (ETDRS < 70 = 40.5 mL/min/1.73m2 ± 16.9 mL/min/1.73m2 vs. ETDRS > 70 = 42.9 mL/min/1.73m2 ± 14.7 mL/min/1.73m2); and urine microalbumin of 63.7 mcg/L ± 79.4 mcg/L (ETDRS < 70 = 70.8 mcg/L ± 83.6 mcg/L vs. ETDRS > 70 = 53.4 mcg/L ± 76.7 mcg/L). Lipid values were as follows: triglycerides of 140 mg/dL ± mg/dL 93.9 mg/dL in the whole cohort (ETDRS < 70 = 137.2 mg/dL ± 88.3 mg/dL vs. ETDRS > 70 = 142.9 mg/dL ± 98.5 mg/dL); total cholesterol of 171.3 mg/dL ± 46.6 mg/dL (ETDRS < 70 = 167.3 mg/dL ± 46.8 mg/dL vs. ETDRS > 70 = 175.5 mg/dL ± 46.1 mg/dL), LDL-C of 96.3 mg/dL ± 38.5 mg/dL (ETDRS < 70 = 94.3 ± 38.2 mg/dL vs. ETDRS > 70 = 98.4 mg/dL ± 38.8 mg/dL); and HDL-C of 47.8 mg/dL ± 16.9 mg/dL (ETDRS < 70 = 45.8 mg/dL ± 15.7 mg/dL vs. ETDRS > 70 = 50 mg/dL ± 17.8 mg/dL). Creatinine (P = .047), and HDL-C levels (P = .028) were the only metabolic indicators significantly different between the two VA groups; however, there was clearly a trend in BUN levels (P = .056). All metabolic indicators and comparisons by ETDRS BCVA groups can be seen in Table 4.

Metabolic Indicators and Comparisons by ETDRS GroupMetabolic Indicators and Comparisons by ETDRS Group

Table 4:

Metabolic Indicators and Comparisons by ETDRS Group

Intravitreal drug chosen for DME management initiation was also recorded. In the overall cohort, 605 patients (94.8%) initiated treatment with bevacizumab, 11 (1.7%) with ranibizumab, 19 (3%) with aflibercept, and one (0.1%) with steroids. In the ETDRS less than 70 group, bevacizumab was used in 305 patients (93.6%), ranibizumab in five (1.5%), aflibercept in 14 (4.3%), and steroids in one (0.3%). Moreover, in the ETDRS 70 or greater group, 300 patients (96.2%) used bevacizumab, six (1.9%) used ranibizumab, and five (1.6%) used aflibercept; no patient was treated with steroids.

Discussion

One of the few studies that reported baseline characteristics of patients with diabetic retinopathy (DR) was the ETDRS Report No. 7.7 In Table 5, a comparison of baseline characteristics is made between that 1991 trial and the current study. Sex distribution, diabetic retinopathy type, and systolic blood pressure greater than 160 mm Hg are comparable among studies. However, the current study showed a higher rate of patients with type 2 diabetes and creatinine levels greater than 1.5 mg/dL. Additionally, percentages of Caucasian patients, history of cardiovascular disease, HbA1C greater than 10%, diastolic blood pressure greater than 85 mm Hg, cholesterol higher than 240 mg/dL, and LDL-C levels greater than 160 mg/dL were lower in the current study. The lower level of cardiovascular disease was likely due to the increased use of statin drugs in the populations since ETDRS Report 7 was published. Baigent et al. showed in a prospective meta-analysis of 90,056 individuals a significant reduction (P < .001) incidence of coronary events, with reduction in LDL cholesterol achieved with statin therapy.8

Comparison Between Baseline Characteristics Studies

Table 5:

Comparison Between Baseline Characteristics Studies

Additionally, by separating patients into two cohorts similarly to Protocol T, significant differences between the two populations groups were identified. Patients with ETDRS less than 70 were found to be significantly older, on average, by 2.3 years (64.2 years vs. 61.9 years). Among patients who never encounter disease, age will still cause a functional and anatomical decline due to the culmination of oxidative stress and degenerative products. Pitss et al. in a review article has related a moderate but steady decline in VA, with the aging process in patients older than 60 years of age and up to the age of 80 years.9 In the retina, age causes the loss of cone photoreceptors and decreased retinal cell density over time through oxidative stress-induced vascular inflammation.10,11 Additionally, hyperglycemia accelerates this vascular aging causing increased vascular dysfunction.12

There was a significant difference in the hypertensive population between visual acuity groups, with the ETDRS less than 70 group having a higher number of patients reporting a diagnosis of hypertension. Elevated blood pressure is a well-known cause of retinal pathology and a risk factor for DME, associated with clinical signs such as generalized and focal arteriolar narrowing, arteriovenous nicking, flame-shaped and blot-shaped retinal hemorrhages, cotton-wool spots, and swelling of the optic disc. The vascular changes stem from intimal thickening, hyperplasia of the media wall, hyaline degeneration, and later, breakdown of the blood-retinal barrier brought about by persistent hypertension.13 The neural changes stem from obstruction of axoplasmic transport on the periphery of areas infarcted by vascular damage (in the case of cotton-wool spots) and increased intracranial pressure in malignant hypertension (in the case of optic disc swelling).14 Hypertension may cause damage to retinal vessels independently of, and in addition to, hyperglycemia. Thus, patients with hypertension would be likely to have worse VA than those with normal blood pressure.

Creatinine level was, on average, higher in the ETDRS less than 70 group. Additionally, there was a trend of higher BUN levels in the same group. BUN and creatinine are catabolic products of muscle tissue that is cleared by the kidneys and are widely used as an indicator of renal function. BUN normal range is 7 mg/dL to 20 mg/dL and creatinine is 0.6 mg/dL to 1.1 mg/dL in females and 0.7 mg/dL to 1.3 mg/dL in male patients. Factors such as age, weight, and race may influence values. Renal disease is a common complication of diabetes, and although our study did not find significant differences in HgA1c between the two VA groups, HbA1c may be altered by any condition that shortens erythrocyte lifespan (eg, blood loss, hemolytic anemia) and by renal disease itself.15 In such cases, it may be an unreliable indicator of glycemic control. Elevated creatinine and BUN may be reflective of renal pathology brought on by poorer glycemic control not reflected in HgA1c levels, which would cause retinal damage through the mechanisms described above.

Patients in the ETDRS less than 70 group had higher HDL levels, on average. Previous studies have generally not found a significant association between HDL levels and retinopathy.16–18 The probability of this result being merely due to chance exists. Therefore, further investigation is required to study the correlation between cholesterol levels and VA in patients with DME.

Patients in the ETDRS less than 70 group were on average less likely to use biguanides. The most common biguanide used today is metformin, which improves sensitivity to insulin and strengthens insulin's stimulation of the utilization of glucose. Metformin has been shown to be protective to renal tubular cells by reducing oxidative stress and restoring physiological biochemical pathways.19 In addition, animal models in diabetic nephropathy have demonstrated metformin has a protective effect on renal podocyte injury.20 Similar models have demonstrated metformin has a protective effect against retinal cell death in diabetic retinas compared to control retinas.21 In other animal models, metformin has been shown to phosphorylate the VEGF2 receptor, which decreases VEGF signaling in retinas.22 These data support the idea that metformin may lead to healthier retinal vasculature, which may therefore be less affected by the toxic oxidative stress and capillary damage from hyperglycemia. Thus, it might be hypothesized that patients using metformin may have better VAs than those who do not use the medication.

Strengths of this study include its large number of subjects with available lab data. Additionally, it has a less stringent inclusion criteria compared to randomized clinical trials, reflecting real-world analysis. The data reviewed were collected during the last few years, which presents a more recent viewpoint into patients with DME than previous, older studies. Weaknesses include the retrospective nature of the study and the limit to only one study center for patient data collection. Additionally, further reviews and studies should consider socioeconomic factors, including access to health care, type of insurance, or education level. This may contribute to worse VA on presentation, which may be related to chronicity of disease.

In conclusion, this study presents data on patients presenting with DME for initiation of anti-VEGF injections in routine clinical practice. In this scenario, it was found that older age, hypertension, elevated creatinine, elevated HDL-C, and decreased biguanide use were all positively associated with worse presenting VA.

References

  1. Antonetti DA, Klein R, Gardner TW. Diabetic retinopathy. N Engl J Med. 2012;43(1):13–19.
  2. Elman MJ, Qin H, Diabetic Retinopathy Clinical Research Network et al. Intravitreal ranibizumab for diabetic macular edema with prompt versus deferred laser treatment: Three-year randomized trial results. Ophthalmology. 2012;119(11):2312–2318. doi:10.1016/j.ophtha.2012.08.022 [CrossRef]
  3. Brown DM, Nguyen QD, Marcus DM, et al. Long-term outcomes of ranibizumab therapy for diabetic macular edema: The 36-month results from two phase III trials: RISE and RIDE. Ophthalmology. 2013;120(10):2013–2022. doi:10.1016/j.ophtha.2013.02.034 [CrossRef]
  4. Korobelnik JF, Do DV, Schmidt-Erfurth U, et al. Intravitreal aflibercept for diabetic macular edema. Ophthalmology. 2014;121(11):2247–2254. doi:10.1016/j.ophtha.2014.05.006 [CrossRef]
  5. Wells JA, Glassman AR, Diabetic Retinopathy Clinical Research Network et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Engl J Med. 2015;372(13):1193–1203. doi:10.1056/NEJMoa1414264 [CrossRef]
  6. No authors listed. Focal photocoagulation treatment of diabetic macular edema. Relationship of treatment effect to fluorescein angiographic and other retinal characteristics at baseline: ETDRS report no. 19. Early Treatment Diabetic Retinopathy Study Research Group. Arch Ophthalmol. 1995;113(9):1144–1155. doi:10.1001/archopht.1995.01100090070025 [CrossRef]
  7. No authors listed. Early Treatment Diabetic Retinopathy Study design and baseline patient characteristics. ETDRS report number 7. Ophthalmology. 1991;98(5):741–756. doi:10.1016/S0161-6420(13)38009-9 [CrossRef]
  8. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterol-lowering treatment: Prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet. 2005;366(9493):1267–1278. doi:10.1016/S0140-6736(05)67394-1 [CrossRef]
  9. Pitts DG. Visual acuity as a function of age. J Am Optom Assoc. 1982;53(2):117–124.
  10. Nadal-Nicolás FM, Vidal-Sanz M, Agudo-Barriuso M. The aging rat retina: From function to anatomy. Neurobiol Aging. 2018;61:146–168. doi:10.1016/j.neurobiolaging.2017.09.021 [CrossRef]
  11. Donato AJ, Morgan RG, Walker AE, Lesniewski LA. Cellular and molecular biology of aging endothelial cells. J Mol Cell Cardiol. 2015;89(Pt B):122–135. doi:10.1016/j.yjmcc.2015.01.021 [CrossRef]
  12. Lamoke F, Shaw S, Yuan J, et al. Increased oxidative and nitrative stress accelerates aging of the retinal vasculature in the diabetic retina. PLoS One. 2015;10(10):e0139664. doi:10.1371/journal.pone.0139664 [CrossRef]
  13. Tien YW, McIntosh R. Hypertensive retinopathy signs as risk indicators of cardiovascular morbidity and mortality. Br Med Bull. 2005;73–74:57–70.
  14. McLeod D, Marshall J, Kohner EM, Bird AC. The role of axoplasmic transport in the pathogenesis of retinal cotton-wool spots. Br J Ophthalmol. 1977;61(3):177–191. doi:10.1136/bjo.61.3.177 [CrossRef]
  15. Little RR, Rohlfing CL, Tennill AL, et al. Measurement of Hba1C in patients with chronic renal failure. Clin Chim Acta. 2013;418:73–76. doi:10.1016/j.cca.2012.12.022 [CrossRef]
  16. van Leiden H a, Dekker JM, Moll AC, et al. Risk factors for incident retinopathy in a diabetic and nondiabetic population: The Hoorn study. Arch Ophthalmol. 2003;121(2):245–251. doi:10.1001/archopht.121.2.245 [CrossRef]
  17. Klein BEK, Moss SE, Klein R, Surawicz TS. The Wisconsin Epidemiologic Study of Diabetic Retinopathy: XIII. Relationship of serum cholesterol to retinopathy and hard exudate. Ophthalmology. 1991;98(8):1261–1265. doi:10.1016/S0161-6420(91)32145-6 [CrossRef]
  18. Morton J, Zoungas S, Li Q, et al. Low HDL cholesterol and the risk of diabetic nephropathy and retinopathy: Results of the advance study. Diabetes Care. 2012;35(11):2201–2206. doi:10.2337/dc12-0306 [CrossRef]
  19. Rafieian-Kopaei M, Ghaed-Amini F, Nasri H. Metformin and renal protection. J Isfahan Med Sch. 2014;32(307).
  20. Kim J, Shon E, Kim CS, Kim JS. Renal podocyte injury in a rat model of type 2 diabetes is prevented by metformin. Exp Diabetes Res. 2012;2012:210821. doi:10.1155/2012/210821 [CrossRef]
  21. Kim YS, Kim M, Choi MY, et al. Metformin protects against retinal cell death in diabetic mice. Biochem Biophys Res Commun. 2017;492(3):397–403. doi:10.1016/j.bbrc.2017.08.087 [CrossRef]
  22. Yi QY, Deng G, Chen N, et al. Metformin inhibits development of diabetic retinopathy through inducing alternative splicing of VEGF-A. Am J Transl Res. 2016;8(9):3947–3954.

Comparison Between DME Studies

Clinical PracticeRISE / RIDEVIVID / VISTA
Mean Age63.162.462.9
Male56%56.8%57.9%
Caucasian76.4%79.4%81.8%
Mean Duration of Diabetes, Years (SD)14.7 (8.9)15.7 (9.8)15.7 (9.8)
Mean HbA1c (SD)8.1 (1.8)7.6 (1.4)7.8 (1.5)
HbA1c > 8% (SD)40.7%31.9%34.8%

Comorbidities Summaries and Comparisons by ETDRS Group

FactorComorbidities

NOverallNETDRS < 70NETDRS ≥ 70P Value

Hypertension625319306.034c
  No135 (21.6%)58 (18.2%)77 (25.2%)
  Yes490 (78.4%)261 (81.8%)229 (74.8%)

Cardiovascular625319306.25c
  No442 (70.7%)219 (68.7%)223 (72.9%)
  Yes183 (29.3%)100 (31.3%)83 (27.1%)

Cerebral626319307.14c
  No533 (85.1%)265 (83.1%)268 (87.3%)
  Yes93(14.9%)54(16.9%)39(12.7%)

Peripheral Vascular Disease626319307.92c
  No523 (83.5%)267 (83.7%)256 (83.4%)
  Yes103 (16.5%)52 (16.3%)51 (16.6%)

Renal Insufficiency624319305.058c
  No483 (77.4%)237 (74.3%)246 (80.7%)
  Yes141 (22.6%)82 (25.7%)59 (19.3%)

Systolic Blood Pressure470142.8 ± 22.5244144.2 ± 22.2226141.4 ± 22.7.18a
470(86.0, 223.0)244(86.0, 223.0)226(94.0, 221.0)

Systolic Blood Pressure Within Normal Range?470244226.50c
  Normal range (≤ 140)242 (51.5%)122 (50.0%)120 (53.1%)
  Out of range228 (48.5%)122 (50.0%)106 (46.9%)

Diastolic Blood Pressure46976.3 ± 11.824475.6 ± 12.122577.1 ± 11.5.17a
469(47.0, 118.0)244(47.0, 118.0)225(50.0, 109.0)

Diastolic Blood Pressure Within Normal Range?469244225.61c
  Normal range (≤ 80)316 (67.4%)167 (68.4%)149 (66.2%)
  Out of range153 (32.6%)77 (31.6%)76 (33.8%)

Medications Comparisons by ETDRS Group

FactorMetabolic Indicators

NOverallETDRS < 70ETDRS ≥ 70P Value

Insulin638326312.83c
  No208 (32.6%)105 (32.2%)103 (33.0%)
  Yes430 (67.4%)221 (67.8%)209 (67.0%)

Biguanides638326312.005c
  No363 (56.9%)203 (62.3%)160 (51.3%)
  Yes275 (43.1%)123 (37.7%)152 (48.7%)

Sulfonylureas638326312.25c
  No429 (67.2%)226 (69.3%)203 (65.1%)
  Yes209 (32.8%)100 (30.7%)109 (34.9%)

Metiglinides541n/a
  No5 (100.0%)4 (100.0%)1 (100.0%)

D-Phenylalanine Derivatives638326312.44d
  No632 (99.1%)324 (99.4%)308 (98.7%)
  Yes6 (0.94%)2 (0.61%)4 (1.3%)

Thiazolodinediones638326312.75c
  No613 (96.1%)314 (96.3%)299 (95.8%)
  Yes25(3.9%)12 (3.7%)13 (4.2%)

DPP4 Inhibitors638326312.51c
  No563 (88.2%)285 (87.4%)278 (89.1%)
  Yes75 (11.8%)41 (12.6%)34 (10.9%)

Alpha-Glucosidade Inhibitors638326312n/a
  No638 (100.0%)326 (100.0%)312 (100.0%)

Bile Acid Sequestrants638326312.99d
  No636 (99.7%)325 (99.7%)311 (99.7%)
  Yes2 (0.31%)1 (0.31%)1 (0.32%)

Diet-Controlled Diabetes803248.65d
  No76 (95.0%)31 (96.9%)45 (93.8%)
  Yes4 (5.0%)1 (3.1%)3 (6.3%)

Metabolic Indicators and Comparisons by ETDRS Group

FactorMetabolic Indicators

NOverallNETDRS <70NETDRS ≥ 70P Value

HgbA1c (%)3988.1 ± 1.82038.1 ± 1.81958.0 ± 1.9.43a
398(5.1, 14.5)203(5.1, 14.5)195(5.3, 13.9)

Creatinine (mg/dL)4191.6 ± 1.42181.7 ± 1.62011.4 ± 1.06.047b
419(0.34, 11.3)218(0.34, 11.3)201(0.50, 7.1)

Creatinine Within Normal Range?419218201.43c
  Normal range (F: 0.6–1.1; M:0.7–1.3)200 (47.7%)100 (45.9%)100 (49.8%)
  Out of range219 (52.3%)118 (54.1%)101 (50.2%)

BUN (mg/dL)41526.7 ± 16.621627.9 ± 17.219925.5 ± 15.7.056b
415(5.0, 122.0)216(5.0, 122.0)199(7.0, 91.0)

BUN Within Normal Range?415216199.66c
  Normal range (7–20)183 (44.1%)93 (43.1%)90 (45.2%)
  Out of range232 (55.9%)123 (56.9%)109 (54.8%)

eGFR (mL/min/1.73m2)23541.6 ± 15.912640.5 ± 16.910942.9 ± 14.7.25a
235(5.0, 93.0)126(5.0, 93.0)109(9.0, 59.0)

eGFR Within Normal Range?235126109.99d
  Normal range (≥ 90)1 (0.43%)1 (0.79%)0 (0.0)
  Out of range234 (99.6%)125 (99.2%)109 (100.0%)

Triglycerides (mg/dL)314140.0 ± 93.3162137.2 ± 88.3152142.9 ± 98.5.68b
314(20.0, 585.0)162(20.0, 543.0)152(29.0, 585.0)

Triglycerides Within Normal Range?314162152.75c
  Normal range (≤ 200)256 (81.5%)131 (80.9%)125 (82.2%)
  Out of range58 (18.5%)31 (19.1%)27 (17.8%)

Total Cholesterol (mg/dL)316171.3 ± 46.6163167.3 ± 46.8153175.5 ± 46.1.12a
316(76.0, 333.0)163(76.0, 333.0)153(87.0, 304.0)

Total Cholesterol Within Normal Range?316163153.12c
  Normal range (≤ 240)287 (90.8%)152 (93.3%)135 (88.2%)
  Out of range29 (9.2%)11 (6.7%)18 (11.8%)

LDL-c (mg/dL)32096.3 ± 38.516494.3 ± 38.215698.4 ± 38.8.35a
320(29.0, 237.0)164(29.0, 226.0)156(38.0, 237.0)

LDL Within Normal Range?320164156.90c
  Normal range (≤ 160)296 (92.5%)152 (92.7%)144 (92.3%)
  Out of range24 (7.5%)12 (7.3%)12 (7.7%)

HDL-c (mg/dL)31447.8 ± 16.916145.8 ± 15.715350.0 ± 17.8.028a
314(17.0, 130.0)161(23.0, 130.0)153(17.0, 110.0)

HDL Within Normal Range?314161153.024c
  Normal range (≥ 40)202 (64.3%)94 (58.4%)108 (70.6%)
  Out of range112 (35.7%)67 (41.6%)45 (29.4%)

Urine Protein (g/gL)36125.7 ± 73.819024.9 ± 77.917126.7 ± 69.1.68b
361(0.10, 814.0)190(0.50, 814.0)171(0.10, 550.0)

Urine Protein Within Normal Range?361190171.81c
  Normal range (≤ 8)289 (80.1%)153 (80.5%)136 (79.5%)
  Out of range72 (19.9%)37 (19.5%)35 (20.5%)

Urine Microalbunim (mg/l)2263.7±79.41370.8±83.6953.4±76.7.39b
22(0.91,272.5)13(0.91,272.5)9(4.0,223.0)

Urine Microalbunim Within Normal Range?22139.55c
  Normal range (≤ 30)13 (59.1%)7 (53.8%)6 (66.7%)
  Out of range9 (40.9%)6 (46.2%)3 (33.3%)

Comparison Between Baseline Characteristics Studies

ETDRS Report No. 7Clinical Practice
Male56%53%
Caucasian76.4%60.7%
Type 2 DM68.7%95.6%
Systolic BP >160 mm Hg20.2%24.6%
Diastolic BP > 85 mm Hg39%21.3%
Creatinine > 1.5 mg/dL6.8%30.8%
History of cardiovascular disease48.8%29.3%
PDR38.5%37%
NPDR61.5%63%
HbA1c > 10%30.2%14.5%
Cholesterol > 240 mg/dL26%9.2%
LDL > 160 mg/dL17.9%7.5%
Authors

From Case Western Reserve University School of Medicine, Cleveland (SH, MH, RPS); Cole Eye Institute, Cleveland Clinic Foundation, Cleveland (WHA, TFC, FFC, FQS, RPS); and the Center for Ophthalmic Bioinformatics, Cole Eye Institute, Cleveland Clinic Foundation, Cleveland (TFC, FFC, RPS).

This study was supported by a research grant from Regeneron Pharmaceuticals.

Dr. Singh has received grants and personal fees from Regeneron, Genentech/Roche, and Alcon/Novartis; personal fees from Optos, Zeiss, and Biogen; and grants from Apellis during the conduct of the study. The remaining authors report no relevant financial disclosures.

Dr. Singh did not participate in the editorial review of this manuscript.

Address correspondence to Rishi P. Singh, MD, Cole Eye Institute, Cleveland Clinic Foundation, 9500 Euclid Ave, Mail Code i-32, Cleveland, OH 44195; email: singhr@ccf.org.

Received: May 09, 2018
Accepted: November 02, 2018

10.3928/23258160-20181212-03

Sign up to receive

Journal E-contents