There are many options for treatment for type 2 diabetes including the thiazolidinediones (TZDs), peroxisome proliferator-activated receptor γ agonists that increase insulin sensitivity in the periphery and in the liver.1 Their efficacy is comparable to metformin and sulfonylureas, the typical first-line treatments for type 2 diabetes, and lower A1C about 1.0–1.5% with a low risk for hypoglycemia.2 Unfortunately, TZDs have been associated with various risks, such as cardiovascular events (rosiglitazone), bone fractures, pulmonary edema and heart failure, and potentially bladder cancer (pioglitazone) and diabetic macular edema (DME).1,2 DME is a chronic condition associated with diabetes and the leading cause of blindness; it effects up to 20% of type 2 diabetics.1 Until recently, the results regarding TZDs and their ocular effects on DME have been inconclusive; and in 2010, a cross-sectional baseline data study revealed that thiazolidinediones were not associated with DME.1,3
Idris and colleagues further investigated the incidence of DME in a retrospective cohort study using The Health Improvement Network (THIN) database review from 2000 to 2009. The study assessed the short-term (1 year) and long-term (10 year) impact of TZD use and the incidence of DME accounting for risk factors that may affect vision in patients with type 2 diabetes. Patients included were diagnosed with type 2 diabetes, ≥ 18 years of age, exposed to thiazolidinediones ≥ 6 months, had no history of DME, and were patients of specified medical practices for ≥ 1 year since release of TZDs in the United Kingdom. The total number of patients reviewed was 103,368 (100,141 nonusers and 3,227 TZD users).1
Baseline patient characteristics were similar except for the TZD group had a higher mean A1C and more insulin use and the nonusers had more aspirin and ACE inhibitor use. Results revealed patients exposed to a TZD had a higher risk of DME (p<0.001). Incidence of DME short-term, 1-year, was 1.3% in TZD users versus 0.2% in non-TZD users. After correcting for confounding factors, exposure to a TZD continued to be associated with a significantly increased risk of DME at 1 year (OR 3.3, 95% CI, 2.5-5.0). Both pioglitazone and rosiglitazone were associated with this increased risk (OR 3.6, 95% CI, 2.0 – 6.6 and OR 3.1, 95% CI, 1.9 – 5.1, respectively).1 After ten years of follow up, the increased risk of DME was still evident after adjustment for the confounding factors, missing values, and selection bias, (OR 2.3, 95% CI, 1.7–3.0). Concomitant TZD and insulin use increased DME incidence further (HR 3.0, 95% CI, 1.5–5.9), while incidence decreased with aspirin and ACE inhibitor use (HR 0.6, 95% CI, 0.4–0.9 and HR 0.4, 95% CI 0.2–0.7, respectively).1
This recent study shows an association between the short-term and long-term use of TZDs and incidence of DME. The study investigators acknowledge limitations to their study, such as not assessing the duration of type 2 diabetes or length of exposure to and dose of TZDs. Baseline characteristics of TZD users described greater severity of type 2 diabetes with greater A1C levels and use of insulin with less use of protective medication compared to nonusers. These factors could have confounded data in favor of the association. Adherence to thiazolidinediones was not recordable through the database and could also have been a confounder.1 This study adds to the overwhelming evidence cautioning use of TZDs in certain patient populations.1,2 ...