Pages

Tuesday, June 19, 2012

Avandia, Actos May Raise Risk of Macular Edema


The diabetes drugs pioglitazone (Actos) and rosiglitazone (Avandia) appear to substantially boost risk of macular edema in type 2 disease, an observational study found.
Incidence of diabetic macular edema was 1.3% at 1 year on a thiazolidinedione (TZD) compared with just 0.2% among type 2 diabetes patients not on one of those agents, Iskandar Idris, MD, of Sherwood Forest Hospitals Foundation Trust in Nottingham, England, and colleagues found.
The risk remained 2.3 times higher out to 10 years of follow-up after adjustment for use of other diabetes and cardiovascular drugs, glucose control, and other factors, the group reported online in the Archives of Internal Medicine.
"Clinicians should be vigilant in the clinical screening for diabetic macular edema among those patients taking thiazolidinediones," they wrote, noting that pioglitazone didn't look any safer than rosiglitazone in the study.
The glitazone drugs work by activating peroxisome proliferator-activated receptors(PPAR-gamma), which are also found in the retinal vasculature and may contribute to fluid retention there.
Cases of macular edema reported with the use of TZDs have resulted in regulatory warnings on the potential association, but prior trials and observational studies have produced conflicting results, and causality remains unclear, Sonal Singh, MD, MPH, and Jodi B. Segal, MD, MPH, both of Johns Hopkins, noted in an accompanying commentary.
The new observational data don't make the link definitive either, they cautioned.
For one thing, the study didn't control for duration of use of the drugs or duration of diabetes, due to lack of data on those factors.
Also, "since thiazolidinediones are typically used as second-line agents, the reported differences may reflect the underlying diabetes severity rather than effect of the drug," Singh and Segal wrote.
Significant differences between users and nonusers in these regards were possible but not likely given the propensity score analysis and similar baseline profiles, according to the researchers.
They analyzed a retrospective cohort study of 103,368 type 2 diabetes patients without baseline diabetic macular edema in the Health Improvement Network (THIN) database, a population-based cohort from primary care practices in England and Wales.
The 3,227 patients who used pioglitazone or rosiglitazone were 5.7-fold more likely to develop diabetic macular edema than those who didn't, in the raw analysis (95% CI 4.1 to 7.9).
Adjustment for age, systolic blood pressure, lipid levels, hemoglobin A1c, and use of aspirin, fibrates, insulin, oral anti-diabetic drugs, or renin-angiotensin system blockers, along with propensity score analysis to try to eliminate selection bias, still left a 2.3-fold elevated risk of the eye condition among thiazolidinedione users at 1-year follow-up (95% CI 1.5 to 3.6) and at 10-year follow-up (95% CI 1.7 to 3.0).
The odds ratios were similar at 3.6 for pioglitazone and 3.1 for rosiglitazone at 1 year of follow-up.
The combination of insulin and a thiazolidinedione boosted that adjusted risk to 4.4-fold, whereas concurrent use of aspirin or an ACE inhibitor significantly reduced the risk.
New-user cohort studies with validated exposure and outcome definitions and adjustment for diabetes severity might settle the issue of causality, but meanwhile the characteristic visual symptoms of diabetic macular edema should prompt evaluation and referral as suggested in the current drug labels, Singh and Segal agreed.
Shared decision-making is key when weighing use of a TZD, they noted, pointing to risk of peripheral edema, congestive heart failure, and bone fractures with both drugs, of myocardial ischemia attributed to rosiglitazone, and of bladder cancer with pioglitazone.
Rosiglitazone is available in the U.S. only through a restricted access program and was removed from the market in Europe.
The study was supported by the Sherwood Forest Hospitals Charitable Trust Fund.
The researchers reported having no conflicts of interest to disclose.
Singh and Segal reported having no conflicts of interest to disclose.

1 comment:

  1. The data seems to suggest an association.
    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3114317/?tool=pmcentrez

    But more studies are needed.

    ReplyDelete