PRESENTATIONS FROM THE RETINA SOCIETY

Glycemic Control a Possible Factor in DR Progression

A trend in lower rates in diabetic retinopathy may be indicative of better diabetes management.

A meta-analysis of 39 studies on diabetic retinopathy (DR) found that better glycemic control in diabetic patients "substantially reduces" the risk of DR progression. The analysis, undertaken by a Canadian team of researchers led by John Gonder, MD and B. Pham, the former of the Department of Ophthalmology at the Ivey Eye Institute of the University of Western Ontario in London, as well as Stephanie Duench, MSc. The team presented its data in a poster at the Retinal Society meeting in Boston.

LITERATURE REVIEW

The poster, "Quantify the progression of diabetic retinopathy in type 2 diabetes: a meta-analysis," stated the team's objectives as to "(1) Quantify the progression of diabetic retinopathy in type 2 diabetics and; (2) Predict the long term consequences of disease progression." They began by doing MEDLINE and EMBASE searches for articles relevant to the topic at hand. After whittling down the initial 1189 articles down to 39 using specified inclusion-exclusion criteria, the team had a total of 22 387 study participants in papers published between 1980 and 2006 to analyze.

Data abstraction was performed by 2 independent reviewers, B. Pham and Stephanie Duench, and then the Canadian team ran a 4-state Markov transition model for DR progression that yielded a 4-stage model for DR: S0: no DR; S1: background; S2: pre-proliferative DR; and S3: proliferative DR. The data analysis they performed yielded several results: In a patient population that was 51% male and with a mean age of 54 years. Overall the transition of patients from S0 to S1 per year was 7.5%, from S1 to S2 was 9.6%, and from S2 to S3 was 12.4. Progression was modified by baseline A1C. Using 9% A1C as the cut off for low/high A1C the progression from S1 to S2 and S2 to S3 was 6.0%/12.9% and 6.4%/12.5% respectively. Similarly, comparing the progression for low/high A1C from S0 to S3 over a 10-year period was 3.1%/10% and over a 20-year period was 15.2%/37.0% (95% confidence interval: 6.2%-10.6% and 26.3%-38.7%, respectively).

IMPLICATIONS FOR FUTURE RESEARCH

The research team sees specific uses for the data they generated. "By quantifying disease progression," they write, "the data could be useful for the formulation of guidelines for the management of this microvascular complication in type 2 diabetics." Furthermore, the team concedes that their estimates of DR incidence from this meta-analysis are lower in frequency than rates found in larger studies. The team's final conclusion is that this difference between their meta-analysis and the larger, more established studies is more indicative of better treatment for diabetes than management of retinal complications. RP