Consistency of Treatment Effects with Various Diabetic Retinopathy Outcomes in ACCORD Eye Study

Abstract

 

Purpose:

 

To examine the consistency of medical treatment effects (intensive blood sugar, blood lipid, and blood pressure control vs. standard or placebo) across alternative diabetic retinopathy (DR) outcome measures in the Action to Control Cardiovascular Risk in Diabetes Trial (ACCORD) Eye Study (n=2856).

 

Methods:

 

Previously ACCORD reported beneficial effects of intensive glycemic (IntGly) and lipid (Fenofibrate), but not blood pressure (IntBP), control in reducing DR progression, pre-defined as 3+ steps worsening (ETDRS patient scale) or vitrectomy or laser photocoagulation, baseline to 4 yrs (NEJM 2010, 128(3):312-318). We re-analyzed results for the following alternative progression definitions: (1) for worsening (ETDRS patient scale), 1+, 2+ and 4+ steps; (2) for improvement, 1+, 2+, and 3+ steps; (3) incidence of any DR; (4) incidence of severe non-proliferative DR (NPDR) or worse, and (5) 2+ steps worse on eye scale (worse or either eye at baseline).

 

Results:

 

The table below indicates whether significant treatment effects were found by DR outcome definition and treatment group comparison (analyses adjusted for covariates).  

 

Conclusions:

 

Compared to the published ACCORD findings of significant treatment effects based upon conventional definition DR progression (3+ steps worsening), benefit was also found using other outcome definitions: for glycemic control, at 2+ and 1+ steps worse, and 1+ and 2+ steps better; for fenofibrate, at 4+ steps worse. Consistency of effects was seen also with alternative outcome definitions including the eye scale.

 

Clinical Trial:

 

http://www.clinicaltrials.gov NCT00542178

 

Keywords: diabetic retinopathy • clinical (human) or epidemiologic studies: outcomes/complications