Oxyopia Abstract

March 5, 2004
Friday, 4 PM
489 Minor Hall

Debra A. Schaumberg, ScD, OD, MPH
Assistant Professor of Medicine and Ophthalmology, Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School
Host: Suzanne Fleiszig

Title

"Systemic markers of inflammation and risk of diabetic retinopathy"

Abstract

Purpose: Diabetic retinopathy (DR) is the leading cause of blindness among working aged adults in the US. Hyperglycemic exposure is the major known risk factor; nonetheless, variations in HbA1c explain only 10-34% of variability in DR progression. Several lines of evidence support a role for inflammation in the pathogenesis on DR. We investigated whether systemic markers of inflammation predict the onset or progression of DR, as well as clinically significant macular edema (CSME), retinal hard exudate (HE), and proliferative diabetic retinopathy (PDR).

Methods: We conducted a prospective cohort study. We measured levels of high-sensitivity C-reactive protein (hsCRP), soluble intercellular adhesion molecule-1 (sICAM-1), soluble vascular cell adhesion molecule-1 (sVCAM-1), and soluble tumor necrosis alpha receptor-1 (sTNFR1). Markers were measured in stored baseline serum samples from the 1441 participants in the Diabetes Control and Complications Trial (DCCT), a randomized trial of intensive versus conventional therapy in type 1 diabetes. We divided levels into quintile groups, and using proportional hazards regression models, we tested whether levels of these markers predicted DR outcomes.

Results: Although sICAM-1 was a significant predictor in univariable models (RR=1.51 for top versus bottom fifth; P=0.007), none of the markers significantly predicted progression of DR after controlling for HbA1c and other risk factors. Findings were similar for PDR. For CSME, both hsCRP and sICAM-1 were significant in univariable models, but only hsCRP persisted as a significant risk factor (RR=1.83; P=0.01) after controlling for HbA1c and other risk factors. Both hsCRP (RR=1.78; P=0.004) and sICAM-1 (RR=1.50; P=0.05) predicted HE in multivariable models. When we examined whether higher levels of hsCRP might be a risk factor, we observed significant increased risk of both HE (RR=2.38) and PDR (RR=2.91) comparing hsCRP at or above versus below the 95th percentile.

Conclusions: Findings of some significant associations between markers of inflammation and DR endpoints are consistent with the hypothesis that subclinical inflammation contributes to the pathogenesis or DR. The stronger relationships with CSME and HE point to breakdown of the blood retinal barrier as a possible mechanism of insult. Further studies are needed to determine whether markers of inflammation may be useful as markers of increased risk of DR among people with type 1 diabetes.