Abstract
Purpose: :
To determine if worsening of diabetic eye disease as viewed by photographic diabetic eye screenings is related to Hemoglobin A1c level at the time of initial screening.
Methods: :
A retrospective analysis of patients screened at the Veteran’s Affairs Hospital in Nashville, TN was performed. Patients with a previous history of diabetes type 2 who had at least two screenings between 2003 and 2008 were included. Screenings involved fundus photographs which were read by trained observers and graded as levels of "no disease," "mild npdr," "moderate npdr," "severe npdr," or "proliferative disease." There were no exclusion criteria. Patients were placed in two groups based upon hemoglobin A1c level measured within three months of the initial screening. Group 1 included those whose hemoglobin A1c measured less than 7.5 within three months of initial screening. Group 2 included those whose hemoglobin A1c measured greater than or equal to 7.5. Patients whose disease based upon the above levels worsened were said to "progress." Those whose disease based upon the above levels improved or stayed the same were said to "not progress."
Results: :
103 patients had two consecutive diabetic screenings and hemoglobin A1c measurements within three months of the initial screening. 76 patients had hemoglobin A1c less than 7.5 and were placed in group 1. 27 patients had hemoglobin A1c equal to or greater than 7.5 and were placed in group 2. The average length of time between screenings was 24.8 months (range 9-54). Of 76 patients in group 1, 16 patients (21.1%) progressed, and 60 patients either improved or stayed the same. Of 25 patients in group 2, 9 patients (33.3%) progressed, and 16 either improved or stayed the same. Chi-square analysis demonstrated that neither group was more likely to show progression at the second screening (p=0.3). However, if progression did not include a change from "no disease" to "mild npdr" in either eye, only 2 of 76 patients (2.6%) in group 1 showed progression, while 4 of 25 patients (14.8%) in group 2 showed progression. Chi-square analysis demonstrated that patients in group 2 were more likely to show progression under these criteria (p=0.0395).
Conclusions: :
Hemoglobin A1c greater than or less than 7.5 does not appear to relate to progression of diabetic eye disease as viewed by diabetic eye screenings. If, however, progression does not include a change from "no disease" to "mild npdr," patients with hemoglobin A1c greater than or equal to 7.5 may be more likely to show progression.
Keywords: diabetic retinopathy • clinical (human) or epidemiologic studies: risk factor assessment