Baseline haemoglobin A1c influences retinal function after long-term insulin pump therapy.

Purpose: The purpose of the study was to characterize the long-term effect of insulin pump therapy (CSII) on electroretinography and dark adaptometry and to examine the influence of baseline glycaemic control on retinal function in patients with type 1 diabetes mellitus.

Methods: This prospective observational extension study enrolled 13 patients out of 17 who completed a primary 1-year study of the effect of CSII on retinal function. Twelve patients were still on CSII at follow-up. The extension study included a single examination 3.5 years (range 3.0-4.0 years) after initiation of CSII of one study eye per patient. Procedures included full-field electroretinography (ERG), dark adaptometry, optical coherence tomography, and fundus photography.

Results: Mean ERG amplitudes 3.5 years after initiation of CSII were 15-43 % lower than at baseline (all p < 0.05) and 21-45 % lower than after 1 year on CSII. The mean rate of dark adaptation had returned to baseline after a transient 13 % (p = 0.0024) acceleration at the 1-year visit. Reduction of ERG amplitudes between 1 and 3.5 years was statistically associated predominantly with baseline haemoglobin A1c (HbA1c) ≥ 8.7 % and, to a smaller extent, with HbA1c reductions larger than 1.9 % after initiation of CSII. No significant changes in ERG amplitudes were found in patients with baseline HbA1c < 8.7 % and HbA1c reductions smaller than 1.9 %.

Conclusions: Deterioration of subclinical retinal function from 1 to 3.5 years after initiation of CSII was associated predominantly with poorer metabolic control before initiation of CSII. Analyses of retinal function may supplement structural and morphological characteristics in the study of diabetic complications.