Hyperoxia, hyperglycemia, and photoreceptor sensitivity in normal and diabetic subjects.

The aim of this study was to investigate the effects of an increase in the saturation of blood oxygen (SaO2) and/or serum glucose on photoreceptor sensitivity in normal subjects and in patients with diabetes mellitus. We monitored cone and rod sensitivity by recording dark-adaptation curves to both green and red test stimuli while inhaling either air (20% O2 + 80% N2) or 100% oxygen in 12 normal subjects and 12 diabetic patients with no (10) or mild (2) retinopathy. We also repeated the experiment in 10 of the normal subjects under hyperglycemia (mean serum glucose: 161 mg/dl). Results show that in normal subjects the dark-adapted cone sensitivity is improved by an increase in SaO2 or by hyperglycemia. Final rod sensitivity is unchanged during hyperoxia and during hyperglycemia when measured with a green test spot. However the kinetics of dark adaptation are altered during hyperglycemia, and an increase in final sensitivity is observed when measured with the red test spot. Inhalation of oxygen during hyperglycemia in normal subjects reduces cone sensitivity compared to that found during hyperglycemia alone (Pasteur effect). In diabetic subjects the dark-adapted cone threshold is comparable to that found in normal subjects, and sensitivity also increases with an increase in SaO2. The final rod threshold, however, is impaired compared to that of the control group, and rod sensitivity is improved by increasing the SaO2. The results suggest that the metabolism of rods and cones may differ in normal subjects: in cones, the rate of metabolism can be augmented by increasing the available oxygen or glucose, whereas rods appear more insensitive to increased blood oxygen saturation and hyperglycemia. In diabetic subjects, both cone and rod metabolism can be increased by supplemental oxygen, indicative of an early rod deficit. The study lends weight to the hypothesis that dark-adapted rods in diabetics are hypoxic before the onset of retinopathy.