Retinal oxygen saturation is associated with HbA1c but not with short-term diabetes control, internal environment,...

Purpose of this prospective uncontrolled single-centre pilot study was to find an association of retinal oxygen saturation (SatO₂ ) with acid-base balance (ABB), carboxyhaemoglobin concentration, current plasma glucose concentration (PG), mean PG and PG variability over the last 72 hr, haemoglobin A1c (HbA1c), and other conditions.

Forty-one adults (17 men) with type 1 (N = 14) or type 2 (N = 27) diabetes mellitus, age 48.6 ± 13.5 years, diabetes duration 9 (0.1-36) years, BMI 29.4 ± 6.3 kg/m² , and HbA1c 52 ± 12.7 mmol/mol completed the study. The 4-day study comprised two visits (Day l, Day 4) including 72 hr of continuous glucose monitoring (CGM) by iPro^® 2 Professional CGM (Medtronic, MiniMed, Inc., Northridge, CA, USA). Retinal oximeter Oxymap T1 (Oxymap ehf., Reykjavik, Iceland) was used to assess SatO₂ .

Wilcoxon signed-rank test showed no SatO₂ difference between eyes and visits. Spearman's correlation analysis revealed a significant correlation between arterial SatO₂ and PG variability in type 2 diabetes mellitus, a positive correlation of venous SatO₂ with HbA1c and with finger pulse oximetry. However, no correlation of SatO₂ with ABB, carboxyhaemoglobin, current PG, mean PG over the 72 hr, age, diabetes duration, BMI, lipoproteinaemia, body temperature, systolic and diastolic blood pressure, heart rate, central retinal thickness and retinal nerve fibre layer thickness was found.

This study confirmed the association of venous SatO₂ with long-term but not with short-term diabetes control, ABB and other conditions. The increased SatO₂ and questionable impact of PG variability on retinal SatO₂ is a research challenge.