Retinal oxygen saturation as a non-invasive estimate for mixed venous oxygen saturation and cardiac output

Coronary artery bypass surgery independently associates with retinal vascular oxygen saturation

PURPOSE

The retinal vasculature is the only part of the microcirculation that can be directly studied by non-invasive imaging. Based on the hypothesis that the systemic circulation is reflected in retinal vessels, we investigated if coronary artery bypass grafting (CABG) is related to changes in retinal vascular oxygen saturation (rSatO₂ ).

METHODS

Retinal metabolism was evaluated by Oxymap T1, which simultaneously captures two retinal images at different wavelengths measuring the retinal arteriolar (raSatO₂ ) and venular (rvSatO₂ ) oxygen saturation. Three to 4 days after surgery, we measured the median rSatO₂ after CABG in 38 patients and in 39 healthy controls (operated for cataract).

RESULTS

Coronary artery bypass grafting patients had higher raSatO₂ (median ± standard deviation 93.1 ± 6.7% versus 90.5 ± 11.2%, p = 0.001) and rvSatO₂ (57.4 ± 8.3% versus 53.5 ± 15.4%, p = 0.048) compared to healthy controls. In multivariable linear regression models, raSatO₂ independently associated with CABG (coefficient + 3.6% in CABG patients, p = 0.007), and rvSatO₂ correlated with gender (coefficient + 9.4% for females, p = 0.001) and CABG (coefficient + 8.2% in patients with CABG, p = 0.001).

CONCLUSIONS

Comparing patients with and without cardiovascular disease, raSatO₂ and rvSatO₂ positively and independently associated with CABG, suggesting their potential as non-invasive markers for coronary large artery disease.