Sex differences in microvascular patterns associated with low- and high-risk for cardiovascular disease.

backgroundThe role of the microcirculation in cardiovascular health in males and females is not fully understood. Skin microvascular function can be assessed non-invasively by measuring the response to a brief occlusion of the arm, which produces a hyperemic response that depends on multiple regulatory pathways. Whether the resulting time-resolved microvascular parameters discriminate between low- and high-cardiovascular-risk individuals, and whether the informative features differ between males and females, is not known.

methodWe analyzed time-resolved skin microcirculatory hemoglobin oxygen saturation and perfusion measured using a combined system integrating laser Doppler flowmetry and diffuse reflectance spectroscopy during a standardized occlusion-release protocol (including baseline, occlusion and reperfusion phases with a total duration of 20 min) in 3,177 participants in the Swedish CArdioPulmonary bioImage Study (SCAPIS) microvascular sub-study. Participants were stratified into a low-risk group (Systematic COronary Risk Evaluation 2 (SCORE2) < 5%, no diabetes or previous atherosclerotic cardiovascular disease (ASCVD); n = 1,645) and a high-risk group (SCORE2 > 7.5%, or diabetes, or previous ASCVD, or severe chronic kidney disease; n = 761). Random forest classifiers were trained to discriminate low- from high-risk individuals using the full time-resolved signals, in the combined cohort and separately by sex, with feature importance mapped back to the time axis.

resultsThe classifier achieved area under the curve (AUC) of 0.685 (95% confidence interval (CI): 0.663-0.707) in the full cohort (i.e. including both females and males). Discrimination was primarily driven by the post-occlusion hyperemic phase of oxygen saturation, with lower responses observed in high-risk individuals compared to low-risk individuals. Sex-stratified models achieved similar performance in females (AUC 0.638, 95% CI: 0.596-0.679) and males (AUC 0.644, 95% CI: 0.628-0.660). In males, both perfusion and oxygen saturation contributed to risk discrimination, with low-risk individuals exhibiting a more pronounced hyperemic perfusion response. In females, discrimination was driven mainly by oxygen saturation, while perfusion responses showed substantial overlap between risk groups and limited discriminatory value.

conclusionsTime-resolved microvascular signals provide information for cardiovascular risk discrimination that differs between males and females. In both sexes, high-risk individuals exhibited an attenuated post-occlusion oxygen saturation response, indicating impaired microvascular reactivity. Perfusion contributed additional discriminatory information in males, but showed limited value in females. These findings suggest that perfusion and oxygen saturation reflect partly different aspects of microvascular regulation, with different relevance to cardiovascular risk in males and females.