Dynamics of noninvasively estimated microvascular O2 extraction during ramp exercise

Utilization of near-infrared spectroscopy (NIRS) in clinical exercise testing to detect microvascular abnormalities requires characterization of the responses in healthy individuals and theoretical foundation for data interpretation. We examined the profile of the deoxygenated hemoglobin signal from NIRS {deoxygenated hemoglobin + myoglobin [deoxy-(Hb+Mb)] approximately O(2) extraction} during ramp exercise to test the hypothesis that the increase in estimated O(2) extraction would be close to hyperbolic, reflecting a linear relationship between muscle blood flow (Q(m)) and muscle oxygen uptake (Vo(2)(m)) with a positive Q(m) intercept. Fifteen subjects (age 24 +/- 5 yr) performed incremental ramp exercise to fatigue (15-35 W/min). The deoxy-(Hb+Mb) response, measured by NIRS, was fitted by a hyperbolic function [f(x) = ax/(b + x), where a is the asymptotic value and b is the x value that yields 50% of the total amplitude] and sigmoidal function {f(x) = f(0) + A/[1 + e(-(-c+dx))], where f(0) is baseline, A is total amplitude, and c is a constant dependent on d, the slope of the sigmoid}, and the goodness of fit was determined by F test. Only one subject demonstrated a hyperbolic increase in deoxy-(Hb+Mb) (a = 170%, b = 193 W), whereas 14 subjects displayed a sigmoidal increase in deoxy-(Hb+Mb) (f(0) = -7 +/- 7%, A = 118 +/- 16%, c = 3.25 +/- 1.14, and d = 0.03 +/- 0.01). Computer simulations revealed that sigmoidal increases in deoxy-(Hb+Mb) reflect a nonlinear relationship between microvascular Q(m) and Vo(2)(m) during incremental ramp exercise. The mechanistic implications of our findings are that, in most healthy subjects, Q(m) increased at a faster rate than Vo(2)(m) early in the exercise test and slowed progressively as maximal work rate was approached.