Exercising skeletal muscle blood flow in humans responds to reduction in arterial oxyhaemoglobin, but not to altered free oxygen

1. We hypothesised that reducing arterial oxyhaemoglobin (O(2)Hb(a)) with carbon monoxide (CO) in both normoxia and hyperoxia, or acute hypoxia would cause similar compensatory increases in human skeletal muscle blood flow and vascular conductance during submaximal exercise, despite vast differences in arterial free oxygen partial pressure (P-a,P-O2). 2. Seven healthy males completed four 5 min one-legged knee-extensor exercise bouts in the semi-supine position (30 +/- 3 W, mean +/- S.E.M.), separated by similar to1 h Of rest, under the following conditions: (a) normoxia (O(2)Hb(a) = 195 ml l(-1); P-a,P-O2 = 105 mmHg); (b) hypoxia (163 ml l(-1); 47 mmHg); (c) CO + normoxia (18% COHb(a); 159 ml l(-1); 119 mmHg); and (d) CO + hyperoxia (19% COHb(a); 158 ml l(-1); 538 mmHg). 3. CO + normoxia, CO + hyperoxia and systemic hypoxia resulted in a 29-44% higher leg blood flow and leg vascular conductance compared to normoxia (P < 0.05), without altering blood pH, blood acid-base balance or net leg lactate release. 4. Leg blood flow and leg vascular conductance increased in association with reduced O(2)Hb(a) (r(2) = 0.92-0.95; P < 0.05), yet were unrelated to altered P-a,P-O2. This association was further substantiated in two subsequent studies with graded increases in COHb(a) (n = 4) and NO synthase blockade (n = 2) in the presence of normal P-a,P-O2. 5. The elevated leg blood flow with CO + normoxia and CO + hyperoxia allowed a similar to 17% greater O-2 delivery (P < 0.05) to exercising muscles, compensating for the lower leg O-2 extraction (61%) compared to normoxia and hypoxia (69%; P < 0.05), and thereby maintaining leg oxygen uptake constant. 6. The compensatory increases in skeletal muscle blood flow and vascular conductance during exercise with both a CO load and systemic hypoxia are independent of pronounced alterations in P-a,P-O2 (47-538 mmHg), but are closely associated with reductions in O(2)Hb(a). These results suggest a pivotal role of O-2 bound to haemoglobin in increasing skeletal muscle vasodilatation during exercise in humans.