Heterogeneity of Capillary Blood Flow in Skeletal Muscle: Evidence from Local ¹³³Xe Clearance and Microsphere Trapping

Abstract

Blood flow and its distribution was studied in isolated-perfused and intact dog gastrocnemius muscles at rest and at several levels of rhythmic tetanic exercise. In each experiment, microspheres 15 um in diameter labeled with ¹²⁵I, ¹⁴¹Ce, ⁵¹Cr, ⁸⁵Sr, ⁹⁵Nb or ⁴⁶Sc, were injected into the arterial inflow while clearance of ¹³³Xe from several small intramuscular depots was monitored for determination of local blood flow (Q_Xe). For determination of blood flow from microsphere trapping (Q_m), the muscle was cut into 140 to 340 pieces, of about 700 mg each.

Both at rest and during exercise the Q_m values followed normal distribution around the mean value. Typical Q_m values were O.14 ± O.05 (mean ± SD) and O.75 ± O.30 ml g⁻¹ min⁻¹ for a resting and maximally stimulated isolated-perfused muscle preparation. Q_m results for intact gastrocnemii were similar: O.11 ± O.08 at rest and O.75 ± O.23 ml g⁻¹ min⁻¹ at maximal exercise. Thus, the coefficient of variation (SD/mean) of local blood flow averaged 45% and showed no clear correlation with rest vs. exercise or isolated-perfused vs. intact state.

The total muscle blood flow was simultaneously determined by three independent methods: direct collection of venous effluent (Q_v), ¹³³Xe clearance (Q_Xe) and microsphere trapping (Q_m). The average Q_m was not significantly different from Q_v. Local washout of ¹³³Xe was multiexponential, indicating perfusion or diffusion inhomogeneity at the microlevel. usually three components were found and the Qxe value therefore depended substantially on the component selected for analysis.

The Q_Xe values based on the middle component averaged 40% of Q_v at all blood flow levels. This finding, which is in agreement with previous measurements of muscle blood flow in running dogs, may be related to diffusion limitation of Xe washout, to veno-arterial diffusion shunt or to functional shunt perfusion.