SUMMARY1. In human subjects, sustained static contractions of the knee extensors were performed in one leg with the same absolute (10% of the initial maximal voluntary contraction) and relative (30 % of the maximal voluntary contraction immediately prior to the static exercise) intensities before and during epidural anaesthesia. Epidural anaesthesia reduced strength to 62 + 8 % of the control value and partially blocked sensory input from the working muscles. During contractions performed with the same relative force, the increases in mean arterial pressure and heart rate were greater during control contractions than during epidural anaesthesia. During contractions at the same absolute force, there was no significant difference in magnitude of cardiovascular responses between control contractions and contractions performed during epidural anaesthesia.2. The metabolic role in the exercise pressor reflex was assessed by applying an arterial leg cuff 10 s before cessation of exercise and through the following 3 min of recovery. Although mean arterial pressure and heart rate decreased immediately after cessation of exercise, application of the arterial occlusion cuff resulted in higher post-exercise mean arterial pressure and heart rate values. Control and epidural mean arterial pressures during arterial occlusion were not significantly different.3. The results of this study suggest that the reflex neural mechanism rather than the intended effort (central command) is important in determining the blood pressure and heart rate responses to static exercise in man. That is, when epidural anaesthesia diminishes sensory feedback and produces muscular weakness, central command does not determine the cardiovascular response. This conclusion, however, is opposite to that derived from experiments with partial neuromuscular blockade which demonstrated the importance of central command in determining the cardiovascular response to static exercise (Leonard, Mitchell, Mizuno, Rube, Saltin & Secher, 1985). Taken together, these two studies are complementary and support the concept that both central and reflex neural mechanisms play roles in regulating arterial blood pressure and heart rate during static exercise in man.

SUMMARY1. The cardiovascular responses, heart rate and mean arterial pressure, were followed in seventeen human subjects who performed static handgrip contractions for 2 min at the same absolute force (15% of the initial maximal voluntary contraction strength) before and during partial curarization. In control contractions the rate of perceived exertion was 10 exertion units, 16 units in contractions with tubocurarine which could be maintained and 20 units in contractions that could not be maintained. Control contractions increased mean arterial pressure by 6 mmHg from 89 mmHg while heart rate was unchanged from the resting value of 68 beats min-'. With tubocurarine, larger increases in mean arterial pressure of 11 mmHg and for heart rate of 8 beats min-1 were obtained during maintained contractions, and 15 mmHg and 16 beats min1. respectively, during non-maintained contractions.2. Atropine increased resting heart rate and blood pressure with tubocurarine to 107 beats min-' and 98 mmHg, respectively, in seven subjects. The blood pressure response to exercise with tubocurarine was unaffected by atropine, but the heart rate increase was reduced from 15 to 4 beats min-'.3. Propranolol reduced resting heart rate with tubocurarine to 56 beats min-1 with no effect on blood pressure in seven subjects. The cardiovascular responses to exercise with tubocurarine were unaffected by propranolol. In contrast, phentolamine reduced resting blood pressure with tubocurarine to 80 mmHg without affecting heart rate in seven subjects. Exercise responses with tubocurarine were unaffected by phentolamine. Combinations of atropine and propranolol in fourteen subjects or atropine and phentolamine in five subjects showed similar results during exercise with tubocurarine as with the sole use of the agents used to block autonomic receptors.4. The results suggest that when partial curarization induces a disproportion between the signal from central command and that from exercising muscles, the larger signal arising from central command determines the magnitude of the cardiovascular responses. The centrally generated heart rate response is in part caused by vagal withdrawal. However, the blood pressure response cannot be * Present address: Harry S. Moss Heart Center, University of Texas Southwestern Medical Center, Dallas, TX 75235-9034, USA. J. H. MITCHELL AND OTHERSattenuated by the sole use of a-or fl-receptor adrenergic blockade or combinations of these with atropine. This suggests that there may be greater redundancy in the autonomic control of blood pressure than in the vagal control of heart rate associated with central command during static exercise in man.

Cerebral blood flow (CBF) was determined in humans at rest and during four consecutive unilateral static contractions of the knee extensors. Each contraction was maintained for 3 min 15 s with the subjects in a semisupine position. The contractions corresponded to 8, 16, 24, and 32% of the maximal voluntary contraction (MVC) and utilized alternate legs. CBF (measured by the 133Xe clearance technique) was expressed by a noncompartmental flow index (ISI). Heart rate and mean arterial pressure increased from resting values of 73 (55-80) beats/min and 88 (74-104) mmHg to 106 (86-138) beats/min and 124 (102-146) mmHg, respectively (P less than 0.0005), during the contraction at 32% MVC. Arterial PCO2 and central venous pressure did not change. Corrected to the average resting PCO2, CBF during control was 55 (35-73) ml.100 g-1.min-1 and remained constant during contractions. Cerebral vascular resistance increased from 1.5 (1.0-2.2) to 2.4 (1.4-3.0) mmHg. 100 g.min.ml-1 (P less than 0.025) at 32% of MVC. There was no difference in CBF between the two hemispheres at rest or during exercise. In contrast to dynamic leg exercise, static leg exercise is not associated with an increase in global CBF when measured by the 133Xe clearance technique.