Running head: OSMOTIC MODULATION OF BAROREFLEX DURING HEAT STRESS. Corresponding author:

The aim of this study was to elucidate the interactive effect of central hypovolemia and plasma hyperosmolality on the regulation of peripheral vascular response during heat stress. We compared the response of forearm vascular conductance (FVC) to lower body negative pressure (LBNP) in normosmotic (NOSM) and hyperosmotic (HOSM) conditions in normothermia (NT) and during whole body heating (HT). Seven male subjects were infused with either 0.9% (NOSM) or 3.0% NaCl solution (HOSM), and then were heated by perfusing 42°C water (HT) through water perfusion suits (34.5°C for NT). Sixty minutes later, the subjects were exposed to a progressive LBNP to -40 mmHg in 10-mmHg steps. Plasma osmolality increased by ~11 mosmol/kg H2O in HOSM conditions and plasma arginine vasopressin concentration ([AVP]p) increased to 3.47 ± 0.65 and 5.28 ± 0.68 pg/ml (mean±SEM), in NTand HT-HOSM, respectively. The increase in esophageal temperature was much larger in HT-HOSM (0.90 ± 0.09°C) than HT-NOSM (0.30 ± 0.07°C) (p<0.01) because of the osmotic inhibition of thermoregulatory responses. Mean arterial pressure during LBNP was well maintained in all conditions. The increase in thoracic impedance and the decrease in stroke volume during LBNP were similar in all conditions. FVC before application of LBNP was 8.16 ± 0.86, 8.96 ±1.20, 15.17 ± 2.89, and 17.50 ± 2.04 conductance unit in NT-NOSM, NT-HOSM, HT-NOSM, and HT-HOSM, respectively. FVC in HT was higher than NT (p<0.001), and FVC in NOSM did not differ from HOSM within the thermal conditions. The decrease in FVC during LBNP was similar between NT-NOSM and NT-HOSM, while the reduction in FVC at -40 mmHg in HT-HOSM (-9.99 ± 0.96 conductance unit; 58.8 ± 4.1 %) was significantly larger than in HT-NOSM (-6.02 ± 1.23 conductance unit; 44.7 ± 8.1 %) (p<0.05). In contrast, plasma [AVP]p response to LBNP did not interact with plasma osmolality during normothermia and heat stress. These data indicate that there apparently exists an interactive effect (manuscript #R-00027-2004-R3 final accepted version) Page 3 for plasma hyperosmolality and central hypovolemia on the peripheral vascular response during heat stress, or peripheral vasodilated condition, but not in normothermia. In contrast, effects of plasma hyperosmolality and baroreceptor unloading on AVP secretion were additive in normothermia and during heat stress. (manuscript #R-00027-2004-R3 final accepted version) Page 4