tion studies were performed on euhydrated subjects under standardized clinical conditions (e.g., controlled diet, body posture, skin temperature, inactivity).

Studies have indicated that BIA may not have sufficient accuracy to validly detect moderate dehydration (approximately 7 percent TBW) and loses resolution with isotonic fluid loss (O’Brien et al., 1999). Because fluid, electrolyte, and plasma protein concentrations can have independent effects, BIA can provide misleading values regarding dehydration or hyperhydration status (Gudivaka et al., 1999; O’Brien et al., 2002). Fluid and electrolyte concentrations may have independent effects on the BIA signal, thus often providing grossly misleading values regarding dehydration status (O’Brien et al., 2002). The BIA with a 0/∞ − kHz parallel (Cole-Cole) multifrequency model may have promise to measure body hydration changes if corrections are made for changes in plasma protein concentration (Gudivaka et al., 1999). However, recently a multifrequency BIA with Cole-Cole analysis was reported not to be sensitive to hypertonic dehydration (Bartok et al., 2004).

Plasma osmolality provides a marker of dehydration levels. Osmolality is closely controlled by homeostatic systems and is the primary physiological signal used to regulate water balance (by hypothalamic and posterior pituitary arginine vasopressin secretion), resulting in changes in urine output and fluid consumption (Andreoli et al., 2000; Knepper et al., 2000). Plasma osmolality rarely varies beyond ± 2 percent and is controlled around a set-point of 280 to 290 mOsmol/kg; this set-point increases with aging and becomes more variable among people. Water deprivation (if it exceeds solute losses) increases the osmolality of plasma and of the ECF and thus fluids bathing the hypothalamus. This causes loss of ICF from osmoreceptor neurons, which then signals the release of arginine vasopressin from the hypothalamus and the posterior pituitary. Arginine vasopressin acts on the renal tubules to increase water reabsorption.

Arginine vasopressin release is proportional to increased plasma osmolality and decreased plasma volume. While body water loss will induce plasma volume reduction and increased plasma osmolality, the influence of body water loss on each depends upon the method of dehydration, physical fitness level, and heat acclimatization status (Sawka, 1988; Sawka and Coyle, 1999).

Many studies have measured plasma osmolality of euhydrated sub-