Severe chronic undernutrition is associated with decreased bone turnover and significant bone loss. However, little is known about the short-term effects of nutritional deprivation on bone turnover. To investigate the effects of short-term fasting on bone metabolism and the contribution of acidosis to these changes, 14 healthy women ages 18-26 (mean, 21 +/- 2 (SD years) were randomized to potassium bicarbonate (KHCO3, 2 meq/kg/day in divided doses) to prevent acidosis or control (potassium chloride, 25 meq/day) during a complete 4-day fast. Bone turnover was assessed using specific markers of formation [osteocalcin (OC) and Type I procollagen carboxyl-terminal propeptide (PICP)] and resorption [pyridinoline (PYRX) and deoxypyridinoline (DPYRX)]. Serum bicarbonate levels fell significantly from 27.0 +/- 3.2 to 17.3 +/- 2.6 mmol/L (P < 0.01) in the control group and were decreased compared to patients receiving KHCO3 [17.3 +/- 2.6 vs. 23.4 +/- 2.4 mmol/L, (P < 0.001)]. Serum total and ionized calcium increased significantly in the control group [9.1 +/- 0.1 to 9.4 +/- 0.2 mg/dL (P < 0.01) and 1.20 +/- 0.03 to 1.23 +/- 0.03 mmol/L (P < 0.05), respectively], but not in patients receiving KHCO3. In addition, serum parathyroid hormone (PTH) levels decreased from 32 +/- 17 to 16 +/- 10 pg/mL (P < 0.05) and urinary calcium excretion increased [86 +/- 51 to 182 +/- 103 mg/day (P = 0.01)] in the control group, but not in patients receiving KHCO3. Serum osteocalcin (OC) and procollagen carboxyl-terminal propeptide (PICP) levels decreased significantly after 4 days of fasting from 9.1 +/- 3.4 to 5.5 +/- 4.2 ng/mL (P < 0.01) and 121 +/- 21 to 46 +/- 13 ng/mL (P = 0.0001) respectively in the patients receiving bicarbonate, and from 10.1 +/- 3.3 to 4.0 +/- 2.9 ng/mL (P < 0.01) and from 133 +/- 22 to 47 +/- 19 ng/mL (P < 0.001) respectively in the control group. The decrease in osteocalcin and PICP during fasting was comparable in both treatment groups. By contrast, urinary excretion of PYRX and DPYRX did not change significantly in either group with 4 days of fasting. These data are the first to demonstrate that markers of bone formation decline significantly with short-term fasting, independent of changes in acid-base status. By contrast, these data demonstrate a direct effect of acidosis in stimulating calcium release from bone during short-term fasting and suggest that acidosis may increase mineral dissolution independent of osteoclast activation and PTH in this experimental model of acute starvation.