In this study we retrospectively reviewed results of the first 9 days of treatment with pamidronate at doses of 30 mg (n = 13), 45 mg (n = 9), and 90 mg (n = 13) in an attempt to see what factors influenced the response of serum calcium to pamidronate. The nadir of serum calcium obtained post treatment was correlated with pretreatment levels of nephrogenous cyclic adenosine monophosphate (NcAMP), the renal tubular threshold for phosphate reabsorption (TmPO4), and the renal tubular threshold for calcium reabsorption (TmCa). Using the post treatment serum calcium levels, patients were divided into "good" and "poor" responders depending on whether a normal serum calcium was obtained. Pretreatment NcAMP was significantly correlated with the magnitude of the response of serum calcium (r = 0.45, P = 0.0001). Pretreatment NcAMP was significantly higher in the poor responders (mean +/- SEM): 65.0 +/- 9.4 nmol/liter GF (poor responders) versus 29.6 +/- 6.3 (good responders), P = 0.004. NcAMP as a predictor of the acute response of serum calcium showed a sensitivity of 93% and a specificity of 72%. Pretreatment TmPO4 was negatively correlated with the serum calcium response post treatment (r = -0.41, P = 0.003). However, though TmPO4 tended to be lower in the poor responders, this was not statistically significant [0.65 mmol/liter GF +/- 0.09 (poor responders) versus 0.76 mmol/liter GF +/- 0.06 (good responders)]. As a predictor of the acute response of serum calcium, TmPO4 was less good with a sensitivity of 70% and specificity of 58%. No significant correlation was present between TmCa and the serum calcium response.(ABSTRACT TRUNCATED AT 250 WORDS)