INTRODUCTION: The clearances of oral and intravenous alfentanil are in vivo probes for assessing first-pass and hepatic cytochrome P450 (CYP) 3A activity. Clearance determinations require time-consuming and expensive serial blood samples and quantitative analysis. This investigation tested the hypothesis that a single concentration measurement can accurately predict alfentanil clearance and compared limited sampling results for alfentanil and midazolam. METHODS: Data were analyzed from 2 previous studies in healthy volunteers (n = 10 in each). By use of a dose-finding protocol, subjects received oral alfentanil (23, 30, 43, or 75 microg/kg). By use of a CYP3A modulation protocol, subjects received oral or intravenous alfentanil and midazolam after pretreatment with rifampin (INN, rifampicin) (CYP3A induction), troleandomycin (CYP3A inhibition), grapefruit juice (intestinal CYP3A inhibition), or nothing (control). Alfentanil and midazolam area under the concentration-time curve (AUC) was calculated by noncompartmental analysis. Correlations between AUC and plasma concentration at each time point were evaluated by unweighted and weighted (1/y, 1/y(2)) linear regression, and the optimal weight and sampling time were determined. RESULTS: Correlations between AUC and plasma concentration were best with 1/y(2) weights. Optimal correlations were at 1.5 to 2 hours for each alfentanil dose and also in an all-dose composite analysis. With the CYP3A modulation protocol, correlations for the control, grapefruit juice, rifampin, and troleandomycin sessions were optimal at 2, 3, 1.5, and 12 hours, respectively, for oral alfentanil, at 2, 3, 0.5, and 9 hours, respectively, for intravenous alfentanil, at 5, 5, 1.5, and 13 hours, respectively, for oral midazolam, and at 3, 1.75, 1, and 13 hours, respectively, for intravenous midazolam. In a composite analysis of all treatments, correlations (oral and intravenous) were optimal at 4 to 5 hours for alfentanil and 5 to 6 hours for midazolam. AUC estimated by a single-point concentration was highly predictive of measured AUC (r(2) = 0.91 and 0.93 for oral and intravenous alfentanil, respectively, and r(2) = 0.90 and 0.87 for oral and intravenous midazolam, respectively). CYP3A induction or inhibition was detected equally well by single-point alfentanil concentrations and by formal AUC determinations. CONCLUSION: A single plasma concentration can predict alfentanil AUC and hence clearance, independent of dose, and thus appears to be an effective and efficient strategy for assessing CYP3A modulation over a broad spectrum of enzyme activity.