The pharmacologic effects of intrathecal sufentanil (ITS) beyond what is clinically administered (10 microg) are not known. We observed 18 healthy, young, adult female volunteers who received 12.5, 25, or 50 microg of ITS in a randomized, double-blind fashion for 11 h. Analgesia was assessed by pressure algometry at the tibia. Respiratory function was assessed by pulse oximetry, respiratory rate, arterial blood gas, the ventilatory response to CO2, and a respiratory intervention score (RIS). The incidence and severity of side effects also were documented. Serum sufentanil levels were measured for 4 h after ITS administration. We found that ITS produced statistically significant changes in algometry, doubling the pressure required to produce moderate pain. However, doses of ITS greater than 12.5 microg failed to produce proportionate increases in the duration or intensity of analgesia. All doses of ITS produced significant respiratory depression, but only the RIS was significantly related to ITS dose. Neither respiratory rate nor sedation reliably predicted hypoxemia. Supplemental oxygen by nasal cannula consistently prevented pulse oximeter readings below 90%. Serum sufentanil concentrations were related to ITS dose in a statistically significant manner, reached clinically significant concentrations, and followed a time course similar to analgesia and measures of respiratory depression. However, there was no significant increase in measured analgesia associated with the increases in serum sufentanil concentrations. We conclude that in our volunteer model of lower extremity pain, administering ITS in doses larger than 12.5 microg does not improve the speed of onset, magnitude, or duration of analgesia and only causes dose-related increases in serum sufentanil concentrations, which may augment respiratory depression.