We attempted to determine the optimal test dose of epinephrine for use with epidural anesthesia in awake patients premedicated with clonidine. Eighty-eight adult patients were randomized into two groups [oral premedication with clonidine 5 mu g/kg (CLON) or no premedication (CONT)]. Before induction of general anesthesia, heart rate (HR) and blood pressure (BP) were measured for 3 min after the IV injection of 3 mt of 1.5% lidocaine containing epinephrine (0, 1.25, 2.5, 5, 7.5, or 15 mu g) in a randomized, double-blind manner. We calculated 95% confidence intervals for the peak HR and BP increases induced by each dose of epinephrine. At 7.5 mu g, epinephrine induced a significantly greater increase in HR and BP in CLON than in CONT. The 95% confidence interval for the HR change induced by 7.5 mu g of epinephrine in CLON was nearly the same as the accepted standard dose of epinephrine (15 mu g) in CONT. We conclude that premedication with clonidine enhances HR and BP responses to the IV administration of epinephrine-containing epidural test solutions. Consequently, 7.5 mu g of epinephrine may be sufficient to enable detection of accidental injection into a blood vessel in awake patients premedicated with clonidine 5 mu g/kg. Implications: Clonidine, a commonly used preanesthetic medication, alters patients' cardiovascular responses to drugs such as epinephrine. Our randomized, double-blind study suggests that, in awake patients receiving oral clonidine premedication, 7.5 mu g of epinephrine (half the usual dose) is adequate as an indicator of accidental injection into the epidural vessels during epidural anesthesia.