OBJECTIVE To evaluate epicutaneous application of 5% lidocaine-prilocaine and 30% lidocaine cream anesthetics for neonatal circumcision. METHODS The efficacy of 5% lidocaine-prilocaine and 30% lidocaine creams was compared in a randomized, double-blind, placebo-controlled trial. Sixty-one neonates were randomly assigned to one of three groups: 5% prilocaine-lidocaine (n = 20), 30% lidocaine (n = 20), and a control group that received an acid-mantle cream (n = 21). Heart rate, oxygen saturation, and crying time were monitored before, during, and after circumcision. Blood pressure was measured before and after circumcision. RESULTS Mean peak heart rates for the 5% lidocaine-prilocaine, 30% lidocaine, and control groups (+/- standard deviation) were 146 +/- 16, 157 +/- 10, and 164 +/- 16 beats per minute, respectively. During four of six active phases of circumcision, the 5% lidocaine-prilocaine group suppressed significant increases in heart rate better than 30% lidocaine, which was more effective than control (dorsal clamp, P < .001; bell clamp on, P = .001; tightening, P = .001; bell clamp off, P < .001). During tightening of the bell clamp, significantly less crying was seen in the 5% lidocaine-prilocaine group (13 +/- 12 seconds) compared with 30% lidocaine (24 +/- 14 seconds) and controls (38 +/- 27 seconds) (P < .001). The group that received 5% lidocaine-prilocaine also had no significant increase in systolic (t = 1.6, P = .12) or diastolic (t = 1.9, P = .067, respectively) blood pressure, unlike the group receiving 30% lidocaine (t = 4.8, P = .001 and t = 2.9, P = .009, respectively) and the placebo group (t = 2.5, P = .023 and t = 2.3, P = .032). There were no significant differences in oxygen saturation (alpha = .05, power 0.79). CONCLUSIONS Epicutaneous 5% lidocaine-prilocaine was more effective than 30% lidocaine for neonatal circumcision, better reducing neonatal stress indicators. Lidocaine-prilocaine significantly shortened crying time during one of the most painful phases of circumcision. Both topical anesthetics were more effective than placebo in attenuating the behavioral and physiologic indicators of neonatal pain.