A pilot study was undertaken to compare the efficacy of two regimens of intravenous patient-controlled analgesia (PCA) with remifentanil for labour analgesia. Twenty term parturients requesting labour analgesia were randomized to receive one of two regimens of intravenous remifentanil. The initial settings in both groups consisted of an infusion of 0.025 μg·kg-1·min-1, a PCA bolus of 0.25 μg·kg-1 and a lockout interval of two minutes. In Group A, the infusion was increased in a stepwise manner from 0.025 to 0.05, 0.075 and 0.1 μg·kg-1·min-1 as required; the bolus was kept constant at 0.25 μg·kg-1. In Group B, the bolus was increased from 0.25 to 0.5, 0.75 and 1 μg·kg-1 as necessary; the infusion was kept constant at 0.025 μg·kg-1·min-1. Maternal pain, satisfaction and sedation scores, remifentanil requirement, and side effects were recorded. Mean pain and patient satisfaction scores, and cumulative doses of remifentanil were similar in the two groups. The overall incidence of side effects was greater in Group B (P = 0.0007), with drowsiness observed in 100% of patients, as compared to 30% in Group A (P = 0.003). The minimum oxygen saturation levels were 94.3% ± 2.6% and 92.2% ± 3.8% in Groups A and B respectively (P = 0.19). Although pain and satisfaction scores were similar in both groups, the regimen used in Group A was associated with fewer side effects compared to the Group B dosing regimen. This pilot study suggests that remifentanil intravenous PCA is efficacious for labour analgesia as a bolus of 0.25 μg·kg-1, with a lockout interval of two minutes and continuous infusion of 0.025-0.1 μg·kg-1·min-1. The potential for respiratory depression mandates close respiratory monitoring. Large-scale trials to evaluate safety issues are warranted.