In this pilot randomized, double-blind, cross-over study, the effectiveness and safety of hydromorphone administration by continuous subcutaneous (s.c.) infusion (mode A) and by continuous basal rate s.c. infusion + PCA (mode B) were compared in 8 cancer patients. Patients experimented with each infusion mode during 48 h. Statistical analysis was performed on data collected in 7 patients during 36 h from 22:00 h on day 1 to 10:00 h on day 3 and from 22:00 h on day 3 to 10:00 h on day 5. Mean hydromorphone dose +/- S.D. was 56.6 +/- 30.1 and 40.4 +/- 24.5 mg/36 h for modes A and B, respectively. There was no statistically significant difference observed in mean pain intensity, but the absence of significant difference may be related to the small sample size and high individual variability. Both methods provided adequate overall pain control in most patients. However, a large interindividual variation was detected. Indeed, some patients reported in the subjective questionnaire that they felt marked discomfort during hydromorphone administration with mode B. Only 2 patients chose mode B at the end of the study, but it was interesting to note that those 2 patients were the youngest of the group. This study demonstrated the effectiveness and safety of both modes of hydromorphone administration. The data suggest that it may be possible to identify particular cancer patients which can really benefit from an association of a basal rate infusion and PCA for opiate administration.