OBJECTIVES:To study the pharmacokinetics of netilmicin in Chinese haematology-oncology patients and to determine the pharmacokinetic differences, if any, between this patient subpopulation of Chinese and Caucasians. METHODS:A prospective study was carried out in the adult oncology unit of a major hospital in Hong Kong. During a 6 week period in 1997, all patients commencing on netilmicin therapy were monitored; the patients' demographics, clinical status, netilmicin dose and regimen, and drug administration/blood sampling time were collected. Pharmacokinetic parameters were generated using the USC*PACK package based on specifics of the patients themselves and Caucasians matched for the same patients' parameters using the Bayesian alogrithms. RESULTS:A total of 22 patients were enrolled into the study. Twenty-nine sets of levels were drawn, but only 25 sets from 18 patients (86%) were interpretable. The predicted peak (7.47+/-1.46 microg ml-1 ) and trough levels (1.39+/-0.96 microg ml-1 ) generated by USC*PACK were found to be significantly higher than the levels observed (6.01+/-1.14 microg ml-1 and 0.93+/-0.71 microg ml-1, respectively). Netilmicin clearance, volume of distribution and rate of elimination were all significantly higher in this Chinese subpopulation than those predicted for matched Caucasians. Conclusion Alterations in the netilmicin pharmacokinetics observed in our study population might be related to the disease state and/or ethics of the study patient population. Direct application of Caucasian based population pharmacokinetic parameters to this subgroup of Chinese patients may not be appropriate and may result in underdose.