Arithmetic Versus Harmonic Mean Values for Cyclosporin‐A Pharmacokinetic Parameters

The harmonic and arithmetic mean values for volume of distribution at steady state, half-life, and clearance of intravenous cyclosporin-A were calculated using an index set of 22 renal transplant candidates to determine if harmonic mean values provide a less biased estimate of central tendency than arithmetic mean values. Cyclosporin-A was measured using a nonspecific radioimmune assay method. The arithmetic mean value for volume of distribution was 16% larger than calculations by the harmonic mean method. Similarly, the arithmetic mean half-life and clearance values were larger than harmonic mean values by 10% and 15%, respectively. However, 95% confidence intervals for these pharmacokinetic parameters overlapped. When these mean pharmacokinetic parameter values were used to predict actual values in a test group of 22 renal transplant candidates receiving cyclosporin-A, similar levels of precision were demonstrated by either method. Both methods produced positively biased predictions for volume of distribution and clearance. However, these differences were not significant. These findings suggest there is little practical value for the use of harmonic mean calculations to describe the central tendency of pharmacokinetic parameters of cyclosporin-A under the conditions studied. The value of harmonic mean values for pharmacokinetic parameters in other patient populations or with other assay methods for cyclosporin remain to be studied.