This paper reviews the current data which provide a rationale for the measurement of cyclosporin as a guide to therapy. Methodological problems related to sample matrix and analytical technique are considered, and the most commonly used methods considered. Factors which could influence the clinical interpretation of cyclosporin measurements are examined, including other drug therapy, compliance with therapy, cyclosporin metabolites, pharmacokinetic variables and sample timing. It is concluded that, whilst isolated measurements do not offer a definitive diagnostic tool, taken in context they can be of considerable value in optimising therapy.