Whole blood CsA concentrations measured by specific monoclonal RIA (CYCLO-Trac SP whole blood RIA, IncSTAR) were compared with episodes of renal dysfunction (n = 138) and protocol biopsies (n = 52) that occurred within the first 100 days in consecutive renal allograft recipients receiving triple therapy (n = 92). Histological confirmation of events was available in 98% episodes of acute rejection (n = 60/61), 59% of episodes of CsA nephrotoxicity (22/38), and 100% of the diagnoses of acute tubular necrosis (35/35). Mean, minimum, and maximum CsA levels were elevated in CsA nephrotoxicity compared with all other groups (P < 0.001). Interestingly, CsA levels achieved relative to administered dose also increased at the time of CsA nephrotoxicity compared with other groups (P < 0.01). In the context of acute dysfunction, the sensitivity and specificity of mean CsA levels above 400 ng/ml to predict CsA nephrotoxicity were 32% and 89%, respectively. The negative predictive value of a high CsA level to exclude acute rejection was 88% (at 400 ng/ml), 92% (450 ng/ml), and 95% (500 ng/ml). As a marker of effective immunosuppression, CsA levels were not correlated with in vitro proliferation of PHA-stimulated PBL and did not reduce the severity and degree of cellular infiltration in needle core biopsies during rejection. The sensitivity and specificity of a low CsA level (150 ng/ml) in the diagnosis of acute rejection were 31% and 91%, respectively. The majority of episodes of acute dysfunction, including 63% of CsA nephrotoxicity and 59% of acute rejections, occurred with CsA levels between 150 and 400 ng/ml. In summary, when using low dose triple therapy regimens, CsA levels within the range of 150-400 ng/ml were of little diagnostic value in acute allograft dysfunction. In contrast, levels outside this range were useful in the clinical diagnosis of CsA nephrotoxicity and acute allograft rejection.