The aim of the present study was to identify factors determining the delivery to and distribution of aerosolized cyclosporine A (CSA) in the lungs of patients with severe pulmonary allograft rejection. Five such patients inhaled a previously characterized radioaerosol consisting of 4 to 6 cc of CSA (50 mg/ml) in ethanol mixed with technetium-99m (99mTc) bound to human serum albumin, generated by an AeroTech II nebulizer. The total dose of CSA depositing in the lungs was determined with a previously described inspiratory/expiratory mass-balance filter technique. Regional distribution of drug within the lungs was measured using a gamma camera. In addition, the following physiologic parameters were measured: regional volume and ventilation using xenon-133 (133Xe) equilibrium and 133Xe washout, respectively, and regional perfusion using intravenous 99mTc macroaggregates. The relationships between these parameters and regional drug deposition were assessed using linear regression analysis. The lung dose ranged from 20 to 53 mg (0.097 to 0.175 mg CSA deposited per milligram placed in nebulizer). In recipients of single-lung allografts, preferential drug deposition occurred either in the allograft (two patients) or in the native lung (one patient). Marked nonuniformities in regional distribution were also apparent in two double-lung allograft recipients. There was a weak but statistically significant correlation between regional drug deposition and regional ventilation, as measured by 133Xe washout (r = -0.542, p = 0.014), suggesting that although regional ventilation is important, it is not the only factor determining regional deposition in these patients. The correlation between regional drug deposition and regional volume, as measured by 133Xe equilibrium, did not achieve statistical significance (r = 0.433, p = NS). There was, however, a close correlation between regional drug deposition and regional perfusion (r = 0.891, p < 0.0001). Perfusion may therefore be an index of functional, ventilated areas with adequate surface area for deposition. In conclusion, because the variability in regional drug deposition in these patients appears to be due to the interaction of regional ventilation and regional anatomy, direct measurement of regional deposition will be necessary in initial studies of the efficacy of aerosol therapy.