Although ciclosporin is useful for atopic dermatitis (AD), appropriate dosage and therapeutic drug monitoring (TDM) has been performed only by post-prandial ciclosporin administration. We administered ciclosporin pre-prandially to eight severe adult AD patients (four cases of erythrodermic AD, three cases of AD recalcitrant to standard therapy, and one AD case with numerous pruriginous lesions). Blood concentrations of ciclosporin at various dosages were measured and appropriate dosage in terms of therapeutic efficacy was analyzed by using the area under the concentration curve (AUC). AUC was estimated by the C1 (obtained serum concentration of ciclosporin at 1 hour after ciclosporin administration), C2 (concentration of ciclosporin at 2 hours) and C4 (concentration of ciclosporin at 4 hours) concentrations of ciclosporin. The trough levels of ciclosporin with 200 mg/day, 150 mg/day, and 100 mg/day administration were 96.5 ng/ml, 66.4 ng/ml, and 75.3 ng/ml, respectively. The peak serum concentration (C(max)) was obtained at 1 hour (C1) in most cases. The AUC of 0-4 hours (AUC 0-4) were 2099.5 ng · h/ml (200 mg/day), 1782.6 ng · h/ml (150 mg/day) and 1696.2 ng · h/ml (100 mg/day). VAS scores of itching and blood eosinophil counts were decreased significantly by the ciclosporin treatment. Pre-prandial administration of a relatively low dose of ciclosporin for severe atopic dermatitis resulted in a favorable subjective and objective clinical response and the measurement of blood concentration mostly correlated with the effective dosage assessment.