Background and aim: Alternative splicing of human glucocorticoid receptor (hGR) premessenger RNA (mRNA) generates two highly homogenous isoforms, termed hGRalpha and hGRbeta. hGRalpha is a ligand-activated transcription factor, which, in the hormone-bound state, modulates the expression of glucocorticoid-responsive genes by binding to specific glucocorticoid response element DNA sequences. In contrast, hGRbeta may be an endogenous inhibitor of glucocorticoid action and transcriptionally inactive. hGRbeta protein has been known to correlate with the development of glucocorticoid resistance. Glucocorticoids can effectively relieve autoimmune hepatitis (AIH), but some patients with this disease are refractory even when glucocorticoids are administered. The aim of this study was to determine the incidence of hGRbeta mRNA in patients with AIH by reverse transcription polymerase chain reaction (RT-PCR), and to compare the clinical characteristics of hGRbeta-positive and -negative patients with AIH. Materials and methods: RNA was obtained from peripheral blood mononuclear cells (PBMCs) of 62 patients, consisting of 26 with AIH, 10 with primary biliary cirrhosis (PBC), seven with chronic viral hepatitis (CVH), 10 with ulcerative colitis (UC), six with pemphigus, and three with systemic lupus erythematosus (SLE), and 10 healthy volunteers. The total RNA obtained was reverse transcribed, the resulting complementary DNA amplified using specific primers for hGRalpha and hGRbeta. Results: The hGRalpha mRNA was detected in RNA from PBMCs of all patients and healthy volunteers. The hGRbeta mRNA was detected in 15 (57.6%) patients with AIH. This incidence was significantly higher than that for patients with PBC (0%) or CVH (28.6%) or for healthy volunteers (20.0%) ( [Formula: see text] ). Of the hGRbeta-positive and -negative groups, serum ALT and total bilirubin (TB) levels were significantly higher in the positive group ( [Formula: see text] ). The total dose of glucocorticoid was higher in the positive group, but the difference was not statistically significant. However, the average monthly dose was significantly higher in the positive group ( [Formula: see text] ). The rate of relapse of AIH was significantly higher in the positive group (60.0%) than in the negative (10.0%) ( [Formula: see text] ). The rate of usage of immunosuppressive drugs was higher in the positive group (33.3%) than in the negative (18.2%), but the difference was not statistically significant. Conclusions: These data show that hGRbeta expression in PBMCs of patients with AIH assessed by RT-PCR is closely associated with resistance to glucocorticoids which affects the outcome of therapy with this drug.