Article Text
Abstract
Introduction Corticosteroid use, cirrhosis and inflammatory disease are risk factors for osteoporosis. However the occurrence of osteoporosis in Autoimmune Hepatitis (AIH) has been little studied. Our aims were to evaluate retrospectively the effects of the disease and its treatment on bone mineral density (BMD) in AIH and whether the use of bisphosphonates prevents bone loss.
Methods 108 patients, median (range) age 60 (18–80) with AIH for 7 (1–416) months were included. BMD was measured using dual-energy x-ray absorptiometry (DXA) at the hip and/or lumbar spine. Results are compared to manufacturer's reference population and expressed as T- and Z-scores. To assess the effect of intervention, we studied a subgroup of 37 prednisolone-treated patients who had a second DXA, with the first performed at 3 (1–6) months and the second at 28 (15–67) months from diagnosis. Patients were prescribed a bisphosphonate after the 1st DXA if clinically indicated.
Results 20% of patients had osteoporosis and 30% had osteopenia. The mean ± SD Z-score at the hip was 0.1±1 and at the spine 0.1±1.5, which was not significantly different from that of the reference population (p=0.4 for both). Compared to those with normal BMD, patients with osteoporosis had higher fibrosis stage at diagnosis (median 3 vs 5, p=0.006). On multivariate analysis, hip Z-score showed independent associations: negative with log cumulative prednisolone dose (r=−0.25, p=0.02) and positive with body weight (r=0.4, p≤0.001), but there was no association with disease duration. 19 patients had fragility fractures. Patients with a fracture had lower hip and spine T-scores compared to patients without (hip T-score −1.8 vs −0.6, p=0.006, spine T-score −2.2 vs −1.0, p=0.015). In the paired DXA analysis, BMD in hip and spine increased in patients (n=18) commenced on a bisphosphonate after the 1st DXA compared to those who were not (% change in BMD/year at hip +0.3 vs −1.7, p=0.002 and at spine +2.1 vs −2.0, p=0.002), despite the two groups receiving similar current and cumulative prednisolone doses.
Conclusion Mean BMD in treated AIH is not lower than the expected level for age. Prednisolone dose-related bone loss occurs but can be prevented with appropriate use of bisphosphonates.
Competing interests None declared.