Kevin A. KarlsEdward V. LoftusWilliam J. TremaineMiguel E. CabanelaLee J. MeltonWilliam J. Sandborn
Purpose: Osteonecrosis, or avascular necrosis (AVN), is a rare complication of inflammatory bowel disease (IBD). We sought to characterize the relationship between AVN, IBD, and corticosteroid therapy. Methods: A centralized diagnostic index identified all patients (pts) with IBD and AVN evaluated at Mayo Clinic Rochester between 1976 and 2003. Medical records were abstracted for pt demographics, subtype and duration of IBD, treatment received including corticosteroids, other potential AVN risk factors, AVN stage and joints involved, and AVN treatment and outcomes. Results: We identified 52 ulcerative colitis patients (71% male) and 42 Crohn's disease patients (52% male). Median age at AVN diagnosis (dx) was 45.9 years (yrs) (range, 16.5–82.2), and median duration from IBD to AVN dx was 9.2 yrs (0.5–41.2). Of all 94 pts, only 6 pts (6%) had no history of systemic corticosteroid exposure. Steroid-naïve IBD-AVN pts were older at both IBD dx (51.8 yrs) and AVN dx (67.3 yrs) than steroid users. All 88 steroid users had received oral corticosteroids, and 26 had also received IV steroids (24%). Among pts with available data, median cumulative duration of steroid therapy prior to AVN dx was 12.2 months (range, 2–70.2), median daily prednisone dose was 20.4 mg (12.1–50), and median cumulative prednisone dose was 9,205 mg (2,770–42,660). Overall, 41% had a prior arthropathy and 40% had a history of smoking, with similar rates for those with and without corticosteroid use. The mean number of joints involved was 2.3 among those who had received steroids (55% femoral head, 23% upper extremities) versus 1.2 among those who had not (57% femoral head, 0% upper extremities). The prevalence of stage III-IV AVN was 77% in steroid users versus 67% in the steroid-naïve. Median follow-up time after AVN dx was 4.4 yrs (0–27.3). Treatment included arthroplasty in 36%, conservative therapy in 20%, core decompression in 10%, and bone grafting in 7%. Despite treatment, 16% of pts developed worsening symptoms, and 17% had AVN involvement of new joints—none of this progression was seen in the 6 steroid-naïve pts. Conclusions: Osteonecrosis is a rare but serious complication seen in IBD. In a minority of pts, AVN may occur in the absence of corticosteroid therapy. These pts are older at IBD and AVN dx, have fewer joints involved, and have less progression of the joint disease. The etiology of AVN may be multifactorial with contributions from IBD itself, systemic corticosteroid therapy, and other AVN risk factors.
Gregory G. KlingensteinAsit K. ShahRoger LévyDaniel H. Present
Siddhant YadavEdward V. LoftusWilliam J. TremaineJithinraj Edakkanambeth VarayilKevin A. Karls
R. M. CarterMichael G. A. Grace