Osteonecrosis (avascular necrosis) in inflammatory bowel disease: an extraintestinal clinical feature, not caused by corticosteroids

The cause of osteonecrosis (AVN, avascular necrosis) in inflammatory bowel disease (IBD) remains an intriguing puzzle. An extensive meta-analysis of 15 selected studies, detailed here, includes more than 100,000 patients and provides further evidence that corticosteroids are not responsible for AVN in IBD. Other factors may contribute to AVN including cigarette smoking and chronic alcohol use as well as other coexisting disorders, especially with vasculitis, thrombocytosis, or altered blood viscosity. Pathological studies have previously shown the histological ‘footprint’ of Crohn’s disease in the gastrointestinal tract, specifically, granulomatous inflammation in resected necrotic bone with multinucleated giant cells. These studies show that IBD represents a risk factor for AVN, unrelated to treatment with corticosteroids. Bhayana et al. [1] provide a systematic review of osteonecrosis (AVN, avascular necrosis) in inflammatory bowel disease (IBD), including an extensive meta-analysis of 15 selected studies containing over 100 000 individuals in an electronic database. The overall objective was to estimate the prevalence of AVN in IBD and the possible role of corticosteroid use in appearance of AVN in these IBD patients. The results showed a prevalence rate of 10.39 per 1000 IBD patients but lower in the larger studies examined. Importantly, AVN risk could not be attributed to steroid use in these IBD patients. Of course, in adults, more than 10% of all joint replacements have been estimated to be due to osteonecrosis. Historically, AVN was reported as early as 1957 [2] in association with corticosteroid use, and later, in the English language literature [3]. Most had undergone renal transplantation or treatment for systemic lupus erythematous (SLE) and received high doses of steroids [4–6]. Although steroids were used to treat other disorders, the development of AVN was rare, except possibly in malignant lymphoma [7]. Moreover, AVN could not be induced by corticosteroids alone in experimental animals [8]. In IBD, AVN was initially described in rare cases [9–15]. Some, but not all [14,15], were treated with steroids but a consistent association with steroid treatment could not be defined. Furthermore, no correlation to steroid dose in IBD (as suggested in renal transplantation and SLE cases) could be determined. Finally, there was no apparent temporal relationship to the time of steroid treatment and the appearance of AVN in these early cases. In some IBD patients, AVN actually appeared years after steroid treatment had been terminated, or never received [16]. Other considerations for cause of AVN were then raised. For example, other concomitant or coexistent diseases were noted including pancreatitis, chronic alcohol abuse, trauma (possibly with fat embolism), and treatment with lipid emulsions in parenteral nutrition solutions [17]. Even cigarette smoking, common in Crohn’s disease patients, was previously noted to significantly increase AVN risk, independent of steroid use [18]. In this report [1], Bhayana et al. may have also implicated other possible treatments in AVN pathogenesis in IBD. One of these, a prior report from Belgium [19], described AVN of facial bones in IBD, similar to an earlier case report [20]. Although not necessarily included in their treatment analysis, tumor necrosis factor-alpha antibodies could have been indirectly implicated in AVN pathogenesis. Others suggested that the disease, itself, was responsible. In IBD, particularly Crohn’s disease, coagulopathy associated with hyperviscosity and thrombocytosis [21] or vasculitis [22] may be important. Based on pathological study in one patient of a resected hip, the well-defined histological ‘footprint’ of Crohn’s disease was recorded, specifically, granulomatous inflammation with bone necrosis and multinucleated giant cells in an extraintestinal site [23]. For the future, Bhayana et al. [1] may have reopened the door to further important studies on the cause and pathogenesis of AVN in IBD. More precise and added data, perhaps, beginning with their database and focused on these issues may be critical. Acknowledgements None. Conflicts of interest There are no conflicts of interest.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.