A nationwide cohort study of inflammatory bowel disease, histological activity and fracture risk

Abstract

Background: Individuals with inflammatory bowel disease (IBD) are at increased risk of fracture. It is unclear if this risk varies by recent histological activity.

Aims: To determine the fracture risk in IBD during periods with and without histological inflammation.

Methods: We studied a nationwide cohort of 54,591 individuals diagnosed with IBD in 1990-2016 with longitudinal data on ileo-colorectal biopsies. Fractures were identified by inpatient and hospital-based outpatient diagnoses. We derived Cox regression estimated hazard ratios (HRs) for fracture during 12 months following a histological inflammation (vs. histological remission) record after adjusting for socio-demographics, comorbidities, IBD duration, IBD-related surgery and hospitalization. We adjusted sensitivity analyses for medical IBD treatment including corticosteroids.

Results: Mean age of patients was 44.0 (SD = 18.3) and 45.5 (SD = 17.1) years at biopsy with histological inflammation and remission, respectively. For histological inflammation, there were 1.37 (95% CI 1.29-1.46) fractures per 100 years' follow-up versus 1.31 (95% CI 1.19-1.44) for remission (adjusted [a]HR 1.12; 95% CI 1.00-1.26; p = 0.04). HRs were similar with histological inflammation of Crohn's disease (1.11; 95% CI 0.91-1.36) and ulcerative colitis (1.18; 95% CI 1.02-1.36). Estimates were consistent across age groups. An overall small excess risk of any fracture remained after accounting for corticosteroids. A more prominently raised fracture risk was observed in corticosteroid-naïve IBD patients with histological inflammation versus histological remission (aHR 1.41; 95% CI 1.07-1.85). The aHR of hip fracture following histological inflammation was 1.29 (95% CI 0.87-1.92).

Conclusions: Histological inflammation in IBD predicted a small increase in short-term fracture risk. Measures to reduce disease activity may reduce fracture risk in IBD.

FIGURE 1

Risk of fracture during 0 to <12 months of histological inflammation versus histological remission in inflammatory bowel disease (IBD). *Adjusted for age at biopsy (i.e. the start of the exposure period), sex, calendar year, education level, country of birth, disease duration, hospital admission for IBD, IBD‐related surgery and the Charlson comorbidity index. Histological inflammation and histological remission as defined by biopsies originating from the ileum or colorectum (topographic codes T65, T67 and T68) with histology codes listed in Table S5. Biopsy date equals the start of the exposure period. Montreal classifications of extent and location of disease at diagnosis are detailed in Table S2. CI, confidence interval; CD, Crohn's disease; IR, incidence rate; IRR, incidence rate ratio; IBD‐U, Inflammatory bowel disease‐unclassified; PY, person‐years; UC, ulcerative colitis.

FIGURE 2

Risk of hip fracture during 0 to <12 months of histological inflammation versus histological remission in inflammatory bowel disease (IBD). *Adjusted for age at biopsy (i.e. the start of the exposure period), sex, calendar year, education level, country of birth, disease duration, any history of inpatient IBD care, IBD‐related surgery and the Charlson comorbidity index. Histological inflammation and histological remission are defined by biopsies originating from the ileum or colorectum (topographic codes T65, T67 and T68) with histology codes listed in Table S5. Biopsy date equals the start of the exposure period. Montreal classifications of extent and location of disease at diagnosis are detailed in Table S2. CI, confidence interval; CD, Crohn's disease; IR, incidence rate; IRR, incidence rate ratio; IBD‐U, inflammatory bowel disease‐unclassified; PY, person‐years; UC, ulcerative colitis.