Osteoporosis in adults with axial spondyloarthritis

Abstract

Osteoporosis was the most prevalent comorbidity in individuals with axial spondyloarthropathy (axSpA) in a large multinational study. Ireland was not included in that study, so it was unknown if the profile of comorbidities seen internationally was the same in the Irish axSpA population. The aim of this thesis was to comprehensively examine osteoporosis in an Irish axSpA cohort.

In pursuing this aim, four cross-sectional studies, a narrative review and a systematic review and meta-analysis were carried out. Obesity was the most prevalent comorbidity in cross-sectional Study 1, affecting 27% of individuals, using Ankylosing Spondylitis Registry of Ireland (ASRI) data, with 734 participants. Less than 20% of individuals had been assessed for osteoporosis; prevalence in those assessed was 23%.

A subsequent narrative review identified that conventional methods of assessing bone mineral density (BMD) of the spine are inaccurate in individuals with axSpA, due to interference from structural damage. This formed the basis of the following three cross-sectional studies.

Study 2 compared lateral and posterior-anterior (PA; conventional) projections of dual-energy x-ray absorptiometry (DXA) in their ability to examine BMD of the spine. A cohort of 100 individuals were included in this study. Lateral DXA of the spine was significantly lower than PA DXA in this population. Structural damage interfered with the ability of PA, but not lateral, DXA to assess the spine. There was a higher prevalence of low BMD in women and human-leucocyte antigen (HLA)-B27 positive individuals.

Study 3 explored the association between biomarkers and BMD in adults with axSpA. Increased bone turnover correlated with lower BMD at all sites measured. Testosterone inversely correlated with BMD at PA spine. Higher serum urate was associated with higher BMD. All relationships between biomarkers and osteoporosis lost significance when important confounders were controlled for.

Study 4 aimed to assess if quantitative ultrasound (QUS) of the heel was suitable as an osteoporosis triage tool in adults with axSpA. QUS identified those individuals who were at low risk of having low BMD with 90% confidence, which would have allowed 27% of DXAs to be avoided. QUS was unable to confidently identify individuals with osteoporosis.

Study 5 was a Systematic Review & Meta-Analysis, which aimed to appraise and synthesise randomised controlled trials (RCTs) and quasi (q)-RCTs examining the efficacy of pharmacological and non-pharmacological interventions on BMD in adults with axSpA. Moderate level evidence supported a conditional recommendation for the use of alendronate for low BMD of the femoral neck. The balance of evidence did not support the use of tumour necrosis factor inhibitors (TNFi) at either spine or hip. We identified no RCTs examining non-pharmacological interventions. There is a need for high-quality RCTs examining both pharmacological and non-pharmacological interventions on BMD in axSpA.