Multimorbid degenerative lumbar spinal stenosis with knee or hip osteoarthritis: prevalence and associated characteristics

Research output: ThesisPh.D. thesis

39 Downloads (Pure)


The overall aim of this PhD thesis was to provide foundational data to help answer the questions of how often multimorbid lumbar spinal stenosis (LSS) with knee or hip osteoarthritis (OA) occurs and what characteristics are associated with this multimorbid presentation. To answer these questions, three studies were conducted:

Study I was a systematic review with the primary objective of investigating the prevalence of multimorbid LSS with knee or hip OA based on varying case definitions. This review found there
is an overall lack of available literature with sufficient data to estimate the multimorbid prevalence, but that published studies suggest 5-54% of people may have coexisting LSS and knee OA and 0-35% of people may have coexisting LSS and hip OA. The majority of estimates are derived from samples of people with LSS and prevalence estimates are highly dependent on the case definitions used for both LSS and OA. The secondary objective was to investigate factors associated with multimorbid prevalence estimates, but the dearth of published literature using similar case definitions for LSS and OA prevented the conduct of metaregression analyses. The study protocol has been published in Systematic Reviews (DOI:10.1186/s13643-020-01478-4) and the final study has been accepted for publication in BMC Musculoskeletal Disorders (DOI:10.1186/s12891-022-05104-3).

Study II was based on data from the Good Life with osteoArthritis in Denmark (GLA:D®) database including people seeking treatment for knee or hip OA. Participants completed 11 self-report questions on symptoms commonly associated with LSS. We found that 8-40% of participants with knee OA self-reported LSS symptoms, depending on the specific symptom. Likewise, 11-50% of people with hip OA self-reported the different LSS symptoms. Using these symptoms to operationalize adapted versions of two published clinical criteria for symptomatic LSS, less than 3% and 5% of people with knee and hip OA were considered to have clinical LSS, respectively. This study has been published in Osteoarthritis and Cartilage (DOI:10.1016/j.joca.2021.07.012).

Study III was based on the same GLA:D® cohort of people seeking care for knee or hip OA. Here, we aimed to identify characteristics associated with comorbid LSS symptoms. We performed multiple multivariable logistic regression analyses in each cohort independently, using comorbid LSS symptom status as the dependent variable. We also conducted sensitivity analyses to ensure our findings were robust to a different definition of comorbid LSS symptoms and different model building procedures. In both cohorts, we found consistent evidence across models and sensitivity models that sick leave in the past year, back pain in the last month, and knee or hip symptom duration greater than 2 years, were associated with comorbid LSS symptoms. We also found a similar pattern of other associated characteristics between cohorts, but these characteristics did not show consistent associations across all models and sensitivity models. This study is currently in review for publication.

Overall, this thesis found a significant proportion of people may experience multimorbid LSS with knee or hip OA and has uncovered patient characteristics associated with multimorbid
presentations. Future studies accounting for the methodological limitations uncovered in this thesis are needed to better inform the clinical care of people with multimorbid LSS with knee or hip OA.

Original languageEnglish
Awarding Institution
  • University of Southern Denmark
  • Hartvigsen, Jan, Principal supervisor
  • Roos, Ewa Maria, Supervisor
  • Jensen, Rikke Krüger, Supervisor
  • Ammendolia, Carlo, Supervisor, External person
Publication statusPublished - 11. May 2022


Dive into the research topics of 'Multimorbid degenerative lumbar spinal stenosis with knee or hip osteoarthritis: prevalence and associated characteristics'. Together they form a unique fingerprint.

Cite this