Abstract
Objective We aimed to explore the prevalence of degenerative disc disease (DDD) in patients with psoriatic arthritis (PsA) aged < 50 years and to describe the factors associated with its development. We also examined the association between radiographic axial imaging findings and inflammatory back pain (IBP) and mechanical back pain.
Methods We included patients with PsA aged < 50 years who were followed at our prospective observational cohort. We defined DDD as intervertebral disc space narrowing, spur formation, facet joint arthrosis, and spondylolisthesis on anteroposterior and lateral plain radiographs of the cervical and thoracolumbar spine. To identify factors associated with the development of DDD, we used multivariate Cox regression analysis. We used generalized estimating equations (GEEs) to test the association between imaging findings (isolated DDD, isolated axial disease, and both) and the type of back pain.
Results Of 814 patients included in the study, 316 (38.8%) were observed to have DDD on plain radiographs of the spine. Factors associated with the development of DDD included older age (hazard ratio [HR] 1.08, P < 0.01), male sex (HR 1.52, P = 0.03), diabetes mellitus (HR 2.35, P = 0.045), and IBP (HR 2.03, P < 0.01). Being employed (vs unemployed), higher BMI, calcaneal spurs, and targeted disease-modifying antirheumatic drug use showed a trending association with DDD. In the GEE analysis, none of the abnormal imaging findings were significantly associated with back pain or IBP.
Conclusion DDD is common in young patients with PsA, and its development may be associated with demographic features, comorbidities, and disease-related factors. IBP does not reliably distinguish between axial PsA and DDD.
Psoriatic arthritis (PsA) is a complex inflammatory disease with heterogeneous clinical features that may complicate skin or nail psoriasis.1 The disease can have peripheral and axial musculoskeletal manifestations, including arthritis, enthesitis, dactylitis, spondylitis, and sacroiliitis.1,2
It is well established that the incidence of degenerative disc disease (DDD) increases with age.3 For instance, it affects 10% of the male population at the age of 50 years and up to 50% at the age of 70 years.3 However, initial degeneration of the intervertebral discs has been reported to present ultrastructurally as early as adolescence.4,5 Additionally, some analyses emphasize the common occurrence of disc degeneration in imaging studies starting in the fourth and fifth decades of life.6,7
In general, spine conditions can lead to both inflammatory back pain (IBP) and mechanical back pain (MBP), which may share overlapping features.8,9 Back symptoms are common in PsA and tend to be of an inflammatory nature.8,10 Depending on the criteria used, 25% to 70% of patients with PsA may have axial involvement, which is characterized by inflammation and postinflammatory structural changes of the spine and/or sacroiliac joints.10 However, patients with PsA may also experience MBP, often resulting from injuries or derangements of spinal anatomical structures.11,12 It is now widely recognized that identifying axial PsA based solely on the type of back pain is unreliable.8,13,14
Despite the widely accepted understanding that inflammatory arthritides increase the risk of developing degenerative disease,15 previous studies have not adequately addressed the occurrence and significance of DDD in young adults with PsA. Therefore, this study aimed to characterize early-onset DDD in patients with PsA aged < 50 years and to explore its associations with back pain, demographic factors, and disease features.
METHODS
Setting. The study was conducted at the Gladman-Krembil Psoriatic Arthritis Research Program, where patients with PsA are recruited and prospectively followed. Patients are enrolled if they have psoriasis and inflammatory musculoskeletal disease. Over 99% fulfill the Classification Criteria for Psoriatic Arthritis (CASPAR).16 Patients are evaluated at the time of recruitment into the clinic (baseline) and every 6-12 months according to a standard protocol that includes a detailed history, physical examination, and laboratory assessment. At each protocol visit, a rheumatologist records the presence of back pain, as well as the individual criteria used to determine if it is IBP or MBP, including age at onset, duration, presence of morning stiffness and its duration, improvement with exercise or rest, alternating buttock pain, pain in the second half of the night with improvement on getting out of bed, and responsiveness to nonsteroidal antiinflammatory drugs (NSAIDs). Clinical judgment of the type of back pain is left to the discretion of the rheumatologist, who makes the decision after considering the patient’s history and physical examination but not the imaging findings (although practically not blinded to them). Radiographs of peripheral joints and spine are obtained at baseline and every 2 years and are scored according to the modified Steinbrocker method for peripheral joints17 and the modified New York criteria for sacroiliac joints.18 DDD, syndesmophytes, periosteal bone formation, and the presence of calcaneal (Achilles and/or plantar) spurs are also recorded. Radiographs are scored by at least 2 experienced rheumatologists (DDG, VC) by consensus. All information is collected via a web portal and stored in a database. Notably, we have previously shown that the methods of clinical and radiographic evaluation in our cohort are reliable, with no systematic bias with regard to disease severity and follow-up.19
Patient selection. We retrieved data from patients aged < 50 years enrolled in the cohort between January 1978 and June 2023, who had plain radiographs of the spine completed as per protocol. At each radiographic assessment, the presence of DDD was recorded.
DDD. We defined DDD as the presence of intervertebral disc space narrowing, bony spur formation, facet joint arthrosis, and spondylolisthesis on anteroposterior and lateral plain radiographs of the cervical or thoracolumbar spine20 (Figure 1).
Radiographs showing degenerative changes in young patients with psoriatic arthritis. (A) Cervical spine, lateral view in flexion, degenerative changes at the C5/C6 level with facet joint arthrosis. (B) Lumbar spine, lateral view, showing disc space narrowing throughout the lumbar spine. There is anterior bony excrescence at the anterosuperior aspect of the L5 vertebral body.
Data collection. From our program database, we retrieved data at the patients’ first evaluation in the clinic and at 6- to 12-month intervals, including age, sex, ethnicity, employment, age at the diagnosis of psoriasis and PsA, disease duration, smoking status, previous trauma, BMI (calculated as weight in kilograms divided by height in meters squared), comorbidities (hypertension, diabetes mellitus [DM], and hyperlipidemia), and the presence and type of back pain. We summarized musculoskeletal disease activity using the Disease Activity Index for Psoriatic Arthritis (DAPSA).21 We assessed skin disease activity using the Psoriasis Area and Severity Index (PASI). We also retrieved other important disease-related measures including the presence of nail disease, enthesitis, and dactylitis. Laboratory data included erythrocyte sedimentation rate (ESR) and HLA-B27 status. Abnormal ESR was considered in male patients when ≥ 15 mm/h and in female patients when ≥ 20 mm/h. We included the following radiographic features (obtained at 2-year intervals): syndesmophytes, sacroiliitis, periosteal bone formation, calcaneal spurs, erosions (scored per the modified Steinbrocker score), and the radiographically damaged joint count (number of joints with erosions). We also retrieved information on current treatments with NSAIDs, conventional synthetic disease-modifying antirheumatic drugs (DMARDs; methotrexate, sulfasalazine, and leflunomide), and targeted DMARDs (including targeted synthetic DMARDs [apremilast, tofacitinib, and upadacitinib] and biologic DMARDs).
Statistical analysis. We present descriptive statistics for continuous variables (mean [SD] if normally distributed or median [IQR] otherwise) and count data according to frequencies and percentages.
We used multivariate Cox regression models22 to study factors associated with the time to the development of DDD, with the time of clinic entry as the time origin and censoring follow-up when patients reached 50 years of age; patients who had DDD at clinic entry were excluded from the analysis of the onset time. Since the exact point of time at which DDD developed cannot be accurately determined, we used midpoint imputation, with the time defined as halfway between the present (the visit at which it was first recorded) and the previous visit. We applied the last observation carried forward approach to impute data on visits with missing covariate information; the covariate values from the most recently completed assessments were carried forward until a new covariate value was available. We considered the following covariates as time-varying: age, disease duration, being employed, previous trauma, hypertension, DM, hyperlipidemia, BMI, MBP, IBP, PASI, modified Steinbrocker score, periosteal bone formation, calcaneal spurs, axial disease, abnormal ESR, use of NSAIDs, use of conventional synthetic DMARDs, and use of targeted (biologic or synthetic) DMARDs. Sex and HLA-B27 status were used as fixed covariates. For the full model, we included all covariates of interest. The reduced model was obtained by backward elimination.
We used generalized estimating equations (GEE)23 to fit marginal logistic regression models describing the association between axial imaging findings and the type of back pain experienced (IBP vs MBP). For this analysis, we divided the cohort into 4 categories based on findings on plain radiographs: isolated DDD, isolated axial disease (sacroiliitis or syndesmophytes), both DDD and axial disease, and normal spine radiographs. We compared the first 3 categories to the last. The first model tested the association between back pain and the different imaging groups. Two additional models were fitted to test the association between MBP and IBP, each separately, and the imaging categories, after censoring the visits where no back pain was reported. We adjusted all models for age, sex, BMI, NSAID use, and targeted DMARD use.
RESULTS
Prevalence of DDD in young patients with PsA. Of the 814 patients included in this study, 316 (38.8%) were observed to have DDD on plain radiographs of the spine. Sixty-five (20.6%) patients had isolated cervical disease, 117 (37%) had isolated thoracolumbar disease, and 134 (42.4%) had both. One hundred eighty-two (57.6%) patients had DDD at their baseline visit, whereas 134 (42.4%) developed it during a median follow-up of of 6.1 (IQR 2.8-10.7) years.
Patient characteristics. Baseline characteristics of patients with DDD at clinic entry and during follow-up compared to those who did not have nor develop DDD are shown in Table 1. The mean age of our cohort was 36.4 (SD 8.3) years; 476 (58.5%) patients identified as male and 674 (83%) identified as White. The mean BMI was 28.4 (SD 6.2). One hundred ninety-seven patients (24.4%) were observed to have radiographic sacroiliitis and 77 (9.5%) patients were on targeted DMARDs.
Baseline characteristics of patients aged < 50 years with and without DDD.
Factors associated with the development of DDD. Using Cox regression analysis, the following factors were associated with the development of DDD (Table 2): older age (hazard ratio [HR] 1.08, P < 0.01), male sex (HR 1.52, P = 0.03), DM (HR 2.35, P = 0.045), and IBP (HR 2.03, P < 0.01). Notably, being employed (vs unemployed; HR 1.61, P = 0.05), higher BMI (HR 1.02, P = 0.06), calcaneal spurs (HR 1.40, P = 0.09), and use of targeted DMARDs (HR 1.46, P = 0.05) showed a trend toward association with the development of DDD.
Multivariate Cox regression model for factors associated with the development of DDD.
Association of back pain with imaging findings. In the GEE model (Table 3), none of the 3 abnormal imaging categories was significantly associated with having back pain. After censoring the visits where no back pain was reported, MBP was inversely associated with isolated radiographic axial disease (odds ratio [OR] 0.69, P = 0.03) compared to isolated DDD and mixed radiographic features. It also showed a trend for an inverse association with isolated radiographic axial disease compared to normal radiographs (OR 0.69, P = 0.05). IBP was not significantly associated with any of the imaging categories but showed a positive trend for an association with isolated radiographic axial disease (OR 1.42, P = 0.08).
Generalized estimating equations presenting the association between different imaging categories and back pain.
DISCUSSION
In our large longitudinal study, we found that the prevalence of DDD in patients with PsA younger than 50 years was 38.8%. Several factors were associated with the development of DDD, including older age, male sex, DM, and IBP. Being employed (vs unemployed), higher BMI, calcaneal spurs, and use of targeted DMARDs showed a trending association with DDD. When we stratified the data by radiographic findings of the spine, none of the 3 abnormal imaging categories—isolated DDD, isolated axial PsA, and mixed features of both—showed a significant association with having back pain. Further, IBP did not reliably distinguish between axial PsA and DDD.
Although it may be challenging to draw direct comparisons between our study population and patients without PsA, due to factors such as the young age of this cohort, the increased association with obesity, the reliance on a radiographic definition, and the repeat of radiographs at regular intervals, the prevalence of DDD in our cohort appears high. For context, a study using Medicare insurance claim data (including patients aged ≥ 65 years) reported an overall prevalence of diagnosed spinal DDD of 27.3%, which highlights the relatively higher occurrence in our younger cohort.24
The relationship between DDD and both age and BMI is well established.3,25-28 Considering the influence of sex, young men are generally more susceptible to disc degeneration than young women, likely due to increased mechanical stress and a higher risk of physical injury. However, this pattern may shift with advancing age.29 Importantly, DM was associated with DDD in our cohort, even after accounting for BMI. This interesting finding remains compatible with previous reports and may be explained by several harmful changes demonstrated in animal models, including increasing toxic end products of glycation, expression of matrix metalloproteinases, and hyperglycemia-induced inflammation in intervertebral discs.30,31 Although we expected radiographic axial PsA to be associated with the occurrence of DDD, this was not the case. It may be that the sole reliance on radiographs, particularly for the definition of axial disease, has diluted the expected trend; IBP, however, appeared as a risk factor for DDD.
Being employed (vs unemployed) showed a trend toward an association with the development of degenerative changes in our young cohort, potentially explained by the link to mechanical loading.32 The relationship with calcaneal spurs is also not unexpected, given the shared bone-forming pathophysiology between enthesophytes and osteophytes.33 Notably, both DDD and calcaneal spurs are more commonly observed in individuals with obesity, a known comorbidity of PsA.34 Interestingly, the use of targeted DMARDs showed a trending association with DDD development in our cohort. This observation may arise from the tendency to administer more advanced treatment to patients with a higher disease burden; these patients typically experience a more pronounced inflammatory state and may have an increased risk of disc degeneration.
When the association between imaging findings and back pain was examined, none of the 3 abnormal imaging categories was significantly linked with having back pain. This is consistent with data suggesting that both axial PsA and DDD may be clinically silent.10,35,36 After censoring the visits where no back pain was reported, MBP was inversely associated with isolated axial disease, which is not unexpected. IBP, however, was not significantly associated with any of the abnormal imaging categories. There may be several explanations for this finding. First, we have previously shown that rheumatologist-judged IBP or the criteria for IBP developed for radiographic axial spondyloarthritis (used in our study) do not perform well when studying axial involvement in PsA.14 Second, an inflammatory response may also be present in DDD,37,38 thus obscuring the clinical evaluation. Third, obesity is common in psoriatic disease and contributes to back pain.1,28
Given that the presence of IBP is a key factor in making therapeutic decisions for axial spondyloarthritis39 and considering the low reliability of IBP in PsA as well as the frequent occurrence of obesity (and its associated complications) in this disease,1,40 it may be prudent to pursue further diagnostic steps when evaluating complaints of IBP, such as obtaining a magnetic resonance imaging scan.
To our knowledge, our analysis is the first study in the literature to date to describe and characterize the occurrence of DDD in a young adult PsA cohort. The high number of patients, the organized nature of the data, the longitudinal design, and the long-term follow-up period are all important strengths. We do, however, acknowledge some limitations, including the retrospective nature of the study, the single-center design, the lack of non-PsA control patients, and the reliance on radiographs as the sole imaging technique. Additionally, we did not include information on the specific type of employment; certain jobs may be associated with greater mechanical overload. Last, being unable to blind the physician to the imaging findings in the spine and sacroiliac joints may have affected their interpretation of the type of back pain.
In conclusion, DDD is common in young patients with PsA, and its development may be associated with demographic features, comorbidities, and disease-related factors. Although radiographic changes of the spine may be completely asymptomatic, the presence of IBP does not reliably distinguish between axial PsA and DDD. These findings should be taken into consideration, as they contribute to better clinical assessment and therapeutic decision making.
Footnotes
CONTRIBUTIONS
FK: conceptualization, data curation, formal analysis, investigation, methodology, project administration, validation, visualization, writing – original draft, writing – review & editing; SG: conceptualization, data curation, formal analysis, investigation, methodology, resources, software, validation, visualization, writing – original draft, writing – review & editing; DP: data curation, investigation, project administration, resources, software, supervision, validation, visualization, writing – review & editing; RJC: conceptualization, data curation, formal analysis, project administration, investigation, methodology, resources, software, supervision, validation, visualization, writing – review & editing; VC, DDG: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing – original draft, writing – review & editing.
FUNDING
The Gladman-Krembil Psoriatic Arthritis Research Program is funded by the Krembil Foundation and the Schroeder Arthritis Institute. VC is supported by a clinician scientist salary award from the Department of Medicine, University of Toronto.
COMPETING INTERESTS
VC has received research grants from AbbVie, Amgen, and Eli Lilly, and has received honoraria for advisory board member roles from AbbVie, BMS, Eli Lilly, Fresenius Kabi, Janssen, Novartis, and UCB; and his spouse is an employee of AstraZeneca. DDG has received grants and/or consulting fees from AbbVie, Amgen, AstraZeneca, BMS, Eli Lilly, Janssen, Novartis, Pfizer, and UCB. FK, SG, DP, and RJC declare no conflicts of interest relevant to this article.
ETHICS AND PATIENT CONSENT
We obtained ethics review and approval from the University Health Network Research Ethics Board (08-0630). We collected informed consent from all patients at enrollment.
- Accepted for publication January 28, 2025.
- Copyright © 2025 by the Journal of Rheumatology
