Abstract
Objective Diffuse idiopathic skeletal hyperostosis (DISH) is characterized by calcification and ossification, predominantly affecting the axial skeleton. DISH is associated with obesity, diabetes mellitus, and hypertension, all included in the metabolic syndrome (MetS). The aim of this study was to investigate the prevalence of DISH in patients with MetS and the influence of obesity on spinal radiographic changes related to DISH vs degenerative changes.
Methods Patients diagnosed with MetS were categorized into 4 obesity classes (World Health Organization criteria). Based on Resnick criteria, patients were classified as DISH or non-DISH. Radiographs were assessed for degenerative (d)-spondylophytes and DISH-related chunky (Drc)-spondylophytes, and their association with obesity was analyzed.
Results There were 124 patients were included, of whom 42 (33.9%) were identified as having DISH. The mean BMI was 39.8 (SD 6.3), with obesity classes distributed as follows: preobese (n = 7), obesity class I (n = 19), II (n = 38), and III (n = 60). Higher obesity classes (II and III) were significantly associated with an increased number of Drc-spondylophytes in all patients. In DISH, obesity class ≥ II was linked to a greater number of Drc-spondylophytes (14.8 [SD 10.3] vs 9.7 [SD 7.7]), whereas no association was found between obesity and d-spondylophytes. Overall, patients with DISH had fewer d-spondylophytes than those without DISH (2.7 [SD 2.7] vs 4.1 [SD 4.1]; P = 0.04).
Conclusion One-third of patients with MetS were classified as having DISH. In contrast to degenerative spine disease, obese MetS classes II and III were significantly associated with radiographic spinal changes attributed to DISH.
Diffuse idiopathic skeletal hyperostosis (DISH) is a skeletal disorder characterized by the calcification and ossification of ligaments and entheses, particularly along the thoracic part of the spine. DISH was first recognized by Forestier et al in 1950, who distinguished it from radiographic axial spondyloarthritis (axSpA; formerly termed ankylosing spondylitis).1 The disorder was later defined more precisely by Resnick and Niwayama, who characterized it radiographically by flowing ossifications alongside the anterolateral aspect of at least 4 contiguous vertebrae.2 This radiographic feature, often described as “chunky osteophytes,” remains the cornerstone of DISH diagnosis today.
DISH is not a rare condition, particularly among the elderly, with prevalence estimates ranging from 3.8% to 30% in individuals over 50 years old, and is more commonly diagnosed in men.3-14 The association of the disorder with aging is well documented, with prevalence increasing significantly after the age of 60.4,14-17
Despite its frequency, the pathophysiology of DISH remains incompletely understood. Until today, the condition is diagnosed exclusively based on radiographic criteria. Several metabolic factors, such as obesity, impaired glucose tolerance, diabetes mellitus, dyslipidemia, and hypertension, which contribute to the diagnosis of metabolic syndrome (MetS), have been associated with DISH.18-20
Notably, the association between DISH and obesity seems to be strong, with many studies consistently showing higher BMI in patients with DISH compared to patients without DISH.3-5,7,9,12,18,21-27 DISH has also been linked to cardiovascular disorders, reinforcing the hypothesis that it is not merely a localized skeletal disorder.18,27-30
The aim of this study was to investigate the prevalence of DISH in patients with MetS and to investigate the influence of obesity on spinal radiographic changes related to DISH vs degenerative changes.
METHODS
The analysis included information from in-and outpatients who visited our hospital group (Department of Internal Medicine and Department of Rheumatology, Rheumazentrum Ruhrgebiet) between January 2019 and September 2021. The primary selection process was based on International Classification of Diseases (ICD)-coded diagnoses. Both type 2 diabetes mellitus and hypertension had to be documented. In addition, patients were included when they presented with a BMI (calculated as weight in kilograms divided by height in meters squared) defined as ≥ 25. Patients were excluded if they had a concomitant diagnosis of axSpA, confirmed by the combination of clinical and imaging findings; if fibromyalgia was documented; or if no radiographs of the spine or chest were available. Patients were admitted for various reasons, including metabolic, cardiac, pulmonary, or renal conditions.
Clinical assessments. Demographics and clinical data including age, sex, weight, height; laboratory data (C-reactive protein, creatinine, HBA1c, uric acid, urea); as well as back pain (numerical rating scale [NRS] 0-10), general body pain (NRS 0-10), and data on medication for hypertension and diabetes mellitus were collected.
For the classification of obesity, patients were categorized, according to the World Health Organization (WHO), in 4 obesity classes31: preobese: BMI 25.0-29.9 (increased); obese class I: BMI 30.0-34.9 (moderate); obese class II: 35.0-39.9 (severe); and obese class Ill: 40.0 (very severe). For statistical analysis, we subdivided obese class III into class IIIa (BMI 40.0-44.9) and class IIIb (BMI ≥ 45),
Imaging assessments. Radiographs (lateral and anterior-posterior) of the cervical, thoracic, and lumbar spine, as well as chest radiographs, were obtained as part of routine clinical care. For inclusion, at least 1 radiograph had to be available. Images were analyzed retrospectively on certified Digital Imaging and Communications in Medicine (DICOM)-compliant workstations. Characteristic radiographic findings of DISH in the spine were defined as complete and incomplete DISH-related chunky (Drc)-spondylophytes in the axial direction, starting at the anterior edges of the vertebrae and following the flow of the anterior and/or lateral longitudinal ligament, whereas horizontally oriented osteophytes were classified as degenerative (d)-spondylophytes32-35 (Figure 1). Drc-spondylophytes were scored with 1 point for incomplete and 2 points for complete bridging. A modified version of the Resnick criteria was applied: DISH was diagnosed based on the presence of Drc-spondylophytes spanning at least 4 contiguous vertebrae, without exclusion based on intervertebral disc degeneration; this is in line with evidence indicating that degenerative changes can coexist with DISH. All other patients were classified in the non-DISH group.2,36-38
Differentiation of Drc-spondylophytes and d-spondylophytes on lateral spine radiographs. (A) Drc-spondylophyte oriented at 32° to the anterior vertebral edge. White arrows indicate additional complete bridging at adjacent levels. (B) d-spondylophyte oriented at 84° to the anterior edge (white arrow). Black arrows mark additional d-spondylophytes at other levels. d-spondylophyte: degenerative spondylophyte; DISH: diffuse idiopathic skeletal hyperostosis; Drc-spondylophyte: DISH-related chunky spondylophyte.
Method of differentiation between Drc-spondylophytes and d-spondylophytes. According to previous studies, we used the following method to differentiate between Drc-spondylophytes and d-spondylophytes: a reference line was drawn along the anterior vertebral edges of the same vertebral body. From this line, a second line was drawn at a 45° angle to the anterior edge.39
Bony changes forming an acute angle of < 45° with the anterior vertebral side was classified as Drc-spondylophytes when its morphology was assessed by the reader as typical for DISH (Figure 1A).
Bony changes with an obtuse angle ≥ 45° were classified as d-spondylophytes when its morphology was assessed by the reader as typical for degenerative spine disease (Figure 1B).
Two readers, trained for reading musculoskeletal radiographs of the spine, evaluated the radiographs blinded for all information (MG and DK). Pathologies were considered positive if identified by both readers. All cases with disagreement were solved in an adjudication session where a third reader was also present (XB).
Statistical analysis. Mean values, SDs, and frequencies are provided descriptively. Comparisons between groups were performed using the t test for independent samples for the mean values and using the chi-square test for frequencies. The correlations between clinical and imaging outcomes were calculated with Spearman correlation coefficient, whereas Cohen d was applied to measure effect sizes. A multivariable linear regression analysis was performed to assess the influence of age, sex, and BMI (independent variables) on the number of Drc-spondylophytes and d-spondylophytes (dependent variables). A P value < 0.05 was considered statistically significant. Statistical analysis was performed with BlueSky Statistics, V.10.3.1.
RESULTS
A total of 207 patient records were screened retrospectively and consecutively. From these, 124 patients met the inclusion criteria and had at least 1 radiograph of the spine or the chest available. These patients were included in the final study sample. Forty-two patients (33.9%) were identified as having DISH based on the Resnick criteria, whereas the remaining 82 patients (66.1%) did not meet these criteria. The mean (SD) age between both groups (68.8 [10.8] vs 64.7 [12.3]; P = 0.07) was comparable. Demographics showed no significant differences for sex, laboratory data, or back pain (Table 1).
Demographic and clinical data in patients with DISH and non-DISH.
In line with the inclusion criteria, no patient was underweight or of normal weight. Mean BMI was 39.8 (SD 6.3), patients were stratified into obesity classes: preobese (n = 7; 5.6%), I (n = 19; 15.3%), II (n = 38; 30.6%), and III (n = 60; 48.4%). Mean (SD) BMI of DISH and non-DISH was also comparable (40.7 [7.1] vs 39.4 [5.9]). Only a few patients in both groups were classified as preobese (DISH: 7.1% vs non-DISH: 4.9%). Although there were some differences in obesity class I (DISH: 7.1% vs non-DISH: 19.5%) and II (DISH: 35.7% vs non-DISH: 28%), class III was comparable (DISH: 50% vs non-DISH: 47.6%) between both groups (Table 1).
In the total cohort, radiographs of the thorax (n = 83), lumbar spine (n = 90), thoracic spine (n = 21) and cervical spine (n = 42) were evaluated. Overall, a total of 3441 vertebral body corners were assessed on conventional radiographs. Forty-two patients were classified as having DISH, whereas 19 patients (23.2%) in the non-DISH group exhibited Drc-spondylophytes but did not fulfill the modified Resnick criteria. In total, we identified 691 Drc-spondylophytes in the DISH group and 449 d-spondylophytes in the non-DISH group. Additionally, 81 Drc-spondylophytes were observed in patients in the non-DISH group. Patients with DISH showed a significantly higher mean (SD) number of vertebral segments affected by Drc-spondylophytes (14.0 [10.1]) compared to the non-DISH group (1.2 [3.0]; P < 0.001). Patients in the DISH group had a mean (SD) of 12.8 (9.6) complete and 2.2 (0.3) incomplete Drc-spondylophytes. In contrast, the non-DISH group had significantly fewer complete Drc-spondylophytes (0.4 [1.4]; P < 0.001) and numerically fewer incomplete Drc-spondylophytes (0.8 [2.7]; P = 0.3), as seen in Table 2. In contrast, compared to the non-DISH group (4.1 [4.1]), patients with DISH presented significantly fewer (2.7 [2.7]) d-spondylophytes (Table 2). Notably, 10 patients with DISH presented incomplete Drc-spondylophytes, and 19 patients of that group had complete Drc-spondylophytes but did not fulfill the modified Resnick criteria.
Comparison of radiographic changes in patients with and without DISH.
Multivariate regression analysis revealed that age, sex, and BMI were significantly associated with the number of Drc-spondylophytes. Age was positively associated with the number of Drc-spondylophytes, with an increase of 0.17 Drc-spondylophytes per year of age (95% CI 0.05 to 0.30; P = 0.007). Female sex was associated with 3.04 fewer Drc-spondylophytes compared to male sex (95% CI −6.05 to −0.03; P = 0.048). BMI showed a significant positive association, with an increase of 0.41 Drc-spondylophytes per unit increase in BMI (95% CI 0.18 to 0.64; Table 3).
Multivariate regression results: influence of age, sex, and BMI on the number of DISH-related chunky spondylophytes.
To further explore the effect of obesity, a subgroup analysis based on WHO obesity classifications was conducted. Patients with MetS classified to obesity classes II and III (BMI ≥ 35.0) had a significantly higher mean (SD) number of Drc-spondylophytes (2.8 [5.5] vs 6.3 [9.3]; P = 0.02) and consequently higher numbers of complete (2.5 [5.3] vs 5.2 [8.6]; P = 0.03) and incomplete (0.3 [0.9] vs 1.1 [2.8]; P = 0.006) Drc-spondylophytes compared to those with lower obesity classes. However, there was no significant difference in the mean (SD) number of d-spondylophytes between these groups (3.6 [4.1] vs 3.6 [3.7]; Table 4). The second analysis focused exclusively on patients who met the Resnick criteria for DISH. Although the number of Drc-spondylophytes was higher in the obesity class II and III (14.8 [10.3] vs 9.7 [7.7]), this difference did not reach statistical significance (P = 0.25). In the subgroups with complete and incomplete Drc-spondylophytes, the number was higher in patients with higher obesity classes, reaching statistical significance only for incomplete Drc-spondylophytes (P = 0.03), as seen in Table 4. Consistent with the findings in the overall population, the number of d-spondylophytes remained comparable between obesity classes in patients with DISH.
Comparison of radiographic damage in patients with preobese and obesity class I with patients with obesity class II and III.
DISCUSSION
In this study, based on data retrieved from daily practice, we observed that approximately one-third of patients with MetS also met the radiographic criteria for DISH, a finding that highlights the intersection of these 2 conditions. Considering that the global prevalence of MetS estimated by the International Diabetes Federation is approximately 25%, varying based on demographic factors such as age, sex, and ethnicity,40,41 our findings underscore the high prevalence of DISH within MetS populations.
A key finding of our study is the independent association of higher BMI with an increased number of Drc-spondylophytes in MetS patients with and without DISH. This relationship persisted even after adjusting for other potential confounders such as age and sex. Notably, patients with MetS in the DISH and non-DISH groups having obesity classes II and III exhibited significantly higher numbers of Drc-spondylophytes compared to those with lower BMI categories. Therefore, our results support the hypothesis that obesity may not only be a comorbid condition but also be a driving factor in the formation of Drc-spondylophytes.3-5,7,9,12,18,21-25,27
In contrast, the number of d-spondylophytes did not significantly differ between high and low BMI groups. This lack of association for d-spondylophytes indicates a specificity of the influence of obesity on DISH-related ossification processes, further distinguishing DISH from other forms of spinal degeneration. The coexistence of Drc-spondylophytes and d-spondylophytes in some patients, as observed in our cohort, aligns with recent literature suggesting that degenerative and DISH-related changes can occur concurrently.5,37,38,42 Although this may initially raise concerns about potential misclassification, the differentiation between Drc-spondylophytes and d-spondylophytes in our analysis was based on clear radiographic criteria. Notably, we did not observe significant differences in demographic data between patients with and without DISH in our cohort with all patients presenting obesity, DM, and hypertension. The strong association between obesity and DISH-related radiographic changes underscores the importance of integrating weight management strategies into the clinical care of patients with MetS to potentially mitigate the progression of Drc-spondylophytes. In summary, this study shows that a relevant proportion of patients with MetS present the typical radiographic features of DISH, with obesity (particularly class ≥ II) associated with the formation of Drc-spondylophytes. This supports the growing body of evidence linking obesity with DISH. Future research should focus on the underlying mechanisms connecting obesity to DISH and explore whether targeted interventions, such as weight management strategies, might help mitigate disease progression.
This study has limitations that should be acknowledged. First, the retrospective design inherently introduces a risk of selection bias, as the data were extracted from medical records, and radiographs were obtained as part of routine clinical care rather than a standardized protocol. The study cohort was limited to patients with MetS, which could have led to an overrepresentation of obesity and related comorbidities. This focus might limit the generalizability of the findings to populations without MetS or those with lower BMIs.
Due to the heterogeneity of inpatient admission diagnoses and the potential effect of acute disease processes on laboratory findings, we excluded laboratory variables from the present analysis. We considered these values inappropriate for investigating long-term structural changes such as new bone formation in the spine.
In conclusion, the absence of longitudinal data precludes conclusions about the temporal relationship between obesity, MetS, and the development of radiographic changes in patients with DISH. It remains unclear whether obesity contributes to the progression of Drc-spondylophytes over time or merely reflects an association.
Footnotes
CONTRIBUTIONS
DK: conceptualization, data curation, formal analysis, investigation, methodology, validation, writing - original draft, writing - review & editing. MG: conceptualization, data curation, formal analysis, investigation, methodology, validation, writing - original draft, writing - review & editing. PS: formal analysis, methodology, validation, writing - original draft, writing - review & editing. UK: formal analysis, methodology, validation, writing - original draft, writing - review & editing. ST: conceptualization, data curation, formal analysis, investigation, methodology, validation, writing - original draft, writing - review & editing. IA: formal analysis, methodology, validation, writing - original draft, writing - review & editing. XB: conceptualization, data curation, formal analysis, investigation, methodology, project administration, supervision, validation, writing - original draft, writing - review & editing.
FUNDING
The study received no financial or industrial support for this work.
COMPETING INTERESTS
The authors declare no conflicts of interest relevant to this article.
ETHICS AND PATIENT CONSENT
The study was approved by the Ethical Committee of the Ruhr-Universität Bochum (reference number 22 7735). Informed written consent was obtained from all patients.
- Accepted for publication June 2, 2025.
- Copyright © 2025 by the Journal of Rheumatology
