You are here

Article Text

Article menu

Download PDFPDF

Spine
Original research
Spatiotemporal changes in imaging features associated with diffuse idiopathic skeletal hyperostosis (DISH)
  1. Dale E. Fournier1,2,
  2. Michele C. Battie2,3 and
  3. Cheryle A. Séguin2,4
  1. 1Health and Rehabilitation Sciences (Physical Therapy), Faculty of Health Sciences, The University of Western Ontario, London, ON, Canada
  2. 2Bone and Joint Institute, The University of Western Ontario, London, ON, Canada
  3. 3School of Physical Therapy, Faculty of Health Sciences, University of Western Ontario, London, ON, Canada
  4. 4Department of Physiology and Pharmacology, Schulich School of Medicine & Dentistry, The University of Western Ontario, London, ON, Canada
  1. Correspondence to Dr Cheryle A. Séguin; cheryle.seguin{at}schulich.uwo.ca

Abstract

Objectives The purpose of our study was to characterise spatiotemporal features of disease progression in people with diffuse idiopathic skeletal hyperostosis (DISH), early-phase DISH, and those not meeting either criterion who had repeated CT scans of the thoracic spine.

Methods A retrospective study was designed in collaboration with the Rochester Epidemiology Project to evaluate completeness of ectopic bridging across the thoracic spine and corresponding disease status over an average of 2.7 years (range from 0.2 to 15.0 years) in a cohort of 83 female and 74 male individuals.

Results Over 15% of individuals displayed changes in imaging features over time that resulted in a revised diagnosis along the continuum of DISH. Early-phase DISH was marked by new involvement of previously unaffected motion segments, estimated to occur over 2.1 years. Advanced presentations of DISH were marked by increased prevalence of complete bridging (average two of three available motion segments), estimated to occur over 2.6–2.9 years. Localised nodules of ectopic mineralisation external to and within the intervertebral disc were regularly observed in early-phase DISH.

Conclusions This is the first characterisation of spatiotemporal features across all phases of DISH, indicating that progression of DISH is characterised by distinct features at different phases along the disease continuum. Localised nodules of mineralisation in the spinal ligaments and within the intervertebral discs coincident with early phases of the disease may be a key factor in the pathogenesis of DISH.

  • Arthritis
  • Classification
  • Osteoarthritis

Data availability statement

Data are available upon reasonable request. The data underlying this article cannot be shared publicly for the privacy of individuals who participated in the study. The data will be shared on reasonable request to the corresponding author.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/rmdopen-2024-004074

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Diffuse idiopathic skeletal hyperostosis (DISH) is a prevalent condition diagnosed through imaging, yet the continuum of changes in features of spine mineralisation is poorly understood.

    WHAT THIS STUDY ADDS

    • Early-phase DISH is driven by involvement of previously unaffected regions of the spine with ectopic bridging.

    • Advanced DISH is characterised by progressive change in the extent of bridging in affected regions.

    • Localised nodules of ectopic mineralisation are associated with the early stages of DISH and were observed in nearly all individuals with early-phase DISH.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Based on these findings, revision to the classification of DISH based on its continuum may guide specific recommendations and/or clinical pathways.

    Introduction

    Diffuse idiopathic skeletal hyperostosis (DISH) is a non-inflammatory enthesopathy of the spine1 that affects an estimated 14% of adult Americans and over one-third of men over 50 years.2 DISH is more prevalent in male than female individuals with advanced age,2 and the underlying aetiology is unknown.3 The clinical presentation of DISH can include physical dysfunction,4–6 obstruction of the airway7 or compromised neural tissues.8–12 The rigid spine associated with DISH also increases the risk of vertebral fractures from low-energy forces.13–16

    The diagnosis of DISH is based on Resnick and Niwayama’s radiographic criteria of ectopic mineral bridges that connect four contiguous vertebrae of the spine with preserved intervertebral disc (IVD) height and absence of bony ankylosis of the apophyseal or sacroiliac joints in the involved areas.17 The severity of DISH can be further categorised based on the number of contiguous bridges detected.2 These mineral formations present heterogeneously as osteophyte-like outgrowths and/or flowing bands of mineralised tissue.18–21

    Early-phase DISH is categorised based on the assessment of completeness of ectopic bridging and the spatial relationship across three contiguous segments.22 The prevalence of early-phase DISH is reported to be 13% of adults with an average age of 73 years and slightly higher prevalence in male compared with female individuals.2

    This study sought to describe the spatiotemporal progression of ectopic mineralisation in the thoracic spine by evaluating repeated computed tomography (CT) scans in a sample of individuals meeting the diagnostic criteria for DISH or early-phase DISH at the time of their most recent scan, and those without either diagnosis. This is the first study to provide a detailed characterisation of the changes in imaging features of ectopic mineralisation in the thoracic spine across the continuum of DISH over time. We specifically focused on the continuum of DISH pathogenesis for individuals with early-phase DISH, moderate and severe DISH. The purpose of the study was to better understand the progression of ectopic mineralisation in the spine to elucidate unique clinical features of DISH and better inform on disease continuum.

    Methods

    Data source

    Data were retrieved from the Rochester Epidemiology Project that collates health information from various clinical encounters for individuals of a catchment population in Minnesota, USA.23 24

    Study design and sample

    We conducted a retrospective descriptive study of repeated CT scans of the thoracic spine that were acquired between 1995 and 2019 to evaluate the progression of ectopic mineralisation in a random sample of adults over the age of 19 years with spine imaging.2 Scans of the thoracic spine were identified using procedural codes (72128-30), and a minimum threshold of 2 months between scans was set. There were no restrictions regarding sex, race, indication for imaging or medical history. The last scan (most recent) was used to categorise individuals into one of four distinct groups: 1) those meeting Resnick and Niwayama’s criteria for DISH (moderate, 3–6 contiguous bridges)17; 2) those meeting Resnick and Niwayama’s criteria for DISH (severe, 7+ contiguous bridges)2; 3) those meeting Kuperus et al’s criteria for early-phase DISH;22 and 4) those who did not meet either criterion (figure 1). In addition, the data source was searched using International Classification of Disease (ICD) codes for DISH (V.9: 721.6; V.10: M48.1*), which identified an additional 13 individuals with repeated CT images for inclusion (figure 1). The last image from individuals with an ICD code for DISH was also evaluated to confirm DISH.

    Figure 1
    Figure 1

    Flow chart of study sample from cross-sectional population. Groups were populated from previously established diagnosis of DISH, early-phase DISH or those who did not meet either criterion. Individuals without repeated images or images less than 2 months apart were excluded. A search for DISH based on ICD-9 and ICD-10 codes revealed a group of 13 individuals with multiple images that were included. The cyan boxes represent the number of individuals with multiple CT images that were viewed. The yellow boxes represent the final number of individuals included in the study. Of which, 13 motion segments from three individuals could not be evaluated because of limitations in field of view. Localised deformity was a reason to exclude 27 motion segments from 15 individuals. Since unique motion segments excluded in one image were also excluded in repeated images from the same individual, a total of 64 motion segments from 18 individuals were removed (1.9% of all motion segments). DISH, diffuse idiopathic skeletal hyperostosis; ICD, International Classification of Disease.

    Image acquisition consisted of thoracic and thoracolumbar spine protocols with reported pixel resolutions ranging from 0.6 to 1.1 mm. QREADS Clinical Image Viewer (V.5.14.0, Mayo Clinic) was used to evaluate the reconstructed sagittal and frontal views (consistent window and level: 2500, 350). All first and last scans were randomised for independent evaluation to reduce order bias, although age and sex were known.

    It was established a priori that scans with significant deformity that disrupted anatomy would be excluded (eg, IVD degeneration with fusion, severe kyphosis or scoliosis). Evidence of ankylosing spondylitis—differentiated by a thin band of ectopic mineralisation and presence of zygapophyseal joint erosion or fusion—was also a reason for exclusion. When a unique motion segment was excluded in one image, it was also excluded in any repeated images from the same individual.

    Image evaluation

    This study builds upon our previous work evaluating the continuum of imaging features observed in DISH.2 A validated scoring system was used to independently evaluate each scan for the completeness of ectopic bridging for every motion segment: score of 0 indicates no bridging; score of 1 indicates anterior formations greater than 1 mm apart; score of 2 indicates an almost complete bridge (less than 1 mm apart); and score of 3 indicates a complete bridge.22 Patterns of ectopic bridging were assessed to determine disease categorisation. Inter-rater reliability indicated substantial to excellent agreement between raters in the diagnosis and bridge scoring.2 The presence of focal mineralised nodules was also recorded for each motion segment, annotated as external or within the IVD.

    Statistical methods

    SPSS Statistics (V.29.0, IBM Corp) was used for analysis. Primary analysis was the comparison between the first and last CT scans. Ordinal bridge score data required non-parametric tests. Paired data between the last and first scan were evaluated using Wilcoxon signed-ranked test for ordinal variables and McNemar test for binary variables. Χ2 and Fisher’s exact tests were used to evaluate differences between groups and sexes, and post-hoc testing was performed using adjusted standardised residuals and Bonferroni’s multiple comparisons. Mann-Whitney U test and Kruskal-Wallis pairwise test with Dunn’s multiple comparisons were used to analyse differences between sexes and groups with quantitative data. Predictive modelling of changes over time was performed using logistic regression and fixed linear regression analysis.

    Results

    Sample characteristics

    From 1536 individuals, 163 had repeated CT scans of the thoracic spine and 157 were included in this study (figure 1). The most recent image was used to categorise individuals within the groups by disease status, resulting in 48 with DISH, 30 with early-phase DISH and 79 without either condition. Overall, 91.7% of the individuals were reported as ‘white’. The predominate indication for CT of the thoracic spine was physical trauma (76%: for example, bodily injury, motor vehicle collision or falls). Other indications were neurological reasons (11%), query of fracture (6%), surgical relevance (5%), or cancer screening and/or health follow-up (2%). There was no difference in the time interval between scans based on sex or disease categorisation. The average time between scans was 2.7 years and ranged from 2 months to 15 years. No age or sex differences were noted between disease groups at the time of the first image. At the last image, individuals with DISH were significantly older than those in the without group (78±14 years vs 68±21 years, p=0.01).

    Evaluation of repeated scans revealed changes in disease categorisation over time

    The continuum of DISH can be categorised based on criteria defining early-phase DISH,22 DISH17 and the severity of DISH.2 Progression along the continuum was observed in 15% of individuals without sex-related differences (figure 2A). Individuals who changed disease categories were older than those who remained unchanged (80±14 years compared with 70±19 years, p=0.02). Increased disease severity was seen in 16% of individuals where severity could progress (i.e., without group, early-phase DISH or moderate DISH in the first scan). Decreased disease severity was observed in 4% of individuals and was associated with cases of surgery or injury between scans that may have disrupted contiguous ectopic bridging. In the time frame assessed, no individuals progressed from the without group to DISH (moderate or severe) or vice versa.

    Figure 2
    Figure 2

    Changes in disease stage along the continuum of DISH over time with repeated scans. (A) Schematic representation of categorisation of individuals based on DISH disease stage highlighting changes between first and last images. (B–D) Three-dimensional renderings highlighting the changes in ectopic bridging over time from three unique individuals using three-dimensional Slicer (V.5.2.2). Representative images of the thoracic spine showing regions associated with the progression from (B) without to early-phase DISH, (C) early-phase DISH to DISH and (D) moderate to severe DISH. Inlays present the bridging score at specific motion segments associated with the first and last images for each individual. Scale bars represent 5 cm. DISH, diffuse idiopathic skeletal hyperostosis.

    The criteria for early-phase DISH are based on evaluation of ectopic mineral bridging at individual motion segments (scored as 0–3) and the pattern of bridge scores across three contiguous motion segments.22 All possible patterns for early-phase DISH were observed at least once in individuals who progressed from the without group to early-phase DISH (i.e., new early-phase DISH), with the most common being contiguous segment scores of 3-2-3 and 3-2-2 (figure 2B). When individuals progressed from early-phase DISH to either moderate or severe DISH (i.e., new DISH), the most common patterns of early-phase DISH in the first scan were 3-2-2, 3-3-2 and 2-2-2 (figure 2C). Individuals who progressed from moderate to severe DISH displayed an average increase of three contiguous bridges (first scan=5, range 3–6; last scan=8, range 7–10) (figure 2D).

    Characterisation of changes over time in individuals who did not meet the criterion for DISH

    In the without group, an average of three motion segments per individual (range 0–11) displayed evidence of ectopic spine bridging in the first scan. In the last scan, almost half of individuals (43.6%) displayed at least one region of three contiguous motion segments with some evidence of ectopic mineralisation. More complete bridges were detected in the last scan, resulting in 11.5% of individuals and 1.3% of possible motion segments. New involvement of at least one motion segment with some form of ectopic spine mineralisation was seen in 56.4% of individuals. Regression analysis predicted the occurrence of ectopic mineral formation at a previously unaffected motion segment in the without group over 5.0 years. In contrast, progressive changes in the bridge score of previously affected regions (noted in 20.8% of individuals) were predicted to occur over 2.4 years. Motion segments from individuals in the without group were used as the reference for logistic regression analysis (table 1). Of note, sex was not a significant predictor of either new involvement or progressive change in bridge score.

    View this table:
    Table 1

    Logistic regression analysis for variables associated with ectopic mineralisation at the level of motion segments

    Characterisation of changes over time in individuals with early-phase DISH

    In the group of 30 individuals meeting the criteria for early-phase DISH in the last scan, the average number of motion segments per individual with ectopic mineralisation increased between scans from 8 to 9 (p=0.07). There was a significant increase in the severity of bridging scores between scans (figure 3A). In the first scan, an average of 1±1 complete bridge was detected in individuals associated with early-phase DISH, which increased to an average of 2±2 complete bridges in the last scan. Overall, the likelihood of new involvement and progressive bridge score change was 1.86 (p=0.06) and 2.41 times (p=0.001) greater in motion segments from individuals with early-phase DISH compared with the without group (table 1).

    Figure 3
    Figure 3

    Quantitative analysis of changes associated with early-phase DISH (eDISH). (A) The distribution of ectopic mineralisation of the thoracic spine in individuals who met the criteria for eDISH in the last image compared with bridging score from the same individual in the first image. Legend indicates bridge score. (B) Graph displays the extent of new involvement relative to time. Fixed linear regression analysis enabled predicative estimates per year. (C) Distribution of bridge score in newly involved regions of individuals with new DISH, or unchanged and new eDISH. (D) Graph displays the number of motion segments with a change in bridge score relative to time. Fixed linear regression analysis enabled predicative estimates per year. (E) Distribution of bridge scores in involved regions of individuals with new DISH, or unchanged or new eDISH. DISH, diffuse idiopathic skeletal hyperostosis.

    Early-phase DISH was analysed in three subgroups: new early-phase DISH (early phase detected on the last scan only), persistent early-phase DISH (detected on both scans) and new DISH (early-phase DISH on the first scan, DISH on the last scan). We first identified motion segments with new ectopic bridge scores (0 in first scan and 1, 2 or 3 in the last). Individuals who progressed to new early-phase DISH showed the highest average number of new ectopic bridges (3 of 5 possible motion segments) compared with individuals with persistent early-phase DISH (1 of 2 possible motion segments) or those with new DISH (0.5 of 2 possible motion segments). Linear regression analysis predicted the time for ectopic mineralisation of a previously unaffected motion segment in new early-phase DISH, persistent early-phase DISH and new DISH as 2.1, 5.8 and 3.1 years, respectively (figure 3B). Of note, individuals with new early-phase DISH and persistent early-phase DISH showed the greatest heterogeneity in bridge scores of the newly involved motion segments compared with individuals with new DISH (figure 3C).

    The progression in severity of bridging of motion segments over time was next assessed in early-phase DISH. Almost half (42.5%) of motion segments with ectopic bridge scores of 1 or 2 in the first image changed between scans. Increased bridge score was more commonly detected in the new early-phase DISH (average: two of five motion segments) and new DISH (average: three of seven motion segments) groups compared with persistent early-phase DISH (average: two of seven motion segments). Linear regression analysis predicted the time associated with increased bridge severity at one motion segment (i.e., one-unit change in the bridging score) in the new early-phase DISH, persistent early-phase DISH and new DISH groups as 3.4, 2.7 and 0.83 years, respectively (figure 3D). A one-unit increase in bridge score was detected in 93.9% of motion segments that changed (figure 3E). Individuals with new DISH had more motion segments that developed complete bridges compared with individuals with new or persistent early-phase DISH (64.7% compared with 25.0% and 44.8%, respectively).

    Characterisation of changes over time in individuals with progression of DISH severity

    Four individuals (two female and two male) displayed complete bridging of all motion segments in both images and as such were removed from analysis. In the group of individuals with DISH, 49% were categorised as severe (seven or more contiguous bridges) at the first scan, which increased to 67% in the last scan. On average, ectopic bridge scores of 1, 2 or 3 were observed in 9 of 11 motion segments from individuals with moderate DISH and 10 of 11 with severe DISH. In the last image, fewer near-complete bridges (p=0.05) and more complete bridges (p=0.03) were observed compared with the first image (figure 4A). When stratified by sex, female individuals showed more complete bridges in the last image compared with the first (p=0.06), whereas male individuals did not (p=0.19). Similar to changes in early-phase DISH, it was most common for bridge scores to increase by one unit (80.7%) than two (15.7%) or three (5.9%). In motion segments from individuals with moderate or severe DISH in the first image, the likelihood of progressive bridge score change was 3.56 (p<0.001) and 4.68 times (p=0.001) greater compared with motion segments from individuals in the without group (table 1).

    Figure 4
    Figure 4

    Quantitative analysis of changes associated with moderate to severe DISH. (A) The distribution of ectopic bridge scores for motion segments of the thoracic spine in individuals who met the criteria for DISH in both scans, comparing bridge scores from the same individual in the first image. Legend indicates bridge score. (B) Graph displays the number of motion segments in each individual with mineralisation present in only the last image relative to time. (C) Graph displays the number of motion segments in each individual with increased bridge scores at the last image relative to time. Graphs are coded based on the diagnostic criteria for each individual (unchanged moderate DISH, progression from moderate to severe DISH, unchanged severe DISH). Fixed linear regression analysis used to predict the rate of new bridge formation or bridge progression. (D) Visualisation of the bridge scores at the motion segment adjacent to regions of contiguous bridging meeting the diagnostic criteria for DISH at the first and last images. Percentages are the number of immediately adjacent motion segments in the last scan that maintained the same bridge score or increased in bridge score compared with the first scan. DISH, diffuse idiopathic skeletal hyperostosis.

    Individuals with DISH were further categorised into three subgroups: persistent moderate DISH, persistent severe DISH and progressive DISH (moderate DISH in the first scan and severe DISH in the last scan). New ectopic bridge scores in previously unaffected regions were seen in 48.0% of individuals with DISH, involving almost one-third (31.6%) of possible motion segments (figure 4B). The average number of motion segments with new ectopic bridges was slightly greater among individuals who progressed from moderate to severe DISH (average: 0.8 of 2 motions segments) compared with individuals who did not change disease categories (persistent moderate DISH, average: 0.6 of 3 and persistent severe DISH, average: 0.6 of 1 motion segments). Linear regression analysis predicted that mineralisation of a new motion segment in individuals with progressive DISH and persistent moderate or severe DISH occurs over 4.5, 3.9 and 6.1 years, respectively.

    The progression in severity of ectopic bridging of motion segments over time was observed in all individuals with progressive DISH, and 84.6% of individuals with persistent moderate DISH and 57.9% of individuals with persistent severe DISH (figure 4C). On average, two of three motion segments showed increased bridge scores in individuals who moved from moderate to severe DISH. Increased bridge scores were observed in 25% of possible motion segments in those with persistent presentations of DISH. Linear regression analysis predicted the time for an increased bridge score at one motion segment (ie, one-unit change in the bridging score) in progressive DISH and persistent moderate DISH as 2.9 and 2.6 years, whereas persistent severe DISH was 7.8 years.

    To specifically assess the progression of mineralisation from regions with contiguous bridging, we analysed motion segments immediately adjacent to regions of ectopic bridging that meet the diagnostic criteria for DISH (figure 4D). In the first scan of individuals with DISH (either moderate or severe), the motion segment immediately adjacent to the region of contiguous bridging showed no bridge in 10%, early bridging in 18% and near-complete bridging in 72% of individuals. In the last scan, nearly half (41.6%) of the adjacent motion segments displayed an increased bridge score, displaying change to a complete bridging in 41% of individuals. Progression to a complete bridge in the adjacent motion segment was more common in individuals in the group who progressed from moderate to severe DISH (55.6%) compared with those in the persistent moderate and severe DISH groups (38.5% and 31.6%, respectively).

    Characterisation of localised nodules of ectopic mineralisation within motion segments

    Distinct from ectopic bridging of the vertebral body, mineralised nodules were observed in proximity to the IVD, which have been previously described as ‘nodular soft-tissue calcification’ (figure 5A).22 Localised mineralised nodules were detected in 63.7% of individuals in the first scan and 61.8% of individuals in the last image, occurring in 21.1% of motion segments from these individuals in the first scan and 22.0% of motion segments in the last image. No sex differences were found. We assessed the prevalence of these formations based on location: i) connective tissues external to the IVD or ii) within the annulus fibrosus of the IVD. Stratification based on localisation revealed that mineralised nodules external to the IVD decreased in prevalence from 40.1% to 33.1% of individuals between scans (p=0.11). In contrast, a significant increase in the prevalence of annulus fibrosus nodules was observed from 39.5% to 49.0% of individuals between scans (p=0.03).

    Figure 5
    Figure 5

    Prevalence and association of localised ectopic mineralisation nodules. (A) Examples of clinical images for localisation of nodules to connective tissues likely external to intervertebral disc (IVD; that is, anterior longitudinal ligament) or within the IVD (i.e., annulus fibrosus), indicated by the yellow dotted circles. (B) The prevalence of localised nodules in individuals across the continuum of DISH. (C) Number of motion segments involved with localised nodules of ectopic mineralisation stratified across the disease continuum. Red line indicates mean and error bars are 95% CIs. Each dot represents an individual. (D) Association between the bridge score of the motion segments and localisation of mineralised nodules in connective tissues at the first image. (E) Graph of the change in nodule between images within the same individual, stratified by localisation of nodules in connective tissues. DISH, diffuse idiopathic skeletal hyperostosis.

    We assessed the relationship between localised nodules and the continuum of DISH. In the first scan, localised nodules were most often detected and more abundant in individuals early in the disease continuum (figure 5B,C). Mineralisation external to the IVD was most often associated with a bridge score of 1 (53.6%), in part because the presence of localised nodules adjacent to the IVD meets the criteria for this bridge score (figure 5D).20 Lastly, we evaluated changes to the detection of localised nodules over time. Slightly more nodules were detected in both the first and last scans compared with nodules that were present in only one scan (figure 5E). A notable proportion of nodules external to the IVD were detected only in the first scan. In contrast, slightly more nodules within the IVD were only observed in the last scan.

    Discussion

    In this study, we characterised imaging features of ectopic mineralisation along the continuum of DISH in 157 individuals using repeated CT scans of the thoracic spine (an average of 2.7 years apart). Between the first and last images, 15% of individuals displayed imaging features that led to a change in diagnosis over an average period of 3.7 years. Our analysis revealed that progression to early-phase DISH is characterised by the involvement of previously unaffected motion segments over a predicted 2.1 years. In contrast, progression to DISH is characterised by increased bridge severity at previously affected motion segments, where a one-unit increase in ectopic bridging score is predicted to occur every 0.83 years. Lastly, a one-unit increase in previously affected motion segments in individuals from moderate to severe DISH was predicted to occur over an average of 2.9 years. In addition to bridging of motion segments, localised mineralised nodules were detected within the IVD, particularly early in the disease continuum. These findings underscore that both spatial and temporal changes are critical to progression through distinct phases along the continuum of DISH.

    Our findings of increased prevalence of complete bridges (from 67.7% to 74.1% of motion segments) in individuals with DISH in both images are in keeping with previous studies reporting increases in complete bridging in individuals with DISH over the course of 5 years from 45% to 56%,20 54% to 59%19 and 31% to 38%.25 These results reinforce the progressive nature of DISH following clinical diagnosis.20 25 Our analysis further showcased the extent of ectopic bridging by highlighting the involvement of early forms of bridging in non-contiguous regions of the thoracic spine with DISH. In the last scan, each type of ectopic mineralisation (complete bridges, incomplete bridges and no bridges) was observed in an average 1±1 non-contiguous motion segments, resulting in an average spine bridge score of 5±4 in these regions.

    Few studies have examined imaging features associated with the progression to DISH.19 20 25–27 The current study characterised the imaging features associated with the transition from early-phase DISH to DISH and reported an increase in the frequency of complete bridging from 20.0% to 43.6% of motion segments over an average of 2.4 years. Kuperus et al reported increased complete bridging in the group they classified as ‘pre-DISH’ (DISH confirmed on last scan and not on first scan) from 11.3% to 31.0% of motion segments.20

    By assessing the continuum of DISH, our findings explicitly characterise features associated with the onset of early-phase DISH, the progression from early-phase DISH to DISH, as well as the progression in severity of DISH. We demonstrated progression to the onset of early-phase DISH is primarily characterised by involvement of new motion segments. Meanwhile, progression in disease severity in individuals with DISH is primarily characterised by more severe ectopic bridging of previously involved motion segments. This information is crucial to better understand the clinical features of DISH throughout the disease and suggests that future therapeutic interventions aimed to prevent onset or delay progression may have efficacy at different time points in the natural course of the disease.

    Our study evaluated both female and male individuals throughout the continuum of DISH. In accordance with the well-known sex-related difference in prevalence of DISH,2 56.3% of individuals with DISH in this study were male. Yet, our direct comparison of female and male individuals in each subgroup of DISH suggests that imaging features associated with progressive changes in DISH do not differ based on sex. The greater prevalence among male individuals may be linked to behaviours and/or societal roles that predispose male individuals to the sequelae of DISH and/or biological sex-based differences such as the influence of hormones. Recently, genetic polymorphisms have been associated with DISH.28 29 As such, genetic factors may be a critical feature that differentiates those at risk of advanced pathological mineralisation which may be independent of biological sex. The relative role of these potential underlying factors in the onset, progression and overall prevalence of DISH warrants further investigation.

    A strength of the current study was the detailed evaluation of all motion segments in the thoracic spine using a validated scoring system.22 These findings enable us to put forward three suggestions to advance the field. First, we speculate that there may be unique patterns of bridging at specific regions of the spine that can facilitate early disease detection beyond the number of contiguous bridges. For example, the presence of complete bridging in the lower thoracic spine paired with involvement of early phases of ectopic bridging in upper regions of the spine. Second, we recommend modification of scoring systems to further categorise complete bridges based on morphological features instead of only completeness of bridging (eg, horizontal outgrowths vs flowing bands).19–21 There is value to include the well-characterised feature of ectopic mineralisation opposing the aorta (seen in axial images)20 21 30–34 in refined scoring systems of DISH. An improved understanding of morphological changes of ectopic mineralisation within individual motion segments may enhance detection of DISH across the continuum or identification of disease onset. Lastly, we postulate that additional features, such as the presence of localised nodules of mineralisation—documented by us and others19–21—constitute important pathological features of DISH that should be considered.

    Previous micro-CT-based characterisation of the radiodensity of ectopic mineralisation in DISH reported on unique features of hyperdense material and dystrophic calcification,21 35 likely corresponding to the nodules observed in this study. An increased prevalence of IVD mineralisation with advanced age has been reported in individuals with DISH, along with more complete bridging of the vertebral bodies.27 Interestingly, our investigation of localised nodules showed a significant association with early-phase DISH compared with advanced presentations of DISH, and a reduced prevalence of localised nodules in the last scan. We speculate that localised nodules may be either resorbed with time or become integrated within the ectopic bridge. It is possible that a relationship between bridge scoring and localised nodules may differentiate pathophysiological stages of DISH.

    The current study would be strengthened by an increased number of individuals included, as well as additional scans for each individual to better capture disease progression. Examination of individuals at definitive time intervals may also further distinguish acute and chronic changes associated with the continuum of DISH. Of note, comorbidities within the patient population examined in the current study were not considered, which may impact the progression of pathological mineralisation. For example, associations between metabolic and cardiovascular factors may influence the progression of DISH over time.3 36–41 The sample studied was also limited to predominately white individuals residing in a single geographical region in the USA, which limits the generalisability of our findings to other or more diverse populations. Lastly, the imaging features described represent only one aspect of the clinical profile and would be strengthened by analysis of the correlation with clinical symptoms or functional impairments.

    In conclusion, our detailed analysis of individuals with notable changes in disease features along the continuum of DISH revealed remarkable patterns of pathological changes based on the stage of disease. Early-phase DISH is characterised by the involvement of previously unaffected motion segments and the presence of discrete, localised nodules of ectopic mineralisation within the IVDs. At early stages, therapies should be aimed at preventing the onset of tissue mineralisation. In contrast, increased DISH severity is characterised by the progressive bridging of motion segments previously affected with ectopic mineralisation. At a later stage, therapies may be effective if they target the processes underlying the growth of ectopic bridges. Clearly defining the pathological changes associated with the continuum of DISH will contribute to improved understanding of disease pathogenesis to facilitate targeted clinical management and enhance clinical outcomes.

    Data availability statement

    Data are available upon reasonable request. The data underlying this article cannot be shared publicly for the privacy of individuals who participated in the study. The data will be shared on reasonable request to the corresponding author.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Our study was approved by IRBs (20-000640; 008-OMC-20).

    Acknowledgments

    This study was made possible using the resources of the Rochester Epidemiology Project, which is supported by the National Institute on Aging of the National Institutes of Health under award number R01AG034676. We thank Dr Alanna Chamberlain, Barbara Abbott, and Jon Ziebarth for their assistance.

    References

    1. ↵
      1. Forestier J,
      2. Rotes-Querol J
      . Senile ankylosing hyperostosis of the spine. Ann Rheum Dis 1950;9:321–30. doi:10.1136/ard.9.4.321
      OpenUrlFREE Full Text
    2. ↵
      1. Fournier DE,
      2. Leung AE,
      3. Battié MC, et al
      . Prevalence of diffuse idiopathic skeletal hyperostosis (DISH) and early-phase DISH across the LifeSpan of an American population. Rheumatology (Oxford) 2024;63:1153–61. doi:10.1093/rheumatology/kead362
      OpenUrl
    3. ↵
      1. Mader R,
      2. Verlaan JJ,
      3. Buskila D
      . Diffuse idiopathic skeletal hyperostosis: clinical features and pathogenic mechanisms. Nat Rev Rheumatol 2013;9:741–50. doi:10.1038/nrrheum.2013.165
      OpenUrlCrossRefPubMed
    4. ↵
      1. Utsinger PD
      . Diffuse idiopathic skeletal hyperostosis. Clin Rheum Dis 1985;11:325–51.
      OpenUrlPubMedWeb of Science
    5. ↵
      1. Al-Herz A,
      2. Snip JP,
      3. Clark B, et al
      . Exercise therapy for patients with diffuse idiopathic skeletal hyperostosis. Clin Rheumatol 2008;27:207–10. doi:10.1007/s10067-007-0693-z
      OpenUrlCrossRefPubMed
    6. ↵
      1. Katzman WB,
      2. Huang M-H,
      3. Kritz-Silverstein D, et al
      . Diffuse idiopathic skeletal hyperostosis (DISH) and impaired physical function: the rancho Bernardo study. J Am Geriatr Soc 2017;65:1476–81. doi:10.1111/jgs.14810
      OpenUrl
    7. ↵
      1. Harlianto NI,
      2. Kuperus JS,
      3. Mohamed Hoesein FAA, et al
      . Diffuse idiopathic skeletal hyperostosis of the cervical spine causing dysphagia and airway obstruction: an updated systematic review. Spine J 2022;22:1490–503. doi:10.1016/j.spinee.2022.03.002
      OpenUrl
    8. ↵
      1. Nathan H
      . Osteophytes of the spine compressing the sympathetic trunk and Splanchnic nerves in the Thorax. Spine (Phila Pa 1976) 1987;12:527–32. doi:10.1097/00007632-198707000-00003
      OpenUrl
    9. ↵
      1. Haskard DO,
      2. Panayi GS
      . Multiple peripheral nerve Entrapment in Forestier’s disease (diffuse idiopathic Skeletal Hyperostosis). Rheumatology 1988;27:407–8. doi:10.1093/rheumatology/27.5.407
      OpenUrlCrossRefPubMed
    10. ↵
      1. Oh GS,
      2. Ju CI,
      3. Kim SW, et al
      . Complete cord injury after minimal trauma in a patient with Forestier’s disease accompanying ossification of the posterior longitudinal ligament. J Korean Neurosurg Soc 2007;42:487–9. doi:10.3340/jkns.2007.42.6.487
      OpenUrlPubMed
    11. ↵
      1. Yamada K,
      2. Satoh S,
      3. Hashizume H, et al
      . Diffuse idiopathic Skeletal Hyperostosis is associated with lumbar spinal stenosis requiring surgery. J Bone Miner Metab 2019;37:118–24. doi:10.1007/s00774-017-0901-0
      OpenUrl
    12. ↵
      1. Tsuji O,
      2. Suda K,
      3. Takahata M, et al
      . Early surgical intervention may facilitate recovery of Cervical spinal cord injury in DISH. J Orthop Surg (Hong Kong) 2019;27. doi:10.1177/2309499019834783
    13. ↵
      1. Okada E,
      2. Yoshii T,
      3. Yamada T, et al
      . Spinal fractures in patients with diffuse idiopathic skeletal hyperostosis: a nationwide multi-institution survey. J Orthop Sci 2019;24:601–6. doi:10.1016/j.jos.2018.12.017
      OpenUrl
    14. ↵
      1. Okada E,
      2. Shimizu K,
      3. Kato M, et al
      . Spinal fractures in patients with diffuse idiopathic skeletal hyperostosis: clinical characteristics by fracture level. J Orthop Sci 2019;24:393–9. doi:10.1016/j.jos.2018.10.017
      OpenUrl
    15. ↵
      1. Dan Lantsman C,
      2. Barkay G,
      3. Friedlander A, et al
      . Whole spine CT scan for the detection of acute spinal fractures in diffuse idiopathic skeletal hyperostosis patients who sustained low-energy trauma. Spine (Phila Pa 1976) 2020;45:1348–53. doi:10.1097/BRS.0000000000003536
      OpenUrl
    16. ↵
      1. Guiot A,
      2. Estublier C,
      3. Gaude M, et al
      . Relationship between diffuse idiopathic skeletal hyperostosis and fragility vertebral fracture: a prospective study in older men. Rheumatology (Oxford) 2021;60:2197–205. doi:10.1093/rheumatology/keaa517
      OpenUrl
    17. ↵
      1. Resnick D,
      2. Niwayama G
      . Radiographic and pathologic features of spinal involvement in diffuse idiopathic skeletal hyperostosis (DISH). Radiology 1976;119:559–68. doi:10.1148/119.3.559
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Baraliakos X,
      2. Listing J,
      3. Buschmann J, et al
      . A comparison of new bone formation in patients with ankylosing spondylitis and patients with diffuse idiopathic skeletal hyperostosis: a retrospective cohort study over six years. Arthritis Rheum 2012;64:1127–33. doi:10.1002/art.33447
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Yaniv G,
      2. Bader S,
      3. Lidar M, et al
      . The natural course of bridging Osteophyte formation in diffuse idiopathic Skeletal Hyperostosis: retrospective analysis of consecutive CT examinations over 10 years. Rheumatology (Oxford) 2014;53:1951–7. doi:10.1093/rheumatology/ket335
      OpenUrlCrossRefPubMed
    20. ↵
      1. Kuperus JS,
      2. Buckens CF,
      3. Šprem J, et al
      . The natural course of diffuse idiopathic Skeletal Hyperostosis in the Thoracic spine of adult males. J Rheumatol 2018;45:1116–23. doi:10.3899/jrheum.171091
      OpenUrlAbstract/FREE Full Text
    21. ↵
      1. Fournier DE,
      2. Norley CJD,
      3. Pollmann SI, et al
      . Ectopic spinal calcification associated with diffuse idiopathic skeletal hyperostosis (DISH): a quantitative micro-ct analysis. J Orthop Res 2019;37:717–26. doi:10.1002/jor.24247
      OpenUrl
    22. ↵
      1. Kuperus JS,
      2. Oudkerk SF,
      3. Foppen W, et al
      . Criteria for early-phase diffuse idiopathic skeletal hyperostosis: development and validation. Radiology 2019;291:420–6. doi:10.1148/radiol.2019181695
      OpenUrl
    23. ↵
      1. St Sauver JL,
      2. Grossardt BR,
      3. Yawn BP, et al
      . Data resource profile: the Rochester epidemiology project (Rep) medical records-linkage system. Int J Epidemiol 2012;41:1614–24. doi:10.1093/ije/dys195
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. Rocca WA,
      2. Yawn BP,
      3. St Sauver JL, et al
      . History of the Rochester epidemiology project: half a century of medical records linkage in a us population. Mayo Clin Proc 2012;87:1202–13. doi:10.1016/j.mayocp.2012.08.012
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Murakami Y,
      2. Morino T,
      3. Hino M, et al
      . Progression of ossification of the anterior longitudinal ligament associated with diffuse idiopathic skeletal hyperostosis by age: a study of computed tomography findings over 5 years. Global Spine J 2021;11:656–61. doi:10.1177/2192568220918817
      OpenUrl
    26. ↵
      1. Murakami Y,
      2. Morino T,
      3. Hino M, et al
      . Imaging features of early diffuse idiopathic skeletal hyperostosis (pre-DISH): analysis of progression of ligament ossification over 5 years by computed tomography. Spine Surg Relat Res 2023;7:443–9. doi:10.22603/ssrr.2022-0226
      OpenUrl
    27. ↵
      1. Phang SY,
      2. Barrett C,
      3. Purcell M
      . A radiological study of the natural history of diffuse idiopathic skeletal hyperostosis (DISH): a story of incomplete fusion. Br J Neurosurg 2023;2023:1–10. doi:10.1080/02688697.2023.2197494
      OpenUrl
    28. ↵
      1. Couto AR,
      2. Parreira B,
      3. Power DM, et al
      . Evidence for a genetic contribution to the ossification of spinal ligaments in ossification of posterior longitudinal ligament and diffuse idiopathic skeletal hyperostosis: a narrative review. Front Genet 2022;13:987867. doi:10.3389/fgene.2022.987867
    29. ↵
      1. Sethi A,
      2. Ruby JG,
      3. Veras MA, et al
      . Genetics Implicates overactive osteogenesis in the development of diffuse idiopathic skeletal hyperostosis. Nat Commun 2023;14:2644. doi:10.1038/s41467-023-38279-x
    30. ↵
      1. Gliner-Ron M,
      2. Bercovich E,
      3. Herman A, et al
      . Osteophytes’ position in subjects with DISH and right-sided aorta: verification of the ‘aortic pulsation protective effect’ theory. Rheumatology (Oxford) 2022;61:4910–4. doi:10.1093/rheumatology/keac183
      OpenUrl
    31. ↵
      1. Verlaan JJ,
      2. Westerveld LA,
      3. van Keulen JW, et al
      . Quantitative analysis of the anterolateral ossification mass in diffuse idiopathic skeletal hyperostosis of the thoracic spine. Eur Spine J 2011;20:1474–9. doi:10.1007/s00586-011-1719-4
      OpenUrlPubMed
    32. ↵
      1. Hiyama A,
      2. Katoh H,
      3. Sakai D, et al
      . Prevalence of diffuse idiopathic Skeletal Hyperostosis (DISH) assessed with whole-spine computed tomography in 1479 subjects. BMC Musculoskelet Disord 2018;19:178. doi:10.1186/s12891-018-2108-5
    33. ↵
      1. Ahmed O,
      2. Ramachandran K,
      3. Patel Y, et al
      . Diffuse idiopathic skeletal hyperostosis prevalence, characteristics, and associated comorbidities: a cross-sectional study of 1815 whole spine CT scans. Global Spine J 2022. doi:10.1177/21925682221136844
    34. ↵
      1. Chen H,
      2. Zhou Q,
      3. Wang S, et al
      . Not all osteophytes are located on the right side of the vertebrae in diffuse idiopathic skeletal hyperostosis: a quantitative analysis in relation to the position of aorta. Orthop Surg 2023;15:2881–8. doi:10.1111/os.13869
      OpenUrl
    35. ↵
      1. Fournier DE,
      2. Kiser PK,
      3. Beach RJ, et al
      . Dystrophic calcification and heterotopic ossification in fibrocartilaginous tissues of the spine in diffuse idiopathic skeletal hyperostosis (DISH). Bone Res 2020;8:16. doi:10.1038/s41413-020-0091-6
    36. ↵
      1. Kiss C,
      2. Szilágyi M,
      3. Paksy A, et al
      . Risk factors for diffuse idiopathic skeletal hyperostosis: a case-control study. Rheumatology (Oxford) 2002;41:27–30. doi:10.1093/rheumatology/41.1.27
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Pillai S,
      2. Littlejohn G
      . Metabolic factors in diffuse idiopathic skeletal hyperostosis - a review of clinical data. Open Rheumatol J 2014;8:116–28. doi:10.2174/1874312901408010116
      OpenUrl
    38. ↵
      1. Pariente-Rodrigo E,
      2. Sgaramella GA,
      3. Olmos-Martínez JM, et al
      . Relationship between diffuse idiopathic skeletal hyperostosis, abdominal aortic calcification and associated metabolic disorders: data from the Camargo cohort. Medicina Clínica (English Edition) 2017;149:196–202. doi:10.1016/j.medcle.2017.07.020
      OpenUrl
    39. ↵
      1. Oudkerk SF,
      2. Mohamed Hoesein FAA,
      3. PThM Mali W, et al
      . Subjects with diffuse idiopathic skeletal hyperostosis have an increased burden of coronary artery disease: an evaluation in the Copdgene cohort. Atherosclerosis 2019;287:24. doi:10.1016/j.atherosclerosis.2019.05.030
    40. ↵
      1. Okada E,
      2. Ishihara S,
      3. Azuma K, et al
      . Metabolic syndrome is a predisposing factor for diffuse idiopathic skeletal hyperostosis. Neurospine 2021;18:109–16. doi:10.14245/ns.2040350.175
      OpenUrl
    41. ↵
      1. Harlianto NI,
      2. Westerink J,
      3. Foppen W, et al
      . Visceral adipose tissue and different measures of adiposity in different severities of diffuse idiopathic skeletal hyperostosis. J Pers Med 2021;11:1–11. doi:10.3390/jpm11070663
      OpenUrl

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors Substantial contributions to the conception or design of the work or the acquisition, analysis or interpretation of data for the work—DEF, MB and CS. Drafting the work or revising it critically for important intellectual content—DEF, MB and CS. Final approval of the version to be published—DEF, MB and CS. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved—DEF, MB and CS. Author acting as guarantor - CS.

    • Funding This study used the resources of the Rochester Epidemiology Project (REP) medical records-linkage system, which is supported by the National Institute on Aging (NIA; AG 058738), by the Mayo Clinic Research Committee, and by fees paid annually by REP users.This study used the resources of the Rochester Epidemiology Project (REP) medical records-linkage system, which is supported by the National Institute on Aging (NIA; AG 058738), by the Mayo Clinic Research Committee, and by fees paid annually by REP users.

    • Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Mayo Clinic.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.