Erosive costovertebral joint arthritis as a novel feature of ankylosing spondylitis

Introduction

Axial spondyloarthritis (SpA) is a chronic inflammatory disease that mainly affects the axial skeleton and has a global prevalence of 0.32–1.4% (1-3). SpA includes non-radiographic axial spondyloarthritis (nr-axSpA) and radiographic axial SpA (r-axSpA), which is also termed ankylosing spondylitis (AS) (2). According to the 1984 modified diagnostic criteria, sacroiliac joint (SIJ) arthritis is regarded as the gold standard for AS diagnosis (4). However, numerous studies have demonstrated that SIJ arthritis in patients with AS lacks specificity and has a limited value in indicating whether the spinal joints are fused (5,6). Furthermore, it mainly takes several years from the onset of clinical symptoms to definitive imaging changes in the SIJ, making the early diagnosis of AS a challenging task (7). Consequently, emphasizing the exploration of tissue lesions beyond the SIJ becomes imperative for the early diagnosis of AS.

Owing to the obscuration caused by adjacent anatomical structures surrounding the thoracic spine, the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS), which is presently employed for assessing osseous structural damage in the spine, is limited to the cervical and lumbar regions. As a result, thoracic lesions are frequently neglected, despite the thoracic spine’s heightened susceptibility to structural damage in AS (8,9). Additionally, standard radiographic images make it difficult to adequately examine the lateral and posterior regions of the vertebrae (10). Prior research has found that thoracic pain in AS patients was linked to arthritis in the costovertebral joints (CVJs) (11-13). Restrictive ventilatory impairment might occur due to limited chest wall movement, as the CVJ involvement can cause gradual joint fusion (14,15). Currently, there is limited knowledge about how often and what clinical factors affect CVJ involvement in AS patients, due to the constraints of radiographic assessments.

The posterolateral aspect of the vertebral body, in conjunction with the closely associated rib head, collectively constitutes the CVJ. Within this junction, hyaline cartilage and a synovial membrane are present (16). These synovial joints are highly susceptible to local and systemic inflammation and spread to adjacent vertebrae. Furthermore, CVJs are important movable joints of the thorax, moving with respiration, as well as causing constant changes in the local mechanical stress environment (17,18). Chronic inflammation and local mechanical stress abnormalities are important mechanisms of periosteal new bone (PNB) formation in AS patients (19-23). Therefore, it is essential to determine whether the CVJ, which is closely associated with thoracic mobility, could be a new biomarker for the early diagnosis of AS.

Initially, we investigated the prevalence of CVJ arthritis in patients diagnosed with axSpA who were previously deemed to exhibit no radiographic abnormalities. This was undertaken to indicate the presence of erosive CVJ arthritis and explore its potential significance in facilitating the early diagnosis of AS, Furthermore, magnetic resonance imaging (MRI) and computed tomography (CT) were utilized for the grade of erosive CVJ arthritis. The aim was to investigate potential associations with the distribution of PNB in the posterolateral aspect of the upper endplate of the vertebral body. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-356/rc).

Methods

Ethics approval

The study protocol was approved by the Ethics Committee of The Affiliated Hospital of Southwest Medical University (No. KY2024431). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. As this study was retrospective in nature and exclusively utilized patients’ imaging data, with rigorous anonymization of identifiable patient information maintained throughout the research process, the requirement for obtaining patient-informed consent was waived.

Participant selection

To examine the prevalence of CVJ arthritis among individuals with nr-axSpA, we identified patients diagnosed with nr-axSpA who fulfilled the Assessment of SpondyloArthritis International Society (ASAS) classification criteria from September 2018 to December 2023 as the participants in our study. The CVJ and SIJ imaging data of patients with no abnormality detected upon physical examination that matched the baseline data in the nr-axSpA group during the same period were selected as the control group. The inclusion criteria were summarized as follows: (I) the patient’s age <45 years; (II) back pain persisting for ≥3 months; (III) human leukocyte antigen B27 (HLA-B27) positivity; and (IV) meeting the aforementioned criteria along with two or more of the following SpA features: inflammatory back pain (IBP), arthritis, enthesitis (heel), uveitis, dactylitis, psoriasis, Crohn’s disease or ulcerative colitis, positive response to non-steroidal anti-inflammatory drugs (NSAIDs), family history of SpA, or elevated C-reactive protein (CRP) level. The exclusion criteria were summarized as follows: (I) observation scope affected by surgical treatment for other conditions; (II) concurrent congenital malformations, tumors, infections, metabolic bone diseases, and so on, impacting outcomes; and (III) suspicion of SIJ arthritis.

Additionally, to investigate the relationship between CVJ arthritis and PNB formation at the upper margin of the vertebral endplate, we utilized clinical and imaging data from patients diagnosed with r-axSpA, also referred to as AS, who satisfied the modified New York criteria within the same timeframe. The inclusion criteria were as follows: (I) severe kyphotic deformity, inability to look straight ahead with both eyes when the knees and hips were straightened, quiescent disease, acceptable general condition, and the absence of obvious underlying diseases that would interfere with anesthesia and surgery; (II) patients’ age ≥18 years; and (III) the availability of MRI and CT imaging data for the T4–L5 vertebrae and the SIJ. The exclusion criteria were as follows: (I) patients who underwent surgical treatment within the scope of observation; (II) patients experiencing congenital malformation, tumor, infection, and so on; and (III) patients with complete fusion of the vertebral bodies that could lead to an unclear display of anatomical structures [Bath Ankylosing Spondylitis Radiology Index (BASRI) score >3].

Clinical assessment and laboratory analysis

Participants’ clinical and demographic data were collected, including age, sex, disease duration, and bone mineral density (BMD). All participants completed three comprehensive questionnaires, involving Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), and Bath Ankylosing Spondylitis Global Index (BASGI). Several blood parameters were recorded, including HLA-B27, CRP, and erythrocyte sedimentation rate (ESR).

MRI and CT parameters

MRI parameters

Whole-spine MRI was performed using 1.5T Magnetom Vision (Signal HDe, GE Healthcare, Tokyo, Japan) via a torso coil, and participants assumed a supine position. The imaging parameters and methods were as follows: The sagittal sequence captured 11–15 slices with a slice thickness of 3–4 mm; the axial sequence involved the acquisition of 200–250 slices with a slice thickness of 3–4 mm. Following the selection of the sagittal section orientation, the subsequent sequences were employed: (I) T1-weighted spin-echo (T1W) [time to recovery (TR), 500–640 msec; time to echo (TE), 15 msec]; (II) short-tau inversion recovery (STIR) (TR, 3,500–4,000 msec, time to inversion, 135–160 msec, TE, 40–60 msec). Evaluation of MR images spanned from C2 to S1, dividing the spine into two sections: the upper half, encompassing the entire cervical and a major portion of the thoracic spine, and the lower half, encompassing the lower part of the thoracic spine and the entire lumbar spine.

The assessment of the SIJ through MRI should adhere to the specified orientations and sequences (24,25). The sequences were summarized as follows: (I) STIR with a TR of 2,500–4,000 and TE of 600; (II) T1W with TR of 500–700 and TE of 10–55. The recommended orientations were summarized as follows: (I) coronal oblique, where slices run parallel to the dorsal surface of the S2 vertebra; (II) semiaxial, with slices perpendicular to the dorsal cortex of the S2 vertebra; (III) maximal slice thickness set at 3 mm; and (IV) interslice gap maintained at 0.3 mm.

CT parameters

All patients underwent a 64-slice plain CT scan from T4 to the SIJ (Light Speed VCT, GE Healthcare), with a layer thickness of 0.625 mm, field of view (FOV) of 550–600 mm, matrix of 512×512 pixels, voltage of 140 kVp, and current of 80 mAs. The original CT imaging data were imported to an Advantage Workstation (Version ADW 4.4, GE Healthcare) for reconstruction and measurement.

Image processing and data analysis

Qualitative assessment of erosive CVJ arthritis

MRI and CT data were reviewed by an orthopedic surgeon (S.Z. with 8 years of experience in spine MRI and CT) and a rheumatologist (S.W. with 20 years of experience in SIJ MRI and CT). Both readers were blinded to clinical and laboratory findings. On the basis of the classification of CVJ arthritis proposed by Cerrahoglu et al. (26), CVJ arthritis grading was determined based on lesion severity through a combination of MRI and CT data (Figure 1A-1H), delineated as follows: Grade 0 indicating normal conditions; Grade I denoting bone marrow edema in the rib head or the corresponding vertebral body; Grade II involving sclerosis, rough margins, moth-eaten changes, or bone cysts in the rib head or the corresponding vertebral body, while maintaining a normal joint space; Grade III indicating joint space narrowing in addition to Grade II features; and Grade IV signifying fusion of the CVJ with the vertebral body at any position (Figure 1I-1R).

Figure 1 Classification of CVJ arthritis. (A-D) The presentation of normal CVJ in X-ray, CT, and MRI images. The arrows indicate normal CVJ. (E-H) The presentation of erosive CVJ arthritis in X-ray, CT, and MR images. The arrows indicate lesion of CVJ. (I,J) Grade 0: normal. (K,L) Grade I: MR image showing bone marrow edema in the rib head and the corresponding vertebrae, whereas the CT scan image displaying no abnormality. (M,N) Grade II: sclerosis, rough edges, insect-like damage, or bone cysts in the rib head or the corresponding vertebral body, with a normal joint space. (O,P) Grade III: Grade II with joint space narrowing. (Q,R) Grade IV: fusion of the CVJ with the vertebral body at any position. CT, computed tomography; CVJ, costovertebral joint; MR, magnetic resonance; MRI, magnetic resonance imaging.

Qualitative assessment the distribution of PNB on the vertebral body

CT scan data were evaluated by two orthopedic surgeons (S.Z. with 8 years of experience in spine CT and S.W. with 21 years of experience in spine CT). PNB was defined as the presence of bony outgrowth at the upper margin of vertebral body, which did not reach the adjacent vertebrae. The upper endplate of the vertebral body was divided into eight 45° fan-shaped areas (I to VIII) on the CT axial image. The synchronous positioning technology was employed to indicate the existence of PNB formation on the axial, coronal, and sagittal images of the same area.

Qualitative assessment of SIJ arthritis

MRI and CT data were reviewed by an orthopedic surgeon (S.Z. with 8 years of experience in spine MRI and CT) and a rheumatologist (S.W. with 20 years of experience in SIJ MRI and CT data). According to the ASAS guidelines, structural damage lesions of the SIJ can appear as: (I) subchondral sclerosis; (II) erosions; (III) fat deposition; and (IV) bony bridges/ankylosis. Combined with the CT grading criteria for SIJ arthritis proposed by Shenkman et al. (27), SIJ arthritis in patients with AS was graded as follows: Grade 0: normal; Grade I: osteoporosis around the joints, mild erosion of the subchondral bone, blurred articular surface, and normal joint space; Grade II: joint surface exhibited blurring, accompanied by subchondral bone destruction, osteoporosis, or sclerosis, while maintaining a normal joint space. Grade III: notable subchondral bone destruction, diffuse sclerosis, a burr-like articular surface, narrowed or uneven joint space, and partial joint rigidity. Grade IV: widespread bone destruction, osteosclerosis, and osteoporosis affecting all joints, leading to complete joint ankylosis.

All readers were trained together to become familiar with identifying different lesions before performing the image analysis, aiming to standardize the image analysis criteria in this study. The readers discussed the imaging data and agreed on the definitions. A unique image number was allocated to each patient and control, thereby ensuring blinding to all patients’ demographics. To assess intra-observer agreement, one orthopedic surgeon (S.Z.) evaluated images twice. To assess interobserver agreement, two readers (S.Z. and X.D.) evaluated 50 randomly selected patients’ erosive CVJ arthritis, PNB of vertebrae, or SIJ arthritis.

Statistical analysis

The Wilcoxon rank-sum test was employed to determine whether there was a difference in the grading of the left and right CVJ arthritis in the same vertebral body. The differences in the formation of PNB in quadrants I and IV, II and III, V and VIII, and VI and VII of the same vertebral body were determined using the dichotomous McNemar test. The variation in the occurrence of PNB formation in distinct sectors of the same vertebral body was evaluated using the R×C contingency table Chi-squared test with post-hoc testing. Given the multiple comparisons undertaken, this study employed the absolute values of the adjusted standardized residuals, setting a threshold at 3 for significance. The difference in the incidence of PNB formation at the posterolateral aspect of the upper endplate of the vertebral body between thoracic and lumbar vertebrae was found. Additionally, the analysis of CVJ arthritis incidence between the AS group and the control group was conducted using the chi-square test. The correlation between the grade of CVJ arthritis and the presence or absence of PNB at the posterolateral aspect of the upper endplate of the vertebral body was expressed by the Cramer’s V coefficient in the chi-square contingency table test. The consistency of the CVJ arthritis and SIJ arthritis grading in the same patient was tested by Kendall’s tau-b rank correlation test. The level of significance was set at P<0.05. Spearman correlation analysis was employed to analyze the correlation between CVJ arthritis grading and BASDAI, BASFI, and BASGI scores. Intra- and inter-observer agreements were assessed by calculating the Cohen kappa coefficient. A kappa of <0.00 was interpreted as minimal agreement, 0.00–0.20 as slight agreement, 0.21–0.40 as fair agreement, 0.41–0.60 as moderate agreement, and 0.61 or over as substantial agreement. The statistical analysis was carried out using SPSS 22.0 software (IBM Corp., Armonk, NY, USA).

Results

A total of 48 patients with nr-axSpA according to ASAS classification criteria, yet without SIJ imaging lesions, were included in this study. Moreover, 53 patients with AS requiring orthopedic surgery were finally enrolled in this study. The BASRI scores were 0 in 6 patients, 1 in 5, 2 in 16, and 3 in 26. Notably, 41 healthy controls undergoing physical examinations were allocated to the control group, matched with the baseline data of the aforementioned two groups over the same period. Participants’ demographic and clinical characteristics are presented in Table 1.

The prevalence and distribution of CVJs arthritis

In the control group, 3 patients were found to have erosive CVJ arthritis, resulting in an incidence of 7.2% (3 out of 41), with 14 CVJs affected, indicating an incidence of 1.9% (14 out of 738). The severity distribution for these cases ranged from degrees 0 to IV, with frequencies of 714, 7, 6, 1, and 0, respectively (Figure 2A). Among patients diagnosed with nr-axSpA according to the ASAS classification criteria, yet without SIJ imaging changes, 14 cases experienced erosive CVJ arthritis, yielding an incidence of 29.2% (14 out of 48). A total of 224 CVJs were affected, representing an incidence of 25.9% (224 out of 864). The severity degrees for these arthritis cases were I in 88, II in 67, III in 65, and IV in 4 cases. The remaining 640 CVJs exhibited no arthritis, scoring a severity level of zero (Figure 2B). Among patients who were diagnosed with AS and required orthopedic surgery, erosive CVJ arthritis was found in 51 patients with an incidence of 96.2% (51/53). Moreover, 740 CVJs were invaded, with an incidence of 77.6% (740/954). The Wilcoxon rank-sum test employed for pairwise comparison of continuous and categorical variables revealed that there was no difference in the grading of CVJ arthritis on the left and right sides of the same vertebral body (Table 2). Notably, the left and right sides were combined, resulting in degrees 0–IV with corresponding frequencies of 214, 88, 129, 251, and 272 (Figure 2C).

Figure 2 The incidence of CVJ arthritis in different groups. (A) The incidence of CVJ arthritis in the control group. (B) The incidence of CVJ arthritis in patients diagnosed with nr-axSpA. (C) The incidence of CVJ arthritis in patients diagnosed with AS who required orthopedic surgery. (D) The segmental distribution of CVJ arthritis in the control group. (E) The segmental distribution of CVJ arthritis in patients diagnosed with nr-axSpA. (F) The segmental distribution of CVJ arthritis in patients diagnosed with AS who required orthopedic surgery. AS, ankylosing spondylitis; CVJ, costovertebral joints; nr-axSpA, non-radiographic axial spondyloarthritis.

The distribution of CVJs arthritis is shown in Figure 2D-2F. The incidence of erosive CVJ arthritis in both AS patients requiring orthopedic surgery and those diagnosed with nr-axSpA was significantly higher than that in the control group (χ²=84.436, P<0.001). Intra- and inter-observer agreements for CVJ arthritis assessment were both regarded “substantial”, with kappa values of 0.86 and 0.82, respectively.

Characteristics of the distribution of PNB at the vertebral body

The upper endplate of the vertebral body was divided into 8 45° fan-shaped areas (I–VIII) on the CT axial image (Figure 3A). The results of the paired McNemar testing of dichotomous variables revealed that there was no difference in the incidence of PNB in quadrants I and IV, II and III, VI and VII, and V and VIII of the same vertebral body; thus, the left and right sides could be combined for calculation purposes. Quadrants I and IV, II and III, VI and VII, and V and VIII were defined as the anterolateral vertebral body, anterior vertebral body, posterior vertebral body, and posterolateral vertebral body, respectively. The heat map displayed a higher incidence of PNB on posterolateral aspect of the upper endplate of the vertebral body adjacent to the CVJ (χ²=1.074, P<0.001; the adjusted standardized residual was 30.3) (Figure 3B,3C, Table 3). The incidence of PNB on the posterolateral aspect of the upper endplate of the vertebral body was significantly higher than that of the lumbar vertebrae (χ²=1.692, P<0.001) (Figure 3D). The heat map displayed that the probability of PNB formation on the posterolateral aspect of the vertebral body’s upper edge increased with the CVJ lesion severity (χ²=565.029; P<0.001; Cramer’s V coefficient, 0.770; P<0.001; Figure 3E). The intra- and inter-observer agreements for the assessment of PNB distribution were both “substantial”, with kappa values of 0.79 and 0.84, respectively.

Figure 3 Characteristics of the PNB distribution on the vertebral body’s upper edge and the correlation between the degree of CVJ arthritis and PNB distribution. (A-C) The vertebral body’s upper endplate was divided into eight 45° fan-shaped sections (I–VIII) on the axial CT images. The heat map and pie chart demonstrate the spatial distribution pattern of PNB in the vertebral body’s upper edge. (D,E) Correlation analysis between the degree of CVJ arthritis and PNB distribution on the posterolateral aspect of the vertebral body’s upper endplate. CT, computed tomography; CVJ, costovertebral joint; PNB, periosteal new bone.

Correlation of erosive CVJ arthritis with BASFI, BASDAI, and BASGI scores

In patients with AS who required orthopedic surgery, the severity of erosive CVJ arthritis was significantly correlated with BASFI score (R²=0.9145, P<0.0001) (Figure 4A) and BASGI score (R²=0.6604, P<0.0001) (Figure 4B), whereas it was not correlated with BASDAI score (R²=0.04413, P=0.1311) (Figure 4C). In the subset of patients diagnosed with axial SpA following the criteria outlined in the ASAS classification, and specifically characterized by the absence of observable changes in SIJ imaging, the severity of erosive CVJ arthritis was not correlated with BASFI score (R²=0.0017, P=0.7814) (Figure 4D) and BASGI score (R²=0.07855, P=0.0537) (Figure 4E), whereas it significantly correlated with BASDAI score (R²=0.6114, P<0.0001) (Figure 4F).

Figure 4 The correlation between the degree of CVJ arthritis and BASFI, BASDAI, and BASGI in different groups. (A,D) The correlation between the degree of CVJ arthritis and BASFI. (B,E) The correlation between the degree of CVJ arthritis and BASGI. (C,F) The correlation between the degree of CVJ arthritis and BASDAI. BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASGI, Bath Ankylosing Spondylitis Global Index; CVJ, costovertebral joint.

The consistency between the grade of CVJ arthritis and the grade of SIJ arthritis

In the analysis of lesion severity consistency between CVJ arthritis and SIJ arthritis within the same patient, the side exhibiting less severe SIJ arthritis was incorporated into the study. The included patients with AS had various degrees of SIJ arthritis, with 6 classified as grade I, 13 as grade II, 26 as grade III, and 8 as grade IV. Intra- and inter-observer agreements for SIJ arthritis assessment were regarded “substantial”, with kappa values of 0.79 and 0.84, respectively. The highest CVJ and SIJ arthritis grades were selected separately for correlation analysis in the same patients, exhibiting a moderate correlation between them (R²=0.589; P<0.001; Figure 5).

Figure 5 Consistency between the CVJ and SIJ arthritis degree in patients with AS. AS, ankylosing spondylitis; CVJ, costovertebral joint; SIJ, sacroiliac joint.

A typical case presentation

A 28-year-old male patient presented with a primary complaint of thoracic back pain and bilateral heel pain persisting for over 9 months. The discomfort intensified during prolonged recumbency and was accompanied by stiffness and lower back discomfort upon rising, which were alleviated by movement. Relevant assessments yielded a BASFI score of 1.3 and a BASDAI score of 4.15. Laboratory results indicated HLA-B27 positivity, with an ESR of 38.4 mm/h and CRP level of 11.2 mg/L. Rheumatoid factor and anti-cyclic citrulline antibodies were negative. Notably, oral celecoxib at a dose of 600 mg/d resulted in significant relief of chest and back pain. Pelvic X-ray, as well as MRI and CT scan of the SIJ, did not reveal any discernible lesions (Figure 6A-6C). Radiographs capturing the full length of the spine did not demonstrate PNB formation (Figure 6D). However, comprehensive spine MRI and CT examinations suggested the presence of varying degrees of arthritis in several CVJs, with the most severe instances reaching grade III according to the grading criteria utilized in the present study (Figure 6E-6G).

Figure 6 A typical case. (A-C) X-ray, CT, and MRI of the sacroiliac joint did not reveal any lesions. (D) No PNB was found in the full-length film of the spine. (E-G) CT and MRI revealed that CVJ arthritis was grade III. CT, computed tomography; CVJ, costovertebral joint; MRI, magnetic resonance imaging; PNB, periosteal new bone.

Discussion

Early diagnosis of AS is of great significance for the prognosis of patients, and the imaging changes of the SIJ remain the gold standard for the diagnosis of AS (4). However, numerous studies have suggested that the SIJ arthritis of AS patients lacks specificity (5,6). Therefore, there is a need to explore new biomarker for early the diagnosis of AS. Notably, AS patients mainly complain of diffuse chest and back pain, stiffness, and discomfort. Due to the reduced attention given to the pathological changes of the posterior elements, such as the CVJs, it has been demonstrated that the abovementioned symptoms may be correlated with the progressive kyphosis caused by spinal fusion, and also attribute these symptoms to fasciitis, muscle fatigue, and so on (26,28,29). In particular, in patients lacking the typical manifestations of SIJ arthritis, these nonspecific symptoms are easily overlooked or misdiagnosed as other systemic diseases (6). Previous studies have confirmed that CVJs and the surrounding ligament joint capsules are densely innervated by sympathetic nerves, and the secretion of pain-stimulating factors in the diseased joints significantly increases, making patients prone to chest and back pain (16,30,31). AS influences synovial joints that exist within a complex biomechanical environment. The interplay of these factors creates an ideal setting for inflammation development and the formation of PNB. Therefore, the present study aimed to evaluate CVJ arthritis as an imaging biomarker to facilitate the early diagnosis of AS.

The thoracic spine is encased by a multitude of adjacent tissues and organs and is enveloped by substantial soft tissue, which led to its exclusion from the mSASSS system traditionally employed to assess the severity of AS. Nonetheless, numerous studies have identified the thoracic spine as potentially the most frequently affected region in AS (8,9). This anatomical complexity poses challenges for the detection of lesions in the costovertebral and costotransverse joints using standard radiographic imaging techniques. In order to surpass the constraints of radiographic assessment, CT has been applied to evaluate the structural damage of the whole spine in patients with axSpA (8,9,32,33). CT scans offer the possibility to assess the entire spinal segment, including the anterior, lateral, and posterior parts of the vertebrae, with the consideration of acceptable radiation exposure (32,33). In patients with radiographic axSpA, the CT Syndesmophyte Score demonstrated a consistent evaluation of new bone formation throughout the spine using CT (8). The lateral and posterior borders of the vertebral body, as well as adjacent articular structures such as the facet joints and CVJs, are also visualized by CT. Whole-spine CT scans in recent studies have demonstrated widespread facet joint erosion and ankylosis in radiographic axSpA, with a notable presence in the thoracic spine (9,32,33). In the present study, it was found that 96.2% of patients diagnosed with AS who required orthopedic surgery had a combination of different degrees of erosive CVJ arthritis, which was consistent with the degree of SIJ arthritis. This emphasizes that both the CVJs and SIJs, which are synovial joints, are highly susceptible to inflammation and abnormal biomechanical loading during the process of AS. In addition, it was revealed that the degree of CVJ arthritis was significantly correlated with BASFI and thoracic mobility in this group of patients. One potential explanation for this phenomenon could be the anatomical configuration of the CVJ, where the rib head, upper, and lower vertebrae are integral components. In the event of localized PNB, there may be a substantial constraint on the activity of the spinal motor unit. Secondly, as the most important movable joint in maintaining thoracic respiratory movement, once it is stimulated by inflammation or abnormal mechanical stress, or PNB formation, the CVJ may significantly restrict thoracic movement and even result in respiratory function limitation. However, as all patients were in a quiescent phase of the disease, this study did not find a significant correlation between the severity of CVJ arthritis and the patient’s BASDAI score. Meanwhile, in patients diagnosed with nr-axSpA according to ASAS criteria, comparison with the control group revealed that the incidence of CVJ arthritis in this group of patients could reach 29.2%, although there were no SIJ imaging changes that could meet the traditional diagnostic criteria. However, given that this patient cohort was in the initial phase of the disease, their primary symptoms included pain in the chest, back, lumbosacral region, buttocks, and other joints, as well as the absence of PNB formation, the degree of CVJ arthritis in this group of patients was significantly positively correlated with the BASDAI score, rather than with the BASFI score. According to the abovementioned research results, it was found that CVJ arthritis is a common lesion that appears in nr-axSpA disease, especially in cases with obvious IBP, rather than SIJ lesions. Moreover, CVJ has the characteristics of wide distribution and multiple targets, thus, it may serve as a potential biomarker to promote the early diagnosis of AS.

The most severe clinical consequence of AS is the formation of intervertebral bone bridges, restricting spinal motion and eventually leading to kyphotic deformity (2,34). Understanding the origin and developmental pattern of PNB can provide valuable insights for clinical diagnosis and treatment (5,35-38). It remains elusive whether the closely connected CVJ is associated with the distribution of PNB. In the present study, CT examination revealed a higher incidence of PNB on the posterolateral side of the upper margin of the thoracic vertebral body in AS patients, whereas no similar spatial distribution difference in the incidence of PNB was found on the upper margin of the lumbar vertebral body in the same patients. Tan et al. pointed out that in the three-column theory of the spine, the posterolateral aspect of the vertebral body, as the main structure of the middle column, bears a greater mechanical load than other parts, thus, PNB formation may occur, and the mechanical load is an important factor for the formation of PNB in patients with AS (35). Although the result of this study is consistent with Tan et al.’s finding, the present study indicated that the incidence of PNB in the posterolateral side of the upper margin of the thoracic vertebral body was closely correlated with the degree of CVJ arthritis. Combined with previous studies, AS patients are mainly accompanied with stiff kyphosis, the center of gravity of the body is moved forward, and the weight-bearing component of the vertebral body is mostly the anterior column (39,40). Notably, CVJs, which are closely adjacent to the posterolateral side of the vertebral body, are an important basis for the formation of secondary PNB, aligning with their anatomical characteristics or the complexity of their physiological functions. At present, the comprehensive understanding of PNB formation in different anatomical regions of AS is far from being explained solely by a single risk factor (1-3,19,23). However, taking into account the anatomical structures adjacent to the site of PNB may provide a reference to further understand the mechanism of PNB formation.

This study had several limitations. Firstly, the included patients had a long disease course; thus, the incidence of CVJ arthritis might be higher than that in other studies. Secondly, patients in the nr-axSpA group had short follow-up time. Finally, although CVJ arthritis was detected, whether subsequent scans will show definitive SIJ arthritis that meet the current New York diagnostic criteria remains undetermined.

Conclusions

In summary, through the application of CT and MRI imaging techniques, it was determined that 29.2% of patients with nr-axSpA, previously thought to exhibit no radiographic alterations, actually presented with varying degrees of CVJ lesions. In patients with AS, the severity of CVJ arthritis demonstrated a positive correlation with the incidence of PNB formation and showed a moderate correlation with SIJ arthritis within the same patient cohort. This finding suggested that CVJ arthritis could potentially serve as a novel biomarker for the diagnosis of AS.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-356/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-356/dss

Funding: This work was jointly funded by Sichuan Science and Technology Program (No. 2022-YFS0628), Research Startup Fund Sponsored Project of Southwest Medical University Hospital (No. 25034), and Natural Science Foundation of Sichuan (No. 2023NSFSC0333).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-356/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are 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. The study protocol was approved by the Ethics Committee of The Affiliated Hospital of Southwest Medical University (No. KY2024431). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. As this study is retrospective in nature and exclusively utilizes patients’ imaging data, with rigorous anonymization of identifiable patient information maintained throughout the research process, the requirement for obtaining patient-informed consent was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ranganathan V, Gracey E, Brown MA, Inman RD, Haroon N. Pathogenesis of ankylosing spondylitis - recent advances and future directions. Nat Rev Rheumatol 2017;13:359-67. [Crossref] [PubMed]
2. Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet 2017;390:73-84. [Crossref] [PubMed]
3. Poddubnyy D, Sieper J. Treatment of Axial Spondyloarthritis: What Does the Future Hold? Curr Rheumatol Rep 2020;22:47. [Crossref] [PubMed]
4. van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361-8. [Crossref] [PubMed]
5. Braun J, Baraliakos X, Buehring B, Kiltz U, Fruth M. Imaging of axial spondyloarthritis. New aspects and differential diagnoses. Clin Exp Rheumatol 2018;36:35-42.
6. Østergaard M. MRI of the sacroiliac joints: what is and what is not sacroiliitis?. Curr Opin Rheumatol 2020;32:357-64.
7. Mau W, Zeidler H, Mau R, Majewski A, Freyschmidt J, Stangel W, Deicher H. Clinical features and prognosis of patients with possible ankylosing spondylitis. Results of a 10-year followup. J Rheumatol 1988;15:1109-14.
8. de Bruin F, de Koning A, van den Berg R, Baraliakos X, Braun J, Ramiro S, van Gaalen FA, Reijnierse M, van der Heijde D. Development of the CT Syndesmophyte Score (CTSS) in patients with ankylosing spondylitis: data from the SIAS cohort. Ann Rheum Dis 2018;77:371-7. [Crossref] [PubMed]
9. Stal R, van Gaalen F, Sepriano A, Braun J, Reijnierse M, van den Berg R, van der Heijde D, Baraliakos X. Facet joint ankylosis in r-axSpA: detection and 2-year progression on whole spine low-dose CT and comparison with syndesmophyte progression. Rheumatology (Oxford) 2020;59:3776-83. [Crossref] [PubMed]
10. Jung SM, Kim MY, Hong YS, Park SH, Kang KY. Costovertebral joint involvement in patients with axial spondyloarthritis. Joint Bone Spine 2023;90:105546. [Crossref] [PubMed]
11. Ellrodt A, Goldberg D, Oberlin F, Huchet B, Cayla J. Erosive arthritis of the costovertebral joint in seronegative spondyloarthropathy. J Rheumatol 1986;13:452-4.
12. Le T, Biundo J, Aprill C, Deiparine E. Costovertebral joint erosion in ankylosing spondylitis. Am J Phys Med Rehabil 2001;80:62-4. [Crossref] [PubMed]
13. Gazitt T, Nassrallah N, Zisman D. Erosive Costovertebral Joint Arthritis-An Uncommon Manifestation of Ankylosing Spondylitis. Arthritis Rheumatol 2021;73:179. [Crossref] [PubMed]
14. Ortancil O, Sarikaya S, Sapmaz P, Basaran A, Ozdolap S. The effect(s) of a six-week home-based exercise program on the respiratory muscle and functional status in ankylosing spondylitis. J Clin Rheumatol 2009;15:68-70. [Crossref] [PubMed]
15. Kanathur N, Lee-Chiong T. Pulmonary manifestations of ankylosing spondylitis. Clin Chest Med 2010;31:547-54. [Crossref] [PubMed]
16. Goldthwait JE. The rib joints. N Engl J Med 1940;223:568-73.
17. HART FD. EMERSON PA, GREGG I. Thorax in ankylosing spondylitis. Ann Rheum Dis 1963;22:11-8. [Crossref] [PubMed]
18. Duprey S, Subit D, Guillemot H, Kent RW. Biomechanical properties of the costovertebral joint. Med Eng Phys 2010;32:222-7. [Crossref] [PubMed]
19. Lories RJ, Derese I, Luyten FP. Modulation of bone morphogenetic protein signaling inhibits the onset and progression of ankylosing enthesitis. J Clin Invest 2005;115:1571-9. [Crossref] [PubMed]
20. Diarra D, Stolina M, Polzer K, Zwerina J, Ominsky MS, Dwyer D, Korb A, Smolen J, Hoffmann M, Scheinecker C, van der Heide D, Landewe R, Lacey D, Richards WG, Schett G. Dickkopf-1 is a master regulator of joint remodeling. Nat Med 2007;13:156-63. [Crossref] [PubMed]
21. Uderhardt S, Diarra D, Katzenbeisser J, David JP, Zwerina J, Richards W, Kronke G, Schett G. Blockade of Dickkopf (DKK)-1 induces fusion of sacroiliac joints. Ann Rheum Dis 2010;69:592-7. [Crossref] [PubMed]
22. Li X, Wang J, Zhan Z, Li S, Zheng Z, Wang T, Zhang K, Pan H, Li Z, Zhang N, Liu H. Inflammation Intensity-Dependent Expression of Osteoinductive Wnt Proteins Is Critical for Ectopic New Bone Formation in Ankylosing Spondylitis. Arthritis Rheumatol 2018;70:1056-70. [Crossref] [PubMed]
23. Jacques P, McGonagle D. The role of mechanical stress in the pathogenesis of spondyloarthritis and how to combat it. Best Pract Res Clin Rheumatol 2014;28:703-10. [Crossref] [PubMed]
24. Sudoł-Szopińska I, Jurik AG, Eshed I, Lennart J, Grainger A, Østergaard M, et al. Recommendations of the ESSR Arthritis Subcommittee for the Use of Magnetic Resonance Imaging in Musculoskeletal Rheumatic Diseases. Semin Musculoskelet Radiol 2015;19:396-411. [Crossref] [PubMed]
25. Maksymowych WP. The role of imaging in the diagnosis and management of axial spondyloarthritis. Nat Rev Rheumatol 2019;15:657-72. [Crossref] [PubMed]
26. Cerrahoglu L, Unlu Z, Can M, Goktan C, Celik P. Lumbar stiffness but not thoracic radiographic changes relate to alteration of lung function tests in ankylosing spondylitis. Clin Rheumatol 2002;21:275-9. [Crossref] [PubMed]
27. Shenkman Y, Qutteineh B, Joskowicz L, Szeskin A, Yusef A, Mayer A, Eshed I. Automatic detection and diagnosis of sacroiliitis in CT scans as incidental findings. Med Image Anal 2019;57:165-75. [Crossref] [PubMed]
28. Pascual E, Castellano JA, López E. Costovertebral joint changes in ankylosing spondylitis with thoracic pain. Br J Rheumatol 1992;31:413-5. [Crossref] [PubMed]
29. Sanzhang C, Rothschild BM. Zygapophyseal and costovertebral/costotransverse joints: an anatomic assessment of arthritis impact. Br J Rheumatol 1993;32:1066-71. [Crossref] [PubMed]
30. Saker E, Graham RA, Nicholas R, D'Antoni AV, Loukas M, Oskouian RJ, Tubbs RS. Ligaments of the Costovertebral Joints including Biomechanics, Innervations, and Clinical Applications: A Comprehensive Review with Application to Approaches to the Thoracic Spine. Cureus 2016;8:e874. [Crossref] [PubMed]
31. Erwin WM, Jackson PC, Homonko DA. Innervation of the human costovertebral joint: implications for clinical back pain syndromes. J Manipulative Physiol Ther 2000;23:395-403. [Crossref] [PubMed]
32. Jung JY, Kim MY, Hong YS, Park SH, Kang KY. Association between facet joint ankylosis and functional impairment in patients with radiographic axial spondyloarthritis. Semin Arthritis Rheum 2021;51:1005-10. [Crossref] [PubMed]
33. Stal R, Sepriano A, van Gaalen FA, Baraliakos X, van den Berg R, Reijnierse M, Braun J, Landewé RBM, van der Heijde D. Associations between syndesmophytes and facet joint ankylosis in radiographic axial spondyloarthritis patients on low-dose CT over 2 years. Rheumatology (Oxford) 2022;61:4722-30. [Crossref] [PubMed]
34. Ward MM, Reveille JD, Learch TJ, Davis JC Jr, Weisman MH. Occupational physical activities and long-term functional and radiographic outcomes in patients with ankylosing spondylitis. Arthritis Rheum 2008;59:822-32. [Crossref] [PubMed]
35. Tan S, Dasgupta A, Yao J, Flynn JA, Yao L, Ward MM. Spatial distribution of syndesmophytes along the vertebral rim in ankylosing spondylitis: preferential involvement of the posterolateral rim. Ann Rheum Dis 2016;75:1951-7. [Crossref] [PubMed]
36. Tan S, Yao J, Flynn JA, Yao L, Ward MM. Quantitative measurement of syndesmophyte volume and height in ankylosing spondylitis using CT. Ann Rheum Dis 2014;73:544-50. [Crossref] [PubMed]
37. Ward MM, Tan S. Better Quantification of Syndesmophyte Growth in Axial Spondyloarthritis. Curr Rheumatol Rep 2018;20:46. [Crossref] [PubMed]
38. Tan S, Yao J, Flynn JA, Yao L, Ward MM. Dynamics of syndesmophyte growth in AS as measured by quantitative CT: heterogeneity within and among vertebral disc spaces. Rheumatology (Oxford) 2015;54:972-80. [Crossref] [PubMed]
39. Song K, Zheng G, Zhang Y, Cui G, Zhang X, Mao K, Wang Y. Hilus pulmonis as the center of gravity for AS thoracolumbar kyphosis. Eur Spine J 2014;23:2743-50. [Crossref] [PubMed]
40. Song K, Zheng G, Zhang Y, Zhang X, Mao K, Wang Y. A new method for calculating the exact angle required for spinal osteotomy. Spine (Phila Pa 1976) 2013;38:E616-20. [Crossref] [PubMed]