Non‐Radiographic Axial Spondyloarthritis

Non-radiographic axial spondyloarthritis (nr-axSpA) is a distinct clinical entity characterized by the absence of sacroiliitis on plain radiographs, despite evidence of active sacroiliac inflammation on magnetic resonance imaging (MRI). Khan and colleagues first described this condition in 1985, referring to it as “spondylitic disease without radiologic evidence of sacroiliitis” 1. Although the concept of nr-axSpA had been recognized prior to this description, the 2009 Assessment of Spondyloarthritis International Society (ASAS) classification criteria brought the entity to the forefront of clinical practice and research. According to these criteria, individuals younger than 45 years with chronic low back pain lasting more than 3 months may be classified as having axial spondyloarthritis if they fulfill either the clinical arm, requiring at least two spondyloarthritis (SpA) features, or the imaging arm, requiring at least one SpA feature in conjunction with imaging evidence of sacroiliitis 2. The prevalence of nr-axSpA in the United States is estimated to be approximately 0.35%, compared with a combined axial SpA prevalence of 0.9%–1.4% 3, 4. In contrast, a survey-based study from Japan estimated the prevalence of radiographic axSpA (r-axSpA) to be 0.0026%, with an even lower prevalence of nr-axSpA at 0.0006% 5. Historically, axial SpA has been regarded as a predominantly male disease; however, more recent data suggest that the male-to-female ratio has narrowed, with several studies reporting near equivalence, particularly within nr-axSpA cohorts 6. Sex-based differences in clinical presentation have also been observed, with women more likely to report widespread pain, including cervical and upper thoracic symptoms that may not meet traditional definitions of inflammatory back pain 7. Women additionally tend to exhibit less radiographic damage, a factor that may contribute to the longer diagnostic delays consistently reported among female patients compared with males 7. Findings from large observational cohorts have further clarified the clinical characteristics of nr-axSpA. In the German Spondyloarthritis Inception Cohort (GESPIC), patients with early nr-axSpA demonstrated active inflammatory disease, with radiographic progression associated with HLA-B27 positivity, elevated C-reactive protein (CRP), and the presence of baseline MRI osteitis 8. The French Devenir des Spondylarthropathies Indifférenciées Récentes (DESIR) cohort showed that most patients with early inflammatory back pain remain non-radiographic over long-term follow-up, with women exhibiting greater pain burden and lower inflammatory markers than men 9. Data from the Corrona registry in the United States similarly demonstrated sex- and inflammation-related differences, identifying a higher proportion of women, lower CRP levels, and longer diagnostic delays among patients with nr-axSpA despite comparable disease activity to r-axSpA 10. The Korean Nonradiographic Axial Spondyloarthritis (KONASPA) registry extended these observations to an Asian population, demonstrating comparable functional limitations in nr-axSpA despite lower objective measures of inflammation 11. The clinical features of nr-axSpA have been less extensively characterized in isolation than those of r-axSpA; nevertheless, both conditions are regarded as part of a continuous disease spectrum and share substantial clinical overlap. The prevalence of inflammatory back pain (IBP) is similar between the two groups, reported in approximately 75%–82% of patients across cohorts, with no consistent differences in individual IBP features 12. Several studies have reported a higher prevalence of peripheral inflammatory arthritis in nr-axSpA compared with r-axSpA (35.7% vs. 17.0%, p = 0.001) 13, whereas pooled analyses demonstrate comparable overall prevalence between groups (27.9% vs. 29.7%) 14. Extra-articular manifestations also show substantial overlap between disease subsets. A meta-analysis of patients with axial SpA reported that acute anterior uveitis occurs more frequently in r-axSpA than in nr-axSpA, with pooled prevalences of approximately 23.0% and 15.9%, respectively 14. In contrast, most available data do not demonstrate consistent differences in the prevalence of enthesitis, dactylitis, or inflammatory bowel disease (IBD) between nr-axSpA and r-axSpA cohorts. Collectively, these differences have been summarized in Table 1. Insidious onset before age 40 years Morning stiffness > 30 min Improvement with exercise and NSAIDs No improvement with rest Nocturnal pain and alternating gluteal pain Similar prevalence in both groups (~75%–82%) 12 GESPIC: nearly universal (ever 100% versus 100%; past 6 months 97.8% versus 94.1%; Calin's criteria 84.6% versus 86.7%; NS) 8 DESIR: required for cohort entry (no between-group comparison) 9 KONASPA and CORRONA: high prevalence, not discriminatory 10, 11 IBP alone has limited specificity for distinguishing nr-axSpA from r-axSpA 12 Asymmetric inflammatory oligoarthritis Predominantly affects lower extremities Some clinical cohorts report higher prevalence in nr-axSpA (35.7% vs. 17.0%, p = 0.001) 13 Meta-analysis shows comparable pooled prevalence (27.9% vs. 29.7%) 14 GESPIC: 40.9% versus 37.4% (ever); 18.2% versus 14.4% (current); NS 8 DESIR: 23.0% versus 25.9% (p = 0.419) 9 KONASPA: 49.4% versus 54.8% (p = 0.567) 11 Inflammation at tendon or ligament insertion sites Associated with HLA-B27 positivity Axial enthesitis more frequently observed in r-axSpA 15 Peripheral enthesitis shows variable findings across cohorts DESIR: 60.1% versus 47.6% (p = 0.003) 9 CORRONA: 47.4% versus 29.0% (p < 0.001) 10 GESPIC: 43.6% versus 39.4% (ever); NS 8 KONASPA: 32.9% versus 16.7% (p = 0.070) 11 Diffuse swelling of an entire finger or toe Overall low prevalence in both groups Some reports suggest slightly higher frequency in nr-axSpA (12.9% vs. 0%) 15 GESPIC: 4.0% versus 6.3%; NS 8 DESIR: 14.6% versus 13.5% (p = 0.731) 9 CORRONA: 12.4% versus 9.0% (p = 0.34) 10 KONASPA: 5.5% versus 3.2% (p = 1.000) 11 Acute anterior uveitis Associated with HLA-B27 Cohort data show numerically higher rates in r-axSpA but no significant differences GESPIC: 12.4% versus 20.9%; NS 8 DESIR: 8.3% versus 12.0% (p = 0.148) 9 CORRONA: 18.6% versus 16.8% (p = 0.69) 10 KONASPA: 17.2% versus 19.4% (p = 0.769) 11 Crohn's disease or ulcerative colitis Some reports suggest higher prevalence in nr-axSpA 13 Overall low prevalence with no consistent differences across cohorts GESPIC: 1.8% versus 2.6%; NS 8 DESIR: 4.1% versus 7.6% (p = 0.070) 9 CORRONA: 6.2% versus 7.4% (p = 0.68) 10 KONASPA: 5.0% versus 3.2% (p = 1.000) 11 The diagnosis of nr-axSpA is based on the integration of appropriate clinical features and supportive imaging findings. Individuals with suspected axial SpA based on clinical presentation should undergo appropriate imaging evaluation. Although the 2009 ASAS criteria are classification rather than diagnostic criteria and no formal diagnostic criteria exist for nr-axSpA, MRI represents the cornerstone imaging modality in the evaluation of suspected nr-axSpA when interpreted within the appropriate clinical context. According to definitions established by the ASAS MRI Working Group, MRI studies performed for the assessment of axial SpA should include fat-suppressed sequences, such as short tau inversion recovery (STIR), which are essential for detecting active inflammatory lesions, including bone marrow edema (BME) or osteitis, characteristic of nr-axSpA 16. A positive MRI, as defined by ASAS, requires the presence of either one area of BME on at least two consecutive slices or at least two areas of BME on a single slice 16. Maksymowych and colleagues proposed alternative criteria for MRI positivity, defining active inflammation as BME present in four sacroiliac (SI) joint quadrants or on three consecutive SI joint slices. Structural lesions on MRI were defined by the presence of erosions in at least three SI joint quadrants or fat lesions in five SI joint quadrants 17. Importantly, not all MRI findings consistent with BME should be attributed to nr-axSpA. Several studies have demonstrated a high prevalence of MRI lesions meeting ASAS definitions in asymptomatic individuals, including approximately 23% of healthy military recruits, 30%–35% of recreational runners, 40% of professional hockey players, and up to 60% of postpartum women 18. Despite these limitations, MRI findings have demonstrated prognostic utility. One study reported a strong correlation between the severity and extent of osteitis on baseline MRI and the subsequent development of radiographic sacroiliitis 8 years later 19. Additionally, greater extent of osteitis on MRI has been associated with improved responses to tumor necrosis factor–alpha inhibitor (TNFi) therapy 20. Laboratory markers such as HLA-B27 status and CRP levels provide supportive information but lack diagnostic specificity for nr-axSpA. Approximately 6% of the general US population is HLA-B27 positive, and not all patients with nr-axSpA carry this allele 4. Elevated CRP levels are more frequently observed in patients with r-axSpA than in those with nr-axSpA, with one study demonstrating CRP levels greater than 5 mg/L in 45.4% of r-axSpA patients compared with 35.6% of nr-axSpA patients (relative risk 1.27; 95% confidence interval 1.16–1.38; p < 0.001) 9. On average, CRP levels are higher in r-axSpA than in nr-axSpA, although CRP reflects systemic rather than localized inflammatory burden 9. Notably, elevated CRP has been associated with an increased risk of progression to r-axSpA and a more robust response to TNFi therapy among patients with nr-axSpA 8, 9. Radiographic progression from nr-axSpA to r-axSpA has been reported in approximately 10%–40% of patients and typically occurs slowly over several years 8. Long-term data from the GESPIC and DESIR cohorts demonstrate that only a minority of patients with nr-axSpA progress to radiographic disease over 8–10 years, reinforcing that many individuals remain non-radiographic despite persistent symptoms 8, 9. Patients who are HLA-B27 positive and have elevated CRP levels appear to be at higher risk of radiographic progression 20. The impact of therapeutic interventions on structural progression in nr-axSpA remains incompletely defined. While some studies suggest that TNFi therapy may slow radiographic progression, other data demonstrate a lack of progression in many patients with nr-axSpA, raising the question of whether this phenotype represents a less structurally progressive subset of disease rather than a self-limiting condition 9, 20. Current evidence does not support withholding treatment solely based on the absence of radiographic changes. The 2019 American College of Rheumatology and Spondylitis Association of America guidelines provide treatment recommendations for axial SpA and do not distinguish between nr-axSpA and r-axSpA. Nonsteroidal anti-inflammatory drugs (NSAIDs) and physical therapy remain first-line therapies for all patients with nr-axSpA, with biologic agents reserved for those with persistent disease activity. Table 2 summarizes the biologic therapies currently approved for the treatment of nr-axSpA. Non-radiographic axial SpA is a clinically important subset within the axial SpA spectrum, characterized by the absence of definitive radiographic sacroiliitis. With an estimated prevalence of approximately 0.35% in the general population, nr-axSpA should be considered in patients with chronic back pain of onset before 45 years of age and features of inflammatory back pain. Although prevalence appears similar between men and women, women are more likely to present with nontraditional clinical features, such as widespread pain, which may contribute to diagnostic delay. There is substantial overlap between the clinical features of nr-axSpA and r-axSpA, and understanding their distribution can aid clinical recognition. Advances in MRI have improved detection of early inflammatory disease; however, nr-axSpA remains under-recognized due to limited awareness and the absence of formal diagnostic criteria. Large observational cohorts, including GESPIC, Corrona, DESIR, and KONASPA, highlight this diagnostic complexity by demonstrating heterogeneous presentations and meaningful disease burden across diverse populations. MRI remains the optimal imaging modality for evaluating early inflammatory lesions and supporting diagnosis, while physical therapy and NSAIDs continue to play central roles in management. Biologic therapies offer additional options for patients with inadequate response to conventional treatment. Alexander Alexandrov: conceptualization, supervision, writing – review and editing. Chen Chao: conceptualization, supervision, writing – review and editing. Anand Kumthekar: conceptualization, supervision, writing – review and editing. All authors have reviewed and approved the final manuscript. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.