The management of axial spondyloarthritis (axSpA) is a complex and rapidly changing field for Canadian rheumatologists. The Spondyloarthritis Research Consortium of Canada (SPARCC) and the Canadian Rheumatology Association (CRA) last updated treatment recommendations for the management of axSpA in 2014.1,2 The axSpA treatment landscape has changed significantly in the interim and now includes new classes of biologic disease-modifying antirheumatic drugs (bDMARDs) and targeted synthetic DMARDs with unique risks and benefits.

In the context of a swiftly evolving field, maintaining up-to-date treatment recommendations becomes increasingly challenging. To address this, CRA/SPARCC is using a living guideline model for these recommendations. In traditional guideline models, the entire set of recommendations is updated periodically, usually with several years passing in between updates. In contrast, a living guideline model allows for individual recommendations to be added or updated as needed, allowing the guidelines to stay relevant and current.3 To establish the living guidelines, a living systematic review is also concurrently generated.4 This model has already been successfully implemented by the CRA for the development of living guidelines in rheumatoid arthritis (RA), available online (https://app.magicapp.org/#/guideline/7413).5

The objective of these recommendations is to provide guidance for the management of axSpA in Canada. The need for this guideline was approved by the Guidelines Committee of the CRA.

METHODS

These treatment recommendations were developed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach.6 The GRADE approach provides an organized method for evaluating the certainty of evidence and grading the strength and direction of recommendations.6 The reporting of this manuscript follows the same structure as the CRA RA treatment recommendations.5,7

Organization and panel composition. The treatment recommendations panel was composed of rheumatologists and researchers, some of whom were SPARCC members; an allied health professional (physiotherapist); methodologists; a person with axSpA, who is also a primary care physician (PCP) and patient advocate; and a rheumatology trainee. Methodological support was provided by Cochrane Musculoskeletal for evidence synthesis. The group had 1 in-person, full-day meeting in 2022, and the rest of the meetings were conducted virtually.

Treatment recommendation funding and management of conflicts of interest. The development of the treatment recommendations was supported by in-kind funding from the CRA, a nonprofit association that represents Canadian rheumatologists. Declarations of potential conflicts of interest (COIs) were collected from all panelists using the International Committee of Medical Journal Editors form.

Development of treatment recommendations. The treatment recommendations were developed using a GRADE-ADOLOPMENT approach.8 GRADE-ADOLOPMENT is a structured approach for the effective adoption or adaptation of existing guidelines for the new development of recommendations. When using a GRADE-ADOLOPMENT approach, existing GRADE evidence tables are used as much as possible to bring efficiency to the guideline process. Evidence to Decision (EtD) profiles are then created to profile the evidence and rationale for the recommendation. EtD profiles are modified as needed to contextualize the recommendation to a different healthcare setting—in this case, Canadian. The treatment recommendations were largely adoloped from GRADE tables presented in the 2019 American College of Rheumatology (ACR)/Spondylitis Association of America/SpA Research and Treatment Network Recommendations for the Treatment of Ankylosing Spondylitis (AS) and Nonradiographic axSpA (nr-axSpA)9 and the Australian Living Guideline for the Pharmacologic Management of Inflammatory Arthritis (https://app.magicapp.org/#/guideline/7265). The literature was reviewed to update the GRADE tables (Supplementary Material, available with the online version of this article). Recommendations were evaluated in terms of specific PICO (patient/population – intervention – comparator/comparator – outcome). They are published on MAGICApp alongside the supporting evidence and clinical contextualization (https://app.magicapp.org/#/guideline/4891).

At each committee meeting, panelists reviewed the evidence profiles and modified EtD frameworks to achieve consensus judgment throughout the process. The direction (ie, to recommend or not) and strength of the final recommendation were determined if > 50% of the panel agreed. The development of a strong recommendation required 80% agreement.10

Addressing health equity in the treatment recommendations. Health equity was explicitly considered for each treatment recommendation by applying a PROGRESS-Plus framework (https://methods.cochrane.org/equity/projects/evidence-equity/progress-plus). PROGRESS-Plus is an acronym that can be used to identify attributes that stratify an individual’s health opportunities and outcomes. The PROGRESS acronym stands for place of residence, race/ethnicity/culture/language, occupation, gender/sex, religion, education, socioeconomic status, and social capital, and Plus includes personal characteristics associated with discrimination (eg, age, disability), relationship features (eg, smoking parents, exclusion from school), and time-dependent relationships (eg, leaving the hospital, respite care). Existing literature was used to address each of the PROGRESS-Plus attributes; if none was found, expert opinion was used.

How to read these treatment recommendations. When using a GRADE approach, recommendations are classified as strong or conditional.6 If a recommendation is classified as strong, it means that all or almost all people with axSpA would choose that intervention. If a recommendation is conditional, it means that the majority of those with axSpA would choose that intervention, but many would not (Table).

Table

Interpretation of strong and conditional recommendations.5

How to use these treatment recommendations. These treatment recommendations are not meant to replace a clinician’s clinical judgment. Rather, they aim to help rheumatologists, healthcare professionals, and patients with axSpA to make decisions regarding treatment. As these are living guidelines, they are expected to change over time and will be updated as new evidence emerges. As such, the CRA website should always be consulted for the most up-to-date version of the treatment recommendations (https://app.magicapp.org/#/guideline/4891).

Public commenting. The public may comment on the treatment recommendations through the CRA website. Public comments will be reviewed regularly and considered by the panel for future updates of the treatment recommendations.

Living treatment recommendations. As these treatment recommendations are living, they will be updated over time. As new data and evidence emerge, new recommendations will be added, and existing recommendations may be changed. As such, users of the recommendations should make sure to consult the latest available version of the recommendations available online. Supplemental journal articles may be periodically published with major modifications to the treatment recommendations to aid with dissemination of the online living treatment recommendations and assist in knowledge translation.

Structure of treatment recommendations. These axSpA treatment recommendations are divided into 2 sections: general management principles and treatment (nonpharmacologic and pharmacologic). General management principles were reached by consensus of the treatment recommendation panel.

General management principles.

Classification of SpA. Management recommendations for SpA are organized under the categories of axSpA and peripheral SpA.11 This set of recommendations addresses axSpA specifically; CRA/SPARCC will be presenting management recommendations for peripheral SpA separately in a future publication.

AxSpA can be further classified as radiographic and nr-axSpA. Since the literature suggests that treatment works similarly for both, the panel has considered them together under the umbrella as axSpA.

Goals of treatment. The goal of treatment is remission or inactive disease. When remission or inactive disease is not possible, the goal is low or minimal disease activity and control of symptoms, prevention of damage, and improvement in quality of life (QOL). Therapy should be adjusted until these goals are reached. Treatment goals must be based on shared decision making between the patient, the rheumatologist, and other care providers. The following recommendations are subdivided based on whether the patient has active or stable disease. By stable disease, we refer to a state of controlled disease activity, in which the axSpA is in a state of remission. Active disease vs remission is usually defined using standardized disease activity scores including the Bath AS Disease Activity Index (BASDAI), Assessment of SpA international Society (ASAS), and AS Disease Activity Score (ASDAS). There is currently no well-defined treat-to-target definition for disease control in axSpA, although many have been suggested (ie, BASDAI < 4, ASDAS < 2.1). The panel suggests that clinicians keep a target in mind when deciding if their patient is in a state of remission, but currently leaves this to the clinician’s judgment. As more data emerge in the field of treating to target in axSpA, the living guidelines will be updated in this area.

Optimal management. Optimal management of SpA includes a combination of nonpharmacological and pharmacological treatments, as well as patient education. The following recommendations address these aspects of management for axSpA. It should be noted that the panel felt that no formal recommendations were necessary for treatment of axSpA with conventional DMARDs (eg, methotrexate [MTX] or sulfasalazine [SSZ]), since their inefficacy in axial disease is well established. For patients with axSpA and peripheral joint involvement, CRA/SPARCC will be publishing treatment recommendations for the management of peripheral SpA at a future date.

Shared decision making. Patient preferences, including balancing risks and benefits, must be incorporated into regulatory decision making and the prescribing of arthritis medications.

Biosimilars. It is appropriate to consider pharmacoeconomic data in formulating decisions on management strategies. The aim is to identify subgroups of patients with the highest burden of disease for whom the additional benefits merit the additional costs. Biosimilars were recently introduced into the Canadian market, and provincial and territorial policymakers adopted different approaches regarding drug reimbursement. The inclusion of the individual biosimilars currently available in Canada is beyond the scope of these guidelines, and for the purpose of the following recommendations, biosimilars were the same as their biologic originator with regard to efficacy and safety.

Postmarketing evaluation of therapeutics. Postmarketing evaluation of newer therapies for axSpA should be implemented to ensure both appropriate access and use of these agents and their safety in an unselected population with longer periods of observation.

Ethical considerations.

Duties of a formulary committee. A formulary committee has a duty to represent the public’s interests in promoting the greatest health benefits possible (the ethical principle of beneficence) as fairly as possible within society’s limited shared resources (the ethical principle of justice) through an open and transparent process and in accordance with the best available evidence (the ethical principle of accountability).

Economic evaluations. Economic evaluations should be comprehensive with a clear analysis of the direct and indirect costs of suboptimal treatment. Ethically, suboptimal treatment is always questionable (the ethical principle of nonmaleficence).

Practice of evidence-based medicine. Healthcare providers (HCPs) must be enabled to practice the highest standard of evidence-based medicine for the benefit of their patients; thus, none of the recommended therapies should be excluded based on access or cost. It is ethically unreasonable to deny an effective therapy because of limited access (eg, allied health professionals, including physiotherapy and occupational therapy). Even expensive therapies that are clinically effective must not be excluded on principle.

Formulary committees should be encouraged to work in conjunction with clinical specialists to develop guidelines for access that promote safe and effective interventions at lower cost where possible, but that also allow clinicians and patients to access necessary therapeutics when other options are not medically appropriate.

Universality and transferability. In a Canadian context, the delivery of health care is a provincial rather than federal responsibility. However, the principles of universality, transferability, and comprehensiveness enshrined in the Canada Health Act, as well as the underlying ethical principle of justice, mandate that treatments approved in 1 province should generally be available to patients in all provinces.

Target groups for treatment recommendations. These management recommendations are intended for rheumatologists; PCPs, internists, pediatricians, and other HCPs; persons diagnosed with axSpA; insurance payers; government agencies; and formularies.

Wait time recommendations.

Consultation wait times. Ideally, patients with chronic back pain (duration ≥ 3 months) with an age of onset prior to 45 should be screened for the presence of axSpA and assessed by an HCP with expertise in SpA within 3 months of referral.

Magnetic resonance imaging wait times. Magnetic resonance imaging (MRI) frequently plays an important role in the diagnosis or assessment of disease activity in SpA. When a rheumatologist orders an MRI for these indications, it should occur within 6 weeks.

Disease monitoring. Specific disease monitoring of patients with SpA in clinical practice should ideally include the following:

Patient history.

Relevant clinical exam (axial or peripheral). For axSpA, spinal mobility should be assessed. All patients should have an assessment of tender joints, swollen joints, enthesitis, and dactylitis.

Baseline screening for hepatitis B virus and other chronic infections, liver disease, renal disease, and malignancy.

Assessment for signs and symptoms of extramusculoskeletal manifestations of SpA (in particular, inflammatory bowel disease [IBD], uveitis, and psoriasis [PsO]).

Assessment of disease activity.

Assessment of function.

Patient global assessment of well-being.

Acute-phase reactants (C-reactive protein [CRP] and/or erythrocyte sedimentation rate [ESR]). HLA-B27 status may be used for diagnosis but testing should not be repeated.

Drug toxicity (including infection and malignancy) and adherence.

Appropriate imaging, including plain radiographs and, when indicated, MRI of the sacroiliac joints, spine, and involved peripheral joints.

QOL assessment.

Participation in activities and work disability.

Frequency of disease monitoring will depend on disease severity, treatment type, and patient preference.

Monitoring and management of extramusculoskeletal manifestations of SpA (ie, IBD, uveitis, PsO) should be in collaboration with respective specialists as needed.

Monitoring and management of comorbid conditions associated with inflammatory arthritis (ie, cardiovascular disease (CVD), hypertension, hyperlipidemia, diabetes mellitus, osteoporosis) and immunization records should be in collaboration with PCPs and respective specialists as needed.

If treatment is ineffective, reassess the diagnosis.

Equity. Health inequities arise because of social, economic, geographic, and political circumstances contributing to unfair and avoidable health outcomes. Addressing health inequities is a clinical and public health priority and an integral part of GRADE’s EtD framework. In the process of developing guidelines, recommendations must ensure improvement, not worsening, of health inequities.10

The 6 priority populations identified for the Canadian RA guidelines were rural and remote, Indigenous, elderly and frail, first-generation immigrant and refugee, low-income and vulnerably housed, and diverse sex and gender populations. The obese population was an additional priority group identified by the axSpA guideline committee.

In axSpA, obesity is associated with higher disease activity and worse physical function and QOL. The odds of responding to a tumor necrosis factor inhibitor (TNFi) is 30% vs 70% in the general population.12 Patients with obesity have more severe disease, a poorer response to therapy, and multiple comorbidities that limit access to care. Weight loss is associated with decreased disease activity and improved response to medication.

The First Nations peoples of Haida Gwaii, Bella Bella, and Bella Coola on the west coast of British Columbia have an increased prevalence of HLA-B27 and sacroiliitis. Studies were conducted in the 1960s and blood was taken from male individuals aged ≥ 20 years (younger boys were in residential schools) in Bella Bella, aged ≥ 15 years in Bella Coola, and aged ≥ 15 years in Haida Gwaii. The prevalence of HLA-B27 positivity was 37%, 40%, and 50%, respectively. In the Haida, the prevalence of sacroiliitis was 9.5%, which was significantly greater than the White population.13,14 Access to care in remote communities is a challenge; however, the effects of colonization and the breech of trust must be reconciled, guided by the recommendation of the Truth and Reconciliation Commission.

Barnabe et al developed an EtD framework to examine potential domains of inequity for the identified priority populations.10 These domains included the priority of the problem, equity considerations for the health benefits of treatment and adverse effects, variations in the importance of the main outcomes facing inequity, resource considerations, treatment acceptability, and the feasibility of the intervention. This framework is applicable and recommendations can be adapted to axSpA.

Under the domains health benefits to treatment and adverse effects, there are gender considerations in treatment in axSpA. Women have a higher age of disease onset, shorter duration of disease, and lower CRP, all of which are predictors for a poorer response to TNFi therapy.15,16 In particular, women with nr-axSpA have been shown to have poorer response to therapy.15,16

A health equity lens has been adopted from the work of the CRA RA guidelines development to inform these treatment guidelines, and the additional considerations for axSpA as discussed above need to be considered.

RESULTS

Nonpharmacologic and pharmacologic treatment of axSpA. These recommendations are divided into 5 sections: (1) active axSpA, (2) stable axSpA, (3) active or stable axSpA, (4) comorbidities, and (5) assessment, screening, and imaging. For a summary of these treatment recommendations, please refer to Figure 1 to Figure 5.

Figure 1.

Active axSpA: first-line treatment. axSpA: axial spondyloarthritis; GC: glucocorticoid; IBD: inflammatory bowel disease; NSAID: nonsteroidal antiinflammatory drug.

Figure 2.

Active axSpA: second-line treatment. a Currently, only upadacitinib and tofacitinib are approved for the treatment of axSpA in Canada. b Currently, JAKi are only available for the treatment of axSpA after the failure of another bDMARD. axSpA: axial spondyloarthritis; bDMARD: biologic disease-modifying antirheumatic drug; IBD: inflammatory bowel disease; IL-17i: interleukin 17 inhibitor; JAKi: Janus kinase inhibitor; MSK: musculoskeletal; MTX: methotrexate; NSAID: nonsteroidal antiinflammatory drug; TNFi: tumor necrosis factor inhibitor.

Figure 3.

Active axSpA: third-line treatment. AS: ankylosing spondylitis; axSpA: axial SpA; DMARD: disease-modifying antirheumatic drug; IL-17i: interleukin 17 inhibitor; JAKi: Janus kinase inhibitor; PICO: patient/population – intervention – comparator/comparator – outcome; TNFi: tumor necrosis factor inhibitor; SpA: spondyloarthritis.

Figure 4.

Stable axSpA. axSpA: axial spondyloarthritis; bDMARD: biologic disease-modifying antirheumatic drug; csDMARD: conventional synthetic DMARD; NSAID: nonsteroidal antiinflammatory drug; TNFi: tumor necrosis factor inhibitor.

Figure 5.

(A) General recommendations for comorbidities, uveitis, and IBD. (B) General recommendations for imaging, screening, assessment, and surgery. ASDAS: Ankylosing Spondylitis Disease Activity Score; axSpA: axial spondyloarthritis; bDMARD: biologic disease-modifying antirheumatic drug; CRP: C-reactive protein; DXA: dual-energy x-ray absorptiometry; ESR: erythrocyte sedimentation rate; ETN: etanercept; IBD: inflammatory bowel disease; JAKi: Janus kinase inhibitor; MRI: magnetic resonance imaging; NSAID: nonsteroidal antiinflammatory drug; TNFi: tumor necrosis factor inhibitor.

Active axSpA

Recommendation 1. In adults with active axSpA, we strongly recommend treatment with NSAIDs over no treatment with NSAIDs. (Strong recommendation for.)

Recommendation 2. In adults with active axSpA, we conditionally recommend continuous treatment with NSAIDs over on-demand treatment with NSAIDs. (Weak recommendation for.)

Given the uncertainty regarding potential disease-modifying effects, the CRA panel unanimously agreed on the recommendation favoring continuous use of NSAIDs in patients with active axSpA. The decision to use NSAIDs continuously may vary depending on the severity of symptoms, side effects, patient preferences, and comorbidities, particularly gastrointestinal and kidney comorbidities and cardiovascular disease.

Recommendation 3. In adults with active axSpA, we conditionally recommend using any available NSAID over the use of a particular NSAID as a preferred choice. (Weak recommendation for.)

Recommendation 4. In adults with active axSpA, we strongly recommend against treatment with systemic glucocorticoids (GCs). (Strong recommendation against.)

Despite the very low level of evidence, the panel made a strong recommendation against GC therapies given the known harms and lack of evidence of benefit.

Recommendation 5. In patients with active axSpA despite the use of NSAIDs, we strongly recommend treatment with TNFi over no treatment with TNFi. (Strong recommendation for.)

Recommendation 6. For axSpA, we conditionally recommend any available TNFi for musculoskeletal manifestations. The limited direct evidence does not suggest the superiority of 1 TNFi over another. However, the panel emphasized the importance of considering axSpA-associated conditions such as uveitis, PsO, and IBD— as this should influence the choice of TNFi—as well as patient preference for frequency and dosing. (Weak recommendation for.)

Recommendation 7. In adults with active axSpA despite NSAIDs, we strongly recommend treatment with interleukin 17 inhibitors (IL-17i) over no treatment. (Strong recommendation for.)

There is high certainty that IL-17i are efficacious in axSpA and have a reassuring safety profile. Additional consideration may be given to those with significant concomitant PsO as these agents are highly efficacious for skin PsO.

Recommendation 8. We conditionally do not recommend treatment with an IL-17i over treatment with a TNFi or vice versa. Randomized controlled trials (RCTs) with biologic-naïve patients with active axSpA compared ixekizumab with adalimumab (ADA) as comparator and demonstrated similar responses. However, these were not powered for the comparison with the TNFi, and were compared only against placebo. Preliminary results from a large head-to-head study suggested that secukinumab is not superior to ADA to improve radiographic outcomes.17 Clinical outcome results were measured but not yet available. (Weak recommendation against.)

Recommendation 9. In adults with active axSpA despite treatment with NSAIDs, we conditionally recommend treatment with Janus kinase inhibitors (JAKi) over no treatment with JAKi. (Weak recommendation for.)

Currently, only upadacitinib and tofacitinib (TOF) are approved for use in axSpA in Canada. JAKi are currently approved in Canada for use only after the failure of another bDMARD. Although there is moderate-certainty evidence suggesting improvement in disease outcomes, there are several populations for whom the panel cautioned against use of JAKi given the US Food and Drug Administration (FDA)-mandated phase IV safety study data and associated black box warning regarding major adverse cardiovascular events (MACE), malignancy, and death.18 We caution the use of JAKi in patients with a history of CVD, malignancy, or blood clots.

Recommendation 10. In adults with active axSpA despite treatment with NSAIDs, we conditionally recommend against treatment with a JAKi over treatment with TNFi. (Weak recommendation against.)

There is no direct evidence suggesting efficacy of 1 agent over another; however, head-to-head trials in other patient populations demonstrate increased risks with TOF.19 There is more experience with the use of TNFi in active axSpA, and recent evidence suggests increased risks of MACE, malignancy, and death with JAKi use compared to TNFi.19 However, the panel notes that most data concerning hazard risk derive from an RA population, enriched for CVD risk factors, and not axSpA.

Recommendation 11. In adults with active axSpA despite treatment with NSAIDs, we conditionally recommend against treatment with JAKi over treatment with IL-17i. (Weak recommendation against.)

Based on the limited and indirect evidence, IL-17i and JAKi both are efficacious in axSpA; however, IL-17i appear to be safer, based on recent evidence suggesting increased risks of MACE, malignancy, and death with JAKi. However, if the patient had concomitant IBD or suspected IBD, use of a JAKi would likely be preferred over an IL-17i.

Recommendation 12. In adults with active axSpA despite NSAIDs and with contraindications to TNFi, we conditionally recommend that either IL-17i or JAKi may be used. (Weak recommendation for.)

As mentioned in recommendation 9, IL-17i appear safer, based on recent data suggesting increased risks of MACE, malignancy, and death with JAKi use. In certain clinical situations, such as the presence of IBD, the use of a JAKi would probably be preferred to that of an IL-17i.

Recommendation 13. In adults with active axSpA despite NSAIDs and with contraindications to TNFi, we recommend against IL-6 inhibitors and apremilast. (Weak recommendation against.)

Recommendation 14. In adults with active axSpA despite treatment with a TNFi, we conditionally recommend switching to a different TNFi rather than adding a conventional synthetic DMARD (csDMARD). (Weak recommendation for.)

However, in people who have failed multiple TNFi, the panel felt that the addition of MTX might be considered, despite the lack of supportive evidence for the theoretical benefit of reducing antidrug antibodies.

Recommendation 15. In adults with active axSpA despite treatment with a first TNFi, we conditionally recommend treatment with IL-17i over treatment with a different TNFi in patients with primary nonresponse to TNFi. We conditionally recommend treatment with a different TNFi in patients with secondary nonresponse to TNFi. (Weak recommendation for.)

Recommendation 16. In patients with active axSpA despite treatment with TNFi, we conditionally recommend the use of JAKi over no treatment. (Weak recommendation for.)

Although there is moderate-certainty evidence suggesting improvement in disease outcomes, there are several populations for whom the panel cautioned against using JAKi, given the FDA-mandated phase IV safety study data and associated black box warning MACE, malignancy, and death. We caution using JAKi in patients with a history of CVD, malignancy, or blood clots.

Recommendation 17. In adults with isolated active sacroiliitis despite treatment with NSAIDs, we conditionally recommend treatment with locally administered parenteral GCs over no treatment with local GCs. (Weak recommendation for.)

Although the evidence had serious risk of bias and imprecision, local GC injection of the sacroiliac joint may be considered both diagnostically and therapeutically in cases where the diagnosis is uncertain or when there is a contraindication to systemic therapy. Additionally, this may be a useful option when weighed against the alternative of escalating systemic treatment due to a flare. This may be limited by availability of clinicians with expertise in computed tomography, MRI, ultrasound, or fluoroscopically guided injections.

Recommendation 18. In adults with axSpA with stable axial disease and active enthesitis despite treatment with NSAIDs, we conditionally recommend against using treatment with locally administered parenteral GCs over no treatment with local GCs. (Weak recommendation against.)

This may be considered for lateral and medial epicondyles, greater trochanter, and pelvic rim, but should be avoided for Achilles, patellar, and quadriceps tendons. Caution should be considered for plantar fascia injections.

Treatment with locally administered GCs might be considered only when other therapies have failed or are contraindicated.

Recommendation 19. In adults with axSpA with stable axial disease and active peripheral arthritis despite treatment with NSAIDs, we conditionally recommend treatment with intraarticular GCs over no treatment with intraarticular GCs. (Weak recommendation for.)

Recommendation 20. In adults with active axSpA, we strongly recommend physical therapy to improve health and functional status. Given the evidence of benefit and the very low likelihood of harm, the panel judged that a strong recommendation was justified. The panel decided not to express a preference between aquatic or land-based exercises. (Strong recommendation for.)

Recommendation 21. The panel conditionally recommends active physical therapy interventions over passive interventions. (Weak recommendation for.)

The ACR 2019 guidelines did not identify studies that compared active with passive interventions. One of the essential goals of physical therapy is to educate patients in self-management using an independent exercise program. There is a body of literature supporting exercise as a cornerstone of management in people with axSpA, with little investigation into the effect of passive modalities on axSpA outcomes. Therefore, active interventions are emphasized over passive interventions. Passive interventions could supplement, but not substitute for, active physical therapy interventions.

Stable axSpA

Recommendation 22. In adults with stable axSpA, the use of regular NSAIDs is conditionally not recommended. (Weak recommendation against.)

The panel felt that the use of regular NSAIDs in stable axSpA had an unfavorable risk-benefit ratio. The role of NSAIDs in treatment of axSpA is to control symptoms, and its use can be continued on that basis if effective. There have been no studies showing the effect of continuous NSAIDs on radiographic progression in stable axSpA.

Recommendation 23. In adults with stable axSpA taking TNFi, regular NSAIDs should not be continued. (Weak recommendation against.)

The panel felt that continuing regular NSAIDs in patients with well-controlled axSpA who were taking TNFi had an unfavorable risk-benefit ratio. There is a lack of evidence to support any additional advantage of adding NSAIDs to well-controlled axSpA for clinical outcomes, and recent evidence suggests the lack of a disease-modifying role for NSAIDs.

Recommendation 24. In adults with stable axSpA on treatment with a TNFi and csDMARDs, the csDMARDs should not be continued. (Weak recommendation against.)

The panel concluded that there is no benefit to continuing csDMARDs in addition to TNFi in patients with stable axSpA. There are no studies addressing this PICO question. Studies demonstrate that TNFi persistence is not enhanced by csDMARDs, in contrast to RA and psoriatic arthritis.20

csDMARDs that may be given to patients already receiving TNFi may include MTX and SSZ. These are generally added to control peripheral symptoms or extramusculoskeletal manifestations such as uveitis, PsO, or IBD. If these manifestations are well controlled with TNFi, the panel felt that continuing csDMARDs in patients with well-controlled axSpA receiving TNFi had an unfavorable risk-benefit ratio.

Multiple retrospective studies have examined the benefits of continuing csDMARDs along with TNFi. The results are not consistent, but some studies show better long-term persistence. As a result, for patients that have failed multiple biologics, some clinicians might consider continuing csDMARDs.

Recommendation 25. In adults with stable axSpA receiving treatment with a biologic, the panel strongly recommends against stopping biologic medications abruptly. (Strong recommendation against.)

Multiple RCTs have looked at this PICO, and all the studies have identified a significantly increased risk of disease flare upon stopping the biologic medication.

Recommendation 26. In adults with stable axSpA who have been in a sustained state of low disease activity or remission for at least 6 months, consideration could be given to reducing the dose or lengthening the interval of administration of a bDMARD. (Weak recommendation for, with some situational exemptions.)

This dose reduction should be performed in a stepwise fashion and in a context of shared decision making between the patient and the physician, provided that the patient has rapid access to rheumatologic care in the case of disease worsening or flare. The taper can be continued if the treatment target is maintained. Abrupt cessation of bDMARDs is not recommended. Patients should be informed of the risks of not being able to return to their baseline state upon reinitiation of their initial dosing.5,7

In the case of patients who do not have rapid access to rheumatologic care, or in cases where reaccessing bDMARDs may be challenging, we conditionally recommend against a dose reduction or discontinuation of bDMARDs.

Recommendations on dose reduction are based on TNFi. It is unknown if this recommendation can be extrapolated to non-TNFi bDMARDs.

Recommendation 27. In adults with stable axSpA, physical therapy is strongly recommended over no physical therapy. (Strong recommendation for.)

Given the evidence of benefits on other health outcomes beyond axSpA (eg, cardiovascular, depression) and low likelihood of harm, the panel judged that a strong recommendation was justified. Physical therapy in stable patients was most important for periodic reassessment and appropriate modifications of home exercises.

Active or stable axSpA

Recommendation 28. In adults with either active or stable axSpA on treatment with TNFi, we conditionally recommend against co-treatment with low-dose MTX. (Weak recommendation against.)

In RA, the likelihood of TNFi discontinuation is lower among patients who receive co-treatment with MTX, perhaps by reducing the development of antidrug antibodies, or by the withdrawal of an effective agent, which is not the case in axSpA. In axSpA, it is less clear whether the duration of TNFi use, and by inference their effectiveness, is similarly prolonged.

Recommendation 29. In adults with active or stable axSpA, we recommended unsupervised back exercise as part of a general endorsement of physical activity, given that desirable consequences were likely to outweigh undesirable consequences. The recommendation is conditional in that unsupervised back exercises should not substitute for initial instruction in back exercises by a physical therapist. (Weak recommendation for.)

Recommendation 30. In adults with active or stable axSpA, we recommend fall evaluation and counseling for patients with osteoporosis, significant spinal fusion, unstable posture, or deficits in balance. (Weak recommendation for.)

It should be noted that falls in patients with axSpA may lead to spinal fractures as well as significant neurological sequelae, highlighting the importance of fall prevention.

Recommendation 31. In adults with active or stable axSpA, we strongly recommend group or individual self-management education over no formal self-management education. (Strong recommendation for.)

Recommendation 32. In adults with active or stable axSpA, we strongly recommend against spinal manipulation. This recommendation is derived from evidence from radiographic axSpA. (Strong recommendation against.)

Recommendation 33. In adults with active or stable axSpA and advanced hip disease, we strongly recommend treatment with total hip arthroplasty over no surgery. (Strong recommendation for.)

Recommendation 34. In adults with active or stable axSpA and severe kyphosis, we conditionally recommend against elective spinal osteotomy. (Weak recommendation against.)

In most patients, the risks of spinal osteotomy outweigh the benefits. Consideration for elective spinal osteotomy may be given for those patients who have limited horizontal vision or those with major physical deficits, or whose kyphosis is causing major psychological harm. Of note, this recommendation does not apply to nonelective surgery, such as stabilization of spinal fractures or release of nerve root compression.

Comorbidities

Although this was not addressed by any specific PICO, the panel would like to emphasize the importance of close co-management of comorbidities with relevant subspecialists, such as gastroenterologists, dermatologists, and ophthalmologists.

Recommendation 35. In adult patients with axSpA with recurrent attacks of uveitis, ophthalmology monitoring and treatment is strongly recommended over no ophthalmologist care to help reduce the risk of ocular complications, including but not limited to vision loss. (Strong recommendation for.)

It is felt that ophthalmology monitoring and treatment are likely to result in decreased severity, duration, and potential ocular complications of anterior uveitis.

Recommendation 36. In adult patients with axSpA with recurrent attacks of noninfectious anterior uveitis, a prescription for topical steroids for prompt at-home use is recommended over no at-home use in order to decrease the severity and/or duration of anterior uveitis episodes. This may also be beneficial in preventing ocular complications. (Weak recommendation for.)

Patients should be cautioned that in the event of persisting or worsening symptoms despite the use of topical corticosteroids, they should seek advice from their ophthalmologist to rule out other causes of ocular inflammation, including infections.

Recommendation 37. In adult patients with axSpA with recurrent attacks of uveitis, TNFi—particularly the monoclonal TNFi—should be used over the other biologics. (Weak recommendation for.)

Monoclonal antibodies (particularly ADA and infliximab [IFX]) have demonstrated superiority over soluble TNF receptors (etanercept [ETN]). Certolizumab pegol and golimumab have shown encouraging results, albeit with less robust data than ADA and IFX. IL-17 antagonists appear to be associated with an increased risk of anterior uveitis in comparison with monoclonal TNFi.

Recommendation 38. In adults with axSpA who develop uveitis while being treated with TNFi receptor antagonist, we strongly recommend switching the TNFi from a soluble receptor antagonist (ie, ETN) to a monoclonal antibody to help treat and reduce recurrences of uveitis. (Strong recommendation for.)

Recommendation 39. In adults with axSpA who are already being treated with a monoclonal antibody, we conditionally recommend switching the TNFi to an alternative monoclonal antibody in the event of recurrent or refractory uveitis. Consideration may also be given to adding in a csDMARD (eg, MTX) if appropriate. (Weak recommendation for.)

Recommendation 40. In adults with axSpA and IBD, regular NSAID use is typically not recommended due to its potential to worsen IBD symptoms. However, the panel felt that for patients whose IBD is in remission, NSAIDs may be used cautiously for up to 2 weeks to help control arthritic symptoms in a shared decision-making approach with gastroenterology. (Weak recommendation for.)

Recommendation 41. In adults with axSpA and IBD, we conditionally recommend using celecoxib over other NSAIDs (ie, nonselective cyclooxygenase inhibitors) in a shared decision-making approach with gastroenterology. (Weak recommendation for.)

Recommendation 42. In adults with axSpA and IBD, we conditionally recommend the use of a monoclonal TNFi or JAKi over other biologic therapies to improve outcomes, given existing data that show the efficacy of monoclonal antibodies against TNF and JAKi in both diseases, the lack of efficacy of soluble TNF receptor and IL-17i in IBD, and the lack of efficacy of IL-23i in axSpA. (Weak recommendation for.)

Assessment, screening, and imaging

Recommendation 43. We conditionally recommend the regular-interval use and monitoring of a validated axSpA disease activity measure. The choice of the disease activity measure is left to the clinician’s discretion. (Weak recommendation for.)

Recommendation 44. We conditionally recommend the regular-interval use and monitoring of CRP and/or ESR over usual care without CRP or ESR monitoring. (Weak recommendation for.)

Recommendation 45. We conditionally recommend against using the treat-to-target strategy, which aims for a target of ASDAS < 1.3 for remission (or < 2.1 for low disease activity), over a treatment strategy based on physician assessment. (Weak recommendation against, adopted/adapted from 2019 ACR guidelines.)

Although ASDAS has been shown to be superior to the BASDAI in the assessment of disease activity in axSpA, not requiring CRP to be tested at the time of the visit makes BASDAI a more feasible tool. Treatment goals should be agreed on through shared decision making between the patient and rheumatologist.

Recommendation 46. In adults with axSpA, we conditionally recommend screening for osteopenia and osteoporosis over no screening. (Weak recommendation for.)

No studies directly address this PICO. The panel was mindful of the fact that patients with advanced axSpA are at higher risk of both osteoporosis and falls. The panel felt that general osteoporosis guidelines should be followed.

Recommendation 47. In adults with axSpA and syndesmophytes or spinal fusion, we conditionally recommend screening for osteopenia and osteoporosis with a dual-energy x-ray absorptiometry (DXA) scan that includes the spine as well as the hip and femoral neck, as opposed to DXA scanning solely of the spine, or solely of the hip or other nonspine sites. (Weak recommendation for.)

Recommendation 48. In adults with axSpA, we strongly recommend against routine screening for cardiac conduction defects with electrocardiograms in asymptomatic patients. (Strong recommendation against.)

This question has not been directly addressed by any study. The panel felt that the risks and costs of regular screening for conduction defects outweighed the benefits due to low incidence of conduction defects. Regular cardiac examination should be performed at each patient visit.

Recommendation 49. In adults with axSpA, we strongly recommend against screening for valvular heart disease with echocardiogram in asymptomatic patients. (Strong recommendation against.)

This question has not been directly addressed by any study. The panel felt that the risks and costs of regular screening for conduction defects outweighed the benefits due to low incidence of conduction defects. Regular cardiac examination should be performed at each patient visit.

Recommendation 50. In adults with axSpA of unclear activity who are receiving a biologic, we conditionally recommend obtaining a spinal or pelvis MRI to assess activity. (Weak recommendation for.)

Since physical and laboratory measures are often normal despite active axSpA, and because symptoms may be nonspecific, it may be difficult to know whether a patient is experiencing inflammation that warrants a change in treatment. MRI has been shown to have a predictive value for treatment response to TNFi therapies, and patients with axSpA who have bone marrow edema tend to have higher responses. Therefore, some patients may benefit from having an MRI to understand whether symptoms are due to inflammation. MRI is not recommended for patients in whom disease activity is either clearly clinically active or clinically stable, or when the results of an MRI would not be expected to change treatment.

Recommendation 51. In adults with stable axSpA, we conditionally recommend against obtaining MRIs of the pelvis or spine to confirm inactive disease. (Weak recommendation against.)

Because the clinical assessment of inflammation in axSpA has many limitations, questions may arise on whether subclinical inflammation that could be detected by MRI is being missed by either the physical examination, symptoms, or laboratory studies. Given the lack of evidence demonstrating that obtaining an MRI in stable patients improves clinical outcomes, the only moderate sensitivity and specificity of MRI-defined abnormalities for measurement of activity in axSpA, the burden of testing, and the concern for possible overtreatment, the panel recommended against obtaining an MRI in this situation. MRI could be considered in circumstances where the clinician and patient differ in their assessment of whether the disease is stable. (Weak recommendation against.)

Recommendation 52. In adults with active or stable axSpA receiving any treatment, we conditionally recommend against obtaining repeat spine radiographs at scheduled intervals as a standard approach. (Weak recommendation against.)

However, structured spine radiographs at preset intervals can be informative in certain settings and can reveal other relevant radiographic features (eg, hip osteoarthritis or degenerative disc disease [DDD]). Obtaining repeat spine radiographs may provide additional information when the clinical features of disease activity are unclear, or in cases deemed appropriate through shared decision making between the patient and clinician.

Recommendation 53. In adult patients with axSpA with clinical enthesitis, we conditionally recommend obtaining an entheseal ultrasound to confirm enthesitis rather than not obtaining one if the ultrasound finding will inform the decision of systemic treatment, such as starting or changing a bDMARD. (Weak recommendation for.)

Recommendation 54. We recommend not performing entheseal ultrasound to detect subclinical enthesitis, rather than performing entheseal ultrasound in adult patients with axSpA without clinical enthesitis. (Weak recommendation against.)

Recommendation 55. We recommend not obtaining routine spine MRI in addition to sacroiliac joint imaging when trying to make a diagnosis of axSpA. However, in certain circumstances (eg, discriminating pain from DDD vs axSpA), spine MRI can be informative. (Weak recommendation against.)

Recommendation 56. In adults with a known diagnosis of axSpA, we do not recommend routinely obtaining an MRI for every patient with active disease. (Weak recommendation against.)

It may be reasonable to order an MRI in patients with inadequate response to therapy to confirm ongoing disease activity or to identify alternative causes of pain. Currently, there is insufficient evidence to suggest that clinicians should target MRI remission, as it is unknown whether patients who notice symptomatic benefit with biologic or JAKi therapy but who continue to show activity on MRI will have worse outcomes if therapy is not switched or increased.