Ankylosing Spondylitis

Ankylosing spondylitis (AS) is the radiographic subset of axial spondyloarthritis (axSpA) — a chronic, immune-mediated inflammatory disease centred on the sacroiliac joints and spine, strongly associated with HLA-B27, that produces inflammatory back pain and progressive stiffness. AS refers specifically to axSpA with definite sacroiliitis visible on a plain radiograph (see Assessment & Diagnosis); non-radiographic axSpA describes the same disease process before, or without ever developing, that radiographic change, and shares essentially the same clinical presentation, extra-articular manifestations, and treatment approach.^[1]

Left inadequately controlled, chronic inflammation at the spine's entheseal insertions can drive new bone formation (syndesmophytes) that bridges adjacent vertebrae over years, producing the classic fused "bamboo spine" and a fixed, stooped posture — a structural outcome that current treatment aims to prevent, not merely to control symptoms.^[2] AS is 1 disease within a broader family — the spondyloarthritis (SpA) spectrum — that also includes psoriatic arthritis, reactive arthritis, inflammatory-bowel-disease-associated arthritis, and undifferentiated peripheral SpA. These conditions share genetic risk, extra-articular manifestations (acute anterior uveitis, inflammatory bowel disease, psoriasis), and much of their treatment ladder, which is why axSpA/AS management is best understood as 1 point on that shared spectrum rather than in isolation.

The hallmark presentation is chronic inflammatory back pain — insidious low back and buttock pain, often alternating sides (reflecting sacroiliitis), with morning stiffness lasting more than 30 minutes that improves with activity and worsens with rest, frequently severe enough to wake the patient in the second half of the night. Onset is typically insidious, before age 45, and most often in the late teens through early 30s (see Assessment & Diagnosis for the specific screening criteria and why 2 different age cutoffs are used for 2 different purposes).^[3] Enthesitis — inflammation where a tendon or ligament inserts into bone — is a distinguishing SpA-spectrum feature, most classically at the Achilles tendon and plantar fascia insertions, and also seen at the costochondral junctions, iliac crests, and greater trochanters. Peripheral arthritis, when present, is typically an asymmetric oligoarthritis favouring the lower limbs (hips, knees, ankles) — a different pattern from rheumatoid arthritis's symmetric small-joint polyarthritis. Dactylitis ("sausage digit," diffuse swelling of an entire finger or toe) can occur, though it is more characteristic of psoriatic arthritis. As axial disease advances over years, reduced spinal mobility and chest expansion progress toward a fixed, stooped ("question mark") posture with loss of lumbar lordosis and increased thoracic kyphosis in untreated or inadequately controlled disease.

• HLA-B27 positivity — the strongest genetic association; present in roughly 90% of White patients with AS (lower in some other ancestries), though most HLA-B27-positive people never develop AS.
• Family history of AS, axSpA, or another spondyloarthritis-spectrum condition.
• Male sex — historically a more pronounced predominance in radiographic AS specifically than in non-radiographic axSpA, which has a more balanced sex distribution.
• Onset before age 45, most commonly in the late teens through early 30s; new-onset AS is uncommon after age 45.
• Cigarette smoking — associated with greater radiographic (structural) progression, higher measured disease activity, and a reduced response to TNF inhibitor therapy.
• A personal history of psoriasis, inflammatory bowel disease, or reactive arthritis — each increases the risk of also developing axial spondyloarthritis.

Differentiating inflammatory from mechanical back pain

Correctly identifying inflammatory back pain is the single most useful screening step for axSpA in a patient with chronic back pain, and changes the entire diagnostic pathway.

The dedicated ASAS inflammatory-back-pain screening tool formalizes this: in a patient with chronic back pain, 4 or more of the following 5 parameters supports inflammatory back pain — onset before age 40, insidious onset, improvement with exercise, no improvement with rest, and pain at night with improvement on rising. Note that this specific screening tool uses an age-40 cutoff, distinct from the age-45 onset cutoff used as the entry criterion for the broader ASAS axSpA classification criteria below — 2 different tools built for 2 different purposes, not a contradiction. This tool performed comparably to older instruments (e.g., the Calin criteria) in validation studies and is a genuinely useful bedside screen, though it is a screening aid rather than a diagnostic test on its own — imaging and specialist assessment remain necessary to confirm axSpA.^[3]

History

• Age at onset, pattern of onset, and response to exercise, rest, and NSAIDs (see the tables above).
• Peripheral joint symptoms, enthesitis symptoms (heel pain, plantar fascia pain), and dactylitis.
• Extra-articular symptom screen: eye redness or pain, bowel symptoms, psoriasis, and family history of any SpA-spectrum condition.

Physical examination

• Modified Schober test — mark the skin at the lumbosacral junction (the dimples of Venus) and 10 cm above; ask the patient to bend forward maximally; an increase of less than 5 cm suggests reduced lumbar spinal flexion.
• Chest expansion — measured at the 4th intercostal space between maximal expiration and inspiration; less than about 2.5 cm is abnormal and supports axial involvement.
• Occiput-to-wall and tragus-to-wall distance — assesses cervical spine and thoracic kyphosis involvement in more advanced disease; both should be close to 0 in a patient able to stand fully upright against a wall.
• Peripheral joint exam for synovitis; entheseal palpation (Achilles insertion, plantar fascia, greater trochanter, costochondral junctions); screen for dactylitis and psoriatic skin/nail changes.

Investigations

• HLA-B27 — supportive but neither necessary nor sufficient on its own; a negative result does not exclude axSpA, especially outside White populations.
• CRP and/or ESR — elevated in only about 40-50% of patients with active axSpA even at diagnosis; a normal CRP/ESR does not exclude active disease, a genuinely important distinguishing point from more uniformly CRP-driven conditions such as rheumatoid arthritis.
• Plain radiographs of the sacroiliac joints — the first-line imaging test; look for the radiographic sacroiliitis defined in the modified New York criteria below.
• MRI of the sacroiliac joints — obtained when radiographs are normal or equivocal and clinical suspicion remains, to detect active inflammation (bone marrow edema) before radiographic change has developed (non-radiographic axSpA).
• CBC, creatinine, and a liver panel — baseline organ-function assessment before starting NSAID or DMARD/biologic therapy.

Modified New York criteria (1984) — the original AS-specific classification

The original disease-defining framework for AS combines 1 radiographic criterion with at least 1 of 3 clinical criteria. It remains useful as a teaching anchor for what "definite AS" (as opposed to non-radiographic axSpA) means, though the newer ASAS axSpA classification criteria below are now used more broadly in practice and research.^[4]

"Definite AS" requires the radiographic criterion plus at least 1 clinical criterion. This framework predates modern MRI-based early detection and, by design, misses non-radiographic axSpA — it remains a valid definition of established, radiographically evident disease rather than a tool for early diagnosis.^[4]

ASAS classification criteria for axial spondyloarthritis

Applies to a patient with 3 months or longer of back pain with onset before age 45. Either arm below is sufficient, and both require imaging or laboratory confirmation rather than a clinical impression alone. Presented here as reference teaching content, not an interactive calculator — applying it correctly requires a hands-on assessment plus imaging/laboratory interpretation.^[5]

SpA features usable toward either arm: inflammatory back pain, arthritis, enthesitis (heel), uveitis, dactylitis, psoriasis, Crohn disease/ulcerative colitis, good response to NSAIDs, family history of SpA, HLA-B27, and elevated CRP.

A patient meeting the imaging arm's radiographic-sacroiliitis definition, plus at least 1 clinical criterion from the modified New York framework above, is specifically classified as AS (radiographic axSpA) rather than non-radiographic axSpA — a distinction that matters for some jurisdictions' funding criteria but not for the treatment approach in this chapter, which applies to both.^[5]

• Control signs, symptoms, and inflammation — pain, stiffness, and fatigue.
• Prevent structural damage: slow or halt radiographic progression and new bone formation (syndesmophyte development) at the spine.
• Preserve spinal mobility, chest expansion, physical function, work capacity, and quality of life.
• Identify and manage extra-articular manifestations (uveitis, inflammatory bowel disease, psoriasis) and comorbidities (cardiovascular risk, osteoporosis) alongside the joint disease, not as an afterthought.
• Regular clinical monitoring with an objective composite disease-activity measure (ASDAS or BASDAI — see Monitoring & Follow-Up), adjusting therapy when disease remains active, is recommended.^[6] A dedicated treat-to-target strategy trial in axSpA did not show a formal treat-to-target protocol to be superior to good-quality usual care judged against the same measures — a genuine, honestly-reported difference from the stronger treat-to-target evidence base in rheumatoid arthritis — so regular, attentive monitoring and responsive treatment adjustment matter more than adherence to any single rigid protocol.^[2]

• Structured exercise is an essential, first-line component of AS treatment — not an optional adjunct — and should be prescribed at diagnosis regardless of current disease activity or drug therapy.^[6] Effective programs combine stretching/flexibility work, strengthening, and aerobic conditioning; supervised, individual, or group-based (including patient-association exercise groups) formats all have evidence of benefit for pain, spinal mobility, and function.
• Encourage extension-based exercises and attention to posture specifically to counter the disease's natural tendency toward a flexed, stooped spine as it progresses — a genuinely disease-specific rationale beyond general fitness advice.
• Both land-based and water-based (aquatic) exercise programs have supporting evidence; swimming is commonly recommended given its low-impact, extension-promoting nature.
• Physiotherapy referral at diagnosis, with periodic reassessment and program adjustment as disease activity and mobility change over time.
• Smoking cessation — reduces radiographic progression risk and improves response to TNF inhibitor therapy, in addition to general cardiovascular benefit.
• Patient education and self-management programs, including connection with patient associations (e.g., the Canadian Spondylitis Association), improve long-term engagement with the exercise-centred treatment plan.
• Occupational/ergonomic counselling — workstation and sleep-posture adjustments to minimize sustained spinal flexion.
• There is no good evidence to support a specific diet or complementary/alternative therapy for disease control.

• NSAIDs can be used in the 1st and 2nd trimesters but should be avoided in the 3rd trimester given antiprostaglandin effects (premature ductus arteriosus closure, oligohydramnios, and interference with labour).
• Sulfasalazine can be continued throughout pregnancy with folic acid supplementation, given sulfasalazine's antifolate effect.
• Certolizumab pegol is compatible with all trimesters given minimal-to-no placental transfer, and is generally the preferred TNF inhibitor when a biologic is needed close to delivery.
• Adalimumab, etanercept, golimumab, and infliximab can generally be continued through the 2nd trimester, with increasing placental transfer in the 3rd trimester — discuss infant live-vaccine timing with the treating team whenever 1 of these was used later in pregnancy.
• IL-17 inhibitors and JAK inhibitors have limited human pregnancy data — avoid unless no adequate alternative controls disease.
• Methotrexate is not used for AS, so its pregnancy contraindication is not a practical issue in this chapter, unlike in rheumatoid arthritis.

Breastfeeding

NSAIDs (short-acting agents preferred), sulfasalazine (except in an infant with hyperbilirubinemia), and certolizumab pegol are compatible with breastfeeding. Data for other bDMARDs and for JAK/IL-17 inhibitors in lactation are limited; biologics generally transfer into breast milk only in small amounts given their large molecular size, and use is decided case-by-case with the treating specialist.

Pediatric

Enthesitis-related arthritis is the pediatric SpA-spectrum category and is a distinct clinical entity with its own specialist-directed approach — this chapter covers adult AS/axSpA only. A patient with childhood-onset axial or enthesitis-predominant inflammatory disease needs pediatric rheumatology referral rather than adult-dosing extrapolation.

Perioperative and anesthesia considerations

Advanced AS with cervical spine involvement and reduced neck extension is a genuine difficult-airway/intubation risk that must be flagged before any procedure requiring general anesthesia — distinct in mechanism from rheumatoid arthritis's atlantoaxial subluxation risk, but requiring the same proactive anesthesia-team notification. Hold a bDMARD before elective surgery, timing withdrawal to the end of that agent's dosing cycle, and restart once the wound shows evidence of healing; hold a JAK inhibitor for at least 7 days before surgery. NSAIDs and sulfasalazine can typically continue through minor surgery; confirm the specific plan with the treating rheumatologist.

Renal and hepatic impairment

NSAIDs require particular caution with chronic continuous use given the AS treatment approach itself — periodically reconfirm renal function, especially in an older patient or one with additional cardiovascular/renal risk factors, and do not assume long-term continuous NSAID therapy is risk-free simply because it is guideline-recommended. Most bDMARD/tsDMARD dosing is not renally or hepatically adjusted, but active infection and organ function should be reconfirmed before every dose escalation.

Older adults

New-onset AS is uncommon after age 45, but a patient can carry years of unrecognized symptoms before diagnosis at an older age. Apply the JAK-inhibitor cardiovascular/malignancy risk stratification specifically at age 65 and older, favouring a bDMARD first unless TNF-inhibitor therapy has already failed or was not tolerated, and reassess the ongoing necessity and renal/GI/cardiovascular safety of continuous NSAID therapy more frequently in this group.

See patients with active disease every 3-6 months; extend the interval once disease is well-controlled. Adjust therapy against persistent active disease rather than continuing an unchanged regimen — see Treatment Targets for the evidence nuance around formal treat-to-target protocols in axSpA specifically.^[2]

Disease activity measurement (ASDAS and BASDAI)

The Ankylosing Spondylitis Disease Activity Score (ASDAS) — a weighted composite of patient-reported back pain, peripheral pain/swelling, duration of morning stiffness, patient global assessment, and CRP or ESR — is the current preferred composite measure; the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), a fully patient-reported 0-10 index across 6 symptom domains, remains widely used, particularly where CRP testing is less accessible. Both use continuously weighted formulas and structured longitudinal tracking best done by the specialist team, and are presented here as reference only, with no self-service calculator.

Spinal mobility and function

Reassess the modified Schober test, chest expansion, and occiput/tragus-to-wall distance (see Assessment & Diagnosis) at least annually, even when symptoms are well-controlled — structural progression can be clinically silent, and a falling mobility measure is itself a reason to reassess disease control rather than assume treatment is working.

Pre-biologic screening

• Tuberculosis: chest x-ray and a TST or IGRA before starting any bDMARD or JAK inhibitor; exclude active TB and start isoniazid prophylaxis in anyone who screens positive before proceeding.
• Hepatitis B and C serology before starting any bDMARD or JAK inhibitor.

Vaccination around immunosuppression

Update vaccinations before starting therapy whenever possible, particularly for any live vaccine. Live vaccines are avoided during bDMARD or JAK-inhibitor therapy; inactivated/recombinant vaccines remain recommended once therapy has started. Use the recombinant (inactivated) herpes zoster vaccine, never the live formulation, in a patient on or starting immunosuppression.

Ocular surveillance

There is no routine surveillance eye-exam schedule for AS, unlike hydroxychloroquine retinopathy monitoring in some other conditions — the actual, effective screening strategy is patient education: every patient should know to seek same-day ophthalmology assessment for any new red, painful eye with photophobia, given the substantial lifetime anterior uveitis risk (see Red Flags & Refer).

Bone health

Consider DXA screening given the elevated fracture risk in AS, particularly with longstanding or advanced (fused) disease; ensure adequate calcium (about 1200 mg/day total intake) and vitamin D (800-2000 units/day), and add a bisphosphonate when osteoporosis is confirmed or fracture risk is significant.

Cardiovascular risk

Routine cardiac screening (e.g., ECG or echocardiography) is not recommended in an asymptomatic patient with AS and no cardiac symptoms — reserve cardiac workup for the red-flag presentations above.^[2] Do apply standard age-appropriate cardiovascular risk-factor management (blood pressure, lipids, glucose, smoking cessation) given AS's elevated background cardiovascular risk independent of any specific cardiac complication.^[1]

Renal and GI monitoring for chronic NSAID therapy

Periodic creatinine and CBC given that long-term continuous NSAID use is the treatment approach itself in axSpA, not an occasional add-on; reassess GI bleeding risk and consider gastroprotection for a patient with additional risk factors.

1. Confirm an adequate NSAID trial before declaring failure — maximum tolerated/labelled dose, given continuously (not PRN), for at least 2-4 weeks, with at least 2 different NSAIDs tried.
2. Confirm adherence — continuous, not intermittent, dosing is a common, reversible cause of apparent nonresponse.
3. Reassess disease activity with a structured measure (ASDAS or BASDAI), not symptoms alone, before escalating.
4. If active disease persists despite adequate NSAID trials, start a biologic — a TNF monoclonal antibody for recurrent uveitis or active/past IBD, an IL-17 inhibitor for significant psoriasis without uveitis/IBD, or either class when no extra-articular manifestation predominates — after completing pre-biologic screening.
5. Give the 1st bDMARD an adequate trial, typically around 12 weeks, before judging inadequate response.
6. If inadequate response to the 1st bDMARD, switch to a 2nd bDMARD — either within the same class or to the other class — reconsidering extra-articular manifestations again when choosing.
7. If inadequate response to 2 bDMARDs (ideally 1 from each class), consider a JAK inhibitor (upadacitinib or tofacitinib), applying the cardiovascular/malignancy risk stratification first.
8. In a patient who has achieved sustained remission or inactive disease on a bDMARD, cautious dose tapering (reduced dose or extended interval), rather than abrupt discontinuation, can be considered — full discontinuation is associated with a high relapse rate.^[6]
9. Persistent, localized peripheral or entheseal symptoms despite good axial disease control can often be addressed with a local corticosteroid injection rather than escalating systemic therapy.
10. At any step, re-examine the original diagnosis if response remains genuinely inadequate despite confirmed adherence and adequate trials.

• Dosing an NSAID only as needed instead of continuously in active axSpA — undermines both symptom control and the guideline-supported treatment rationale.
• Prescribing methotrexate, or relying on sulfasalazine, expecting axial (spinal) benefit — neither has proven efficacy for axial disease; sulfasalazine's evidence is limited to the peripheral-joint component.
• Declaring NSAID failure after trialing only 1 agent — a 2nd, different NSAID deserves a genuine trial first.
• Choosing etanercept over a TNF monoclonal antibody in a patient with active or recurrent anterior uveitis or inflammatory bowel disease — etanercept is meaningfully less effective for these manifestations.
• Choosing an IL-17 inhibitor in a patient with active or strongly suspected inflammatory bowel disease — this class can trigger or worsen IBD.
• Reaching for a JAK inhibitor as the 1st non-NSAID therapy instead of after bDMARD failure.
• Skipping tuberculosis and hepatitis B/C screening before starting a bDMARD or JAK inhibitor.
• Ordering routine cardiac screening in an asymptomatic AS patient with no cardiac symptoms — not supported; use standard risk-factor management instead, and reserve cardiac workup for genuine red-flag presentations.
• Missing an acute uveitis presentation because the patient or clinician did not recognize the eye-spine connection — proactively educate every patient at diagnosis.
• Manipulating the neck, or delaying imaging, after new pain in a patient with a longstanding fused spine — treat as a probable fracture until proven otherwise.
• Treating structured exercise as optional or secondary to medication — it is an essential, first-line component of therapy in its own right.
• Continuing chronic NSAID therapy indefinitely without periodic renal, GI, and cardiovascular risk reassessment.

How to use this table: Doses are adult doses and require patient-specific confirmation, including renal/hepatic adjustment. Select any drug name to open its complete Pharmasist Drug Database page.

NSAIDs — continuous therapy is the treatment approach itself in axSpA

Any of the agents below is a reasonable 1st choice; a 2nd, different NSAID deserves its own genuine trial before concluding NSAID therapy has failed.

Conventional synthetic DMARD — peripheral arthritis only

No proven efficacy for axial (spinal/sacroiliac) disease — not a substitute for a biologic when axial disease is active.

Tumor necrosis factor (TNF) inhibitors

1st-line bDMARD class. A TNF monoclonal antibody (every agent in this class except etanercept) is preferred over etanercept, and over an IL-17 inhibitor, when recurrent anterior uveitis or inflammatory bowel disease is present. Complete TB and hepatitis B/C screening before starting every agent in this class.

Interleukin-17 (IL-17) inhibitors

Equivalent 1st-line biologic option to a TNF inhibitor; a reasonable preferred choice when significant psoriasis is present without uveitis or IBD. Avoid, or use with caution, in active or strongly suspected inflammatory bowel disease — this class can trigger or worsen IBD. Complete the same pre-therapy infection screening as the TNF inhibitors.

Janus kinase (JAK) inhibitors (tsDMARDs)

Reserved for inadequate response to, or intolerance of, a bDMARD — currently the only 2 JAK inhibitors approved for axSpA in Canada. Both carry a class-wide boxed warning for serious infections, malignancy/lymphoma, major adverse cardiovascular events, and thrombosis; use with added caution — favouring a TNF inhibitor or IL-17 inhibitor first — in patients 65 or older, current or past smokers, or those with additional cardiovascular risk factors or a malignancy history.

Local corticosteroid injection

For a flare localized to a small number of peripheral joints, entheseal sites, or the sacroiliac joint itself; not a substitute for adequate systemic disease control.

Pre-biologic infection prophylaxis

Bone-health support

For a patient with confirmed osteoporosis or significant fracture risk (see Monitoring & Follow-Up).

Adjunct analgesia