INTRODUCTION — Axial spondyloarthritis (axSpA) is a potentially disabling inflammatory arthritis of the spine, usually presenting as chronic back pain, typically before the age of 45. It is often associated with one or more of several articular features, including synovitis, enthesitis, and dactylitis. It may also be associated with several non-articular features; these include uveitis, psoriasis, and inflammatory bowel diseases. Patients frequently carry the gene for human leukocyte antigen (HLA)-B27, and patients with active inflammatory disease often have evidence of an elevated acute phase response. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults".)
Patients with axSpA may be classified as having either of two subtypes of axSpA: radiographic axSpA (abbreviated as r-axSpA, historically termed ankylosing spondylitis [AS]) or nonradiographic axSpA (nr-axSpA). Patients with r-axSpA exhibit radiographic abnormalities consistent with sacroiliitis, but such findings are not evident on plain radiography in nr-axSpA. Instead, in patients with nr-axSpA, the diagnosis is supported by evidence of active inflammation of the sacroiliac (SI) joints on magnetic resonance imaging (MRI) and/or a combination of other findings. In clinical practice, distinction between these forms of axSpA in an individual patient has limited impact on management and may not be relevant, although the classification is of interest for epidemiologic and other investigative purposes. (See 'Nomenclature and classification' below.)
The diagnostic evaluation, diagnosis, and differential diagnosis of r-axSpA and nr-axSpA will be reviewed here together as axSpA. The clinical manifestations, treatment, and pathogenesis of this disorder are discussed separately. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults" and "Treatment of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults" and "Pathogenesis of spondyloarthritis".)
NOMENCLATURE AND CLASSIFICATION — Historically, spondyloarthritis (SpA, formerly termed spondyloarthritides or spondyloarthropathies) has been regarded as a family of arthritis that includes ankylosing spondylitis (AS), undifferentiated SpA, reactive arthritis, psoriatic arthritis, juvenile SpA, and arthritis and spondylitis associated with inflammatory bowel diseases, including Crohn disease and ulcerative colitis . AS has been regarded as the prototype of SpA, and the majority of historical publications on SpA focus on AS. (See "Clinical manifestations and diagnosis of peripheral spondyloarthritis in adults" and "Reactive arthritis" and "Clinical manifestations and diagnosis of psoriatic arthritis" and "Spondyloarthritis in children" and "Clinical manifestations and diagnosis of arthritis associated with inflammatory bowel disease and other gastrointestinal diseases".)
The term axial spondyloarthritis (axSpA) became standard nomenclature in 2009, based upon a multinational study of SpA by the Assessment of SpondyloArthritis International Society (ASAS) classifying SpA broadly into two groups [2,3]:
●SpA with predominantly axial involvement is designated as "axial SpA" (axSpA), which comprises radiographic axSpA (r-axSpA or AS), typically with radiographic sacroiliitis on plain radiography, and nonradiographic axSpA (nr-axSpA), without plain radiographic changes of sacroiliitis. Both are part of the same disease. About 10 to 40 percent of patients with nr-axSpA will develop radiographic sacroiliitis within 2 to 10 years of follow-up and classify as r-axSpA . In a clinical sense, patients with radiographically evident disease are not substantially different from those without radiographic signs of axSpA and differentiation between r-axSpA and nr-axSpA is of limited use for prediction of important clinical outcomes [5,6].
●SpA with predominantly peripheral involvement is designated as "peripheral SpA," with symptoms mainly of peripheral arthritis, peripheral enthesitis, and/or dactylitis. (See "Clinical manifestations and diagnosis of peripheral spondyloarthritis in adults".)
All classification criteria, including the 2009/2001 ASAS classification of SpA, the 1984 Modified New York Criteria for AS, the 1990 Amor criteria for SpA, and the 1991 European Spondyloarthritis Study Group (ESSG) criteria for SpA, have been developed for use in research such as epidemiologic studies and therapeutic trials (table 1) [7,8].
All criteria have disadvantages that limit their use as diagnostic tools in practice. Diagnosis is typically made based upon the clinician's judgment about the combination of symptoms, physical signs, and laboratory and imaging findings present in an individual patient . (See 'Diagnosis' below and "Overview of the clinical manifestations and classification of spondyloarthritis", section on 'Classification of spondyloarthritis'.)
GENERAL PRINCIPLES — The diagnosis of axial spondyloarthritis (axSpA) involves recognition by the clinician of a pattern of features that taken together are characteristic of axSpA. The specific combination of historical features and physical, laboratory, and imaging findings can differ between patients, and the diagnosis requires exclusion of other potential causes of these abnormalities. (See 'Evaluation of suspected axial spondyloarthritis' below and 'Diagnosis' below and 'Differential diagnosis' below.)
The likelihood of the diagnosis varies depending upon the specific findings that are present. There is no single historical feature, physical finding, laboratory test, or imaging study with sufficient specificity by itself to establish the diagnosis without the presence of additional abnormalities. Thus, the diagnosis cannot depend only upon a single checklist of features; and classification criteria, which are designed for use in epidemiologic and clinical research, cannot be relied upon to establish a diagnosis of axSpA in practice. (See 'Diagnostic approach' below and 'Important additional diagnostic considerations' below.)
EVALUATION OF SUSPECTED AXIAL SPONDYLOARTHRITIS — The diagnosis of axial spondyloarthritis (axSpA) should be considered in patients with an onset before age 45 of continuous chronic back pain, although few patients with chronic back pain and only a minority of the fraction with the features typical of inflammatory back pain (IBP) are likely to have axSpA . Many other conditions may cause back pain in clinical practice; most often such pain is of shorter duration than is typical of axSpA. (See "Evaluation of low back pain in adults" and 'Differential diagnosis' below.)
When to refer — Patients with chronic back pain and any features suggesting axSpA, including IBP (see 'History' below) or other characteristic symptoms or findings, should undergo a thorough rheumatologic history, physical examination, and selected laboratory testing, with particular attention to features characteristic of this condition (see "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults") as well as those features typical of disorders usually considered in the differential diagnosis. (See 'Differential diagnosis' below.)
For the non-rheumatologist, early referral to a rheumatologist based upon a limited set of symptoms and findings may facilitate rapid diagnosis and institution of effective treatment. Several referral strategies have been proposed. In a large multicenter study with 515 referred patients in nine centers from seven Middle Eastern Arab countries, the value of these strategies in diagnosing axSpA was compared . The optimal referral strategy comprised IBP, good response to nonsteroidal antiinflammatory drugs (NSAIDs), a positive human leukocyte antigen (HLA)-B27, and a positive family history of SpA . Other issues that should prompt rheumatologic referral include a positive test for HLA-B27, the presence of IBP, or findings of sacroiliitis on imaging in patients with chronic back pain starting at an age younger than 45 years [13-15]. However, the presence of any other typical SpA features can also justify referral, depending upon local specialist availability and health care policies .
Patients initially suspected of axSpA should also be monitored during follow-up to detect changes that may refute or strengthen the diagnosis of axSpA.
History — The medical history should include ascertainment of the following: the duration and characteristics of back pain, including features that may suggest IBP; other musculoskeletal manifestations, such as arthritis, tendinitis, enthesitis, and dactylitis; disease of the eyes, such as conjunctivitis and iritis or uveitis; history of skin disease, including psoriasis; gastrointestinal symptoms and diagnoses compatible with inflammatory bowel disease; and, for differential diagnosis, history of musculoskeletal mechanical back pain or "lumbago," back injury, radiculopathy, or surgery. Patients should be asked about a family history of disorders associated with HLA-B27, including uveitis and axSpA or other forms of SpA.
Information should be obtained regarding prior therapies, particularly NSAIDs, including the doses of the medications, dosing schedules, and the response to previous treatments.
The presence of clinical features of SpA at the time of the interview or in the past that increase the probability of a SpA include :
●Inflammatory back pain – Several criteria have been proposed for defining IBP [1,17-19]; we use the "ASAS (Assessment of SpondyloArthritis International Society) expert criteria" to identify patients with IBP . The criteria for IBP are met if at least four of the following five features are present and IBP is described as "suggested" in the presence of three of five features:
•Onset of back discomfort before the age of 40 years
•Improvement with exercise
•No improvement with rest
•Pain at night (with improvement upon arising)
The presence of four of the five features listed above has a sensitivity and specificity of 80 and 74 percent, respectively, for an inflammatory cause of the back pain among patients with chronic back pain of unclear origin and onset of the back pain at <45 years of age who were evaluated by local rheumatologists . However, when IBP is present, and considered independently of other factors, the probability of axSpA among patients with chronic back pain increases only from 5 percent to 14 to 16 percent .
The prevalence of axSpA in the general population is low compared with that of other causes of back pain; thus, the proportion of false-positive diagnoses of axSpA is high if the presence of IBP is used as the sole diagnostic criterion. Approximately 20 percent of individuals with other causes of low back pain complain of symptoms similar to those of IBP, and 40 percent of IBP patients resolve spontaneously without developing axSpA . In addition, a considerable number of patients with axSpA will not have back pain that fulfills these IBP criteria [21,22]. Therefore, absence of IBP does not exclude a diagnosis of axSpA, and this characteristic should be used together with other features of SpA in the diagnostic assessment.
●Alternating buttock pain – Such pain is described as alternating between the left and the right gluteal regions (reflecting alternating sacroiliitis) rather than remaining unilateral.
●Heel pain caused by enthesitis – Heel pain due to enthesitis presents as pain or tenderness at the heel involving the insertion of the Achilles tendon and/or the plantar fascia into the calcaneus.
●Dactylitis – Dactylitis is diffuse swelling of digits of toes or fingers, also termed "sausage digits" (picture 1A).
●Asymmetric arthritis predominantly of the lower limbs – This includes a past or present history of arthritis, especially if it is oligoarticular, asymmetrical, and predominantly of the lower extremities, which is most characteristic of arthritis associated with SpA.
●Anterior uveitis (iritis) – This is most helpful if the uveitis was diagnosed by an expert in uveitis, such as an ophthalmologist, usually by slit-lamp examination.
●Crohn disease or ulcerative colitis – Past or present history of verified Crohn disease or ulcerative colitis, which increases the likelihood of axSpA .
●Psoriasis – Past or present psoriasis, which increases the likelihood of axSpA.
●Good response of pain symptoms to NSAIDs – History of a good response to an NSAID is very useful in the diagnosis of axSpA. Although patients with back pain of many causes may improve when treated with antiinflammatory doses of an NSAID, marked improvement in pain within 24 to 48 hours supports the diagnosis of an axSpA but is not specific by itself as a diagnostic feature [9,17].
●Family history of SpA – A positive family history is defined as the presence in a first- or second-degree relative of a diagnosis of axSpA, uveitis, reactive arthritis, psoriasis, or inflammatory bowel disease. In particular, the presence of a family history of axSpA or uveitis has been useful for identifying axSpA in patients with chronic back pain .
Patients with axSpA may also complain of anterior chest wall/rib cage pain, but this symptom has poor specificity for SpA and is not useful diagnostically.
Physical examination — All patients suspected of axSpA should undergo a thorough physical examination. However, physical findings are not very specific and are more useful for monitoring of the disease than for diagnosis of axSpA. These include examination of the spine for ranges of motion, the joints and the heels for arthritis and enthesitis, and the hands and feet for dactylitis; and examination of the skin and eyes for extraarticular features, including psoriasis and uveitis. Other findings may help identify or exclude an alternative diagnosis (see 'Differential diagnosis' below) or comorbidity and establish a baseline for ongoing clinical assessment of the patient. (See "Treatment of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults".)
Features of the physical examination of particular importance for diagnosis include:
●Peripheral joint involvement – The peripheral joints, including the hands, wrists, elbows, shoulders, hips, knees, ankles, and feet, should be examined to determine the number of swollen and/or tender joints as well those with deformities and/or loss of range of motion.
●Enthesitis at the heel – The most common and readily evident site of enthesitis on physical examination is the heel, where tenderness and/or swelling may involve affected patients at the insertions of the plantar fascia and/or Achilles tendon into the calcaneus (picture 2). However, any enthesis can be affected in axSpA and should be evaluated depending upon the patient's symptoms.
●Dactylitis – The presence of dactylitis, which is diffuse swelling of digits of toes or fingers, should be noted (picture 1A-B). Affected fingers or toes are commonly also termed "sausage digits." (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Dactylitis'.)
●Skin – The skin, including the scalp, and nails should be examined for evidence of psoriasis. (See "Psoriasis: Epidemiology, clinical manifestations, and diagnosis", section on 'Clinical manifestations' and "Psoriasis: Epidemiology, clinical manifestations, and diagnosis", section on 'Diagnosis'.)
●Eyes – Patients with symptoms or signs that suggest possible anterior uveitis, such as eye pain, visual disturbance, or redness, should be referred for evaluation and treatment by an expert in uveitis, such as an ophthalmologist, and require examination using a slit lamp to confirm the diagnosis. (See "Uveitis: Etiology, clinical manifestations, and diagnosis", section on 'Symptoms and findings' and "Uveitis: Etiology, clinical manifestations, and diagnosis", section on 'Diagnosis and referral'.)
Various measurements of mobility of the axial skeleton are routine in practice for assessing restriction of spinal mobility and for monitoring the response to therapy but are subject to too large a variation in normal subjects to be of diagnostic utility [25-28]. Reference intervals for various spinal mobility measurements in Europeans are available through the ASAS website . Norms for the United States population are also available . A measurable occiput-to-wall distance with dorsal kyphosis is unusual in normal individuals in both the United States and Europe.
The considerable variation among healthy individuals in the ranges of motion of the cervical, thoracic, and lower spine is influenced by the population studied, body habitus, age, and sex. Thus, the threshold values described here should be considered as estimates, rather than precise cut-off points, between normal and abnormal [28,29]. The various percentiles are shown in https://www.asas-group.org/clinical-instruments/mobility-curves/. Details of methods are shown in https://www.asas-group.org/education/asas-handbook/.
The following parameters are those generally used to examine axial skeletal mobility in clinical practice:
●Cervical spine – axSpA may cause forward stooping of the thoracic and cervical spine. The degree of flexion deformity can be measured by asking the patient to stand erect with heels and buttocks against a wall and to extend his or her neck, while keeping the chin (mandible) horizontal in an attempt to touch the wall. Almost all normal individuals can touch the wall with the occiput. The distance between the occiput and the wall in a patient with axSpA reflects the degree of cervical deformity (figure 1). In addition, the range of flexion, extension, lateral flexion, and rotation can also be recorded.
●Thoracic spine – The range of motion of the costovertebral joints is measured by the degree of chest expansion. Chest expansion is measured at the fourth intercostal level. The patients are instructed to raise their arms above their heads and to exert a maximal forced expiration followed by a maximal inspiration. Normal expansion is usually >2 cm.
●Lower spine – Range of motion of the lower spine is tested in the sagittal plane by the Schober test, and in the coronal plane by the extent of lateral spinal flexion.
•Schober test – The Schober test measures the forward flexion of the lumbar spine (picture 3). Many modifications of the Schober test measuring anterior flexion of the lumbar spine exist. The one chosen by the ASAS is performed as follows: The subject is asked to stand erect with feet apart to approximately the same width as the shoulders. A mark is placed in the midpoint of an imaginary horizontal line joining the two posterior superior iliac spines, which are close to the dimples of Venus. A second mark is then placed in an imaginary vertical line 10 cm above the first mark. The patient is then instructed to bend forward as much as possible as if to touch the floor with the fingers, keeping the knees straight. The difference between the two upper marks in the erect and forward bent position is then measured. The test is usually performed twice, with the greater measurement being accepted as the more accurate one. A measurement of ≥5 cm is considered normal, but a difference as low as 2.2 cm may be seen in some healthy individuals . The test is thus most useful for serial comparisons of a given individual.
•Lateral spinal flexion – With the patient standing erect with heels and back against a wall and knees and hands extended, the distance between the tip of the middle finger and the floor is measured. The patient is then instructed to bend sideways without bending the knees or lifting the heels. A second measurement is made and the difference between the two is recorded. The final result is the averaged measurements of right and left flexion. Normal is greater than 10 cm.
Laboratory testing — There are no laboratory tests that are by themselves diagnostic of axSpA, although two types of tests can contribute to making the diagnosis (see 'Diagnosis' below). These include:
●HLA-B27 testing – In most studies, HLA-B27 is present in 85 to 95 percent of White patients with axSpA but in only approximately 8 percent of the general population in many geographical regions [30,31]. However, in White northern European populations, the prevalence of axSpA in the HLA-B27-positive population is only approximately 5 percent [32,33]. The frequency of HLA-B27 in nr-axSpA may be slightly lower than in r-axSpA; the estimated prevalence in nr-axSpA in a study from Germany was between approximately 75 and 85 percent [34,35]. Thus, a positive test for HLA-B27 alone is not diagnostic for axSpA, and a negative test for HLA-B27 does not exclude the diagnosis of axSpA [36,37].
HLA-B27 can be useful to increase the confidence of a diagnosis of axSpA in patients in whom plain radiographs or MRI also exhibit abnormalities consistent with axSpA. In patients in whom plain radiographs or MRI are equivocal for axSpA, positive testing for HLA-B27 would only be useful if there are additional parameters present that also suggest the presence of axSpA (eg, heel enthesitis) (see 'History' above and 'Physical examination' above). The presence of three parameters identified as characteristic features of axSpA brings the probability to about 50 percent. In such a scenario, being HLA-B27-positive increases the probability of having axSpA to 80 to 90 percent. More importantly, a negative test for HLA-B27 in these borderline cases reduces the probability of axSpA to a very low level. For non-White populations, the frequency of HLA-B27 and the prevalence of axSpA have to be considered together before estimating the usefulness of HLA-B27 testing [38,39].
HLA-B27 can also be used as a screening tool in primary care in patients presenting with chronic back pain or IBP suspected by the primary clinician as having a significant probability for axSpA, depending upon the availability and the costs of local HLA-B27 testing [11-13,40,41]. The probability of axSpA goes up from 5 to about 30 percent in chronic back pain patients and from 14 to about 60 percent in patients with IBP if HLA-B27 is positive [9,37]. Thus, these patients might warrant further evaluation, including imaging. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Epidemiology' and "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Laboratory findings'.)
●Acute phase reactants – Acute phase reactants, including the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP), should be measured in patients suspected of axSpA. Acute phase reactants are one of the 11 SpA features used for determining the diagnosis of axSpA (see 'Diagnostic approach' below). We prefer use of the CRP if only one such test is obtained. Elevated CRP levels are observed in almost 40 percent of patients with axSpA, with a higher percentage of patients with elevated levels in the subgroup with r-axSpA compared with the nr-axSpA-subpopulation [5,42]. CRP levels are also used as a component of some composite measures of disease activity . However, a normal CRP or ESR does not exclude a diagnosis of axSpA or that the disease is in an active state. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Laboratory findings' and "Treatment of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults".)
●Other laboratory testing – General laboratory testing is not useful diagnostically, except to the extent that it may suggest an alternative diagnosis. (See 'Differential diagnosis' below.)
Imaging studies — The appropriate use and accurate interpretation of sacroiliac and other joint images are extremely important in the diagnosis of axSpA; many rheumatologists rely upon the imaging more than other characteristics . Typically, patients will have a history of chronic back pain, continuous for three or more months, with onset before age 45 and other features suggesting IBP, and they will also have at least one of the historical features or physical findings typical of SpA (eg, enthesitis or uveitis). (See 'History' above and 'Physical examination' above.)
We take the following approach to the use of imaging studies in patients suspected of axSpA based upon the medical history and examination:
●Plain radiography in all patients – All patients suspected of axSpA should have an anterior-posterior (AP) plain radiograph or Ferguson view of the pelvis to visualize the sacroiliac (SI) joints (see 'Plain radiography' below and 'Diagnostic approach' below). Plain film radiography is frequently sufficient to identify sacroiliitis, and this is also advocated because of its lower radiation exposure compared with conventional computed tomography (CT) and as a cost-effective approach that reduces the need for expensive imaging studies . Patients with a contraindication to radiation exposure should instead undergo MRI.
●MRI of the SI joints in patients suspected of axSpA but with non-diagnostic plain radiographs – MRI of the SI joints is usually indicated only in patients without or equivocal evidence of sacroiliitis on plain radiographs in whom axSpA is suspected based upon other symptoms and findings characteristic of SpA to help establish the diagnosis of nr-axSpA. It is important to recognize that positive MRI findings by themselves are not sufficient to make the diagnosis of axSpA in the absence of other features of SpA; thus, an MRI should only be ordered if there is a reasonable degree of suspicion of axSpA.
Furthermore, SI joint MRIs can be difficult to accurately interpret, and studies should ideally be read by a clinician with expertise in interpreting such images . (See 'MRI of sacroiliac joints' below and 'Diagnostic approach' below and 'Important additional diagnostic considerations' below.)
●Additional potentially useful studies
•MRI of the spine – In a small number of patients in whom nr-axSpA is strongly suspected, but the MRI of the SI joints is normal or nondiagnostic, we perform MRI of the most symptomatic region(s) of the spine. In those patients, MRI of the spine enhances the sensitivity of the diagnosis by 15 to 20 percent. (See 'MRI of the spine' below.)
•Ultrasonography – Ultrasound evaluation may be useful for the identification of enthesitis in selected patients. (See 'Imaging of enthesitis' below.)
We generally do not use CT and bone scintigraphy, as their utility in the evaluation of SpA is limited by various factors, including nonspecificity and lack of standardization. CT scanning is more sensitive than plain radiography for the detection of structural changes in the SI joints but does not provide information regarding inflammation . However, if an abdominal CT is available when ordered for a different purpose, it can be helpful if the SI joints can be visualized. Bone scintigraphy reveals high uptake in areas of inflammation. However, it is too nonspecific to be useful for diagnosis of SpA, has limited sensitivity, and is associated with some radiation exposure . Both CT and bone scintigraphy may be of use in selected patients if MRI is unavailable, but they should be interpreted with caution.
Plain radiography — A standard AP plain radiograph of the pelvis should be obtained to evaluate the SI joints. In a patient with chronic back pain, the presence of obviously significant abnormalities of the SI joint(s) on plain radiographs, including erosions, ankylosis, changes in joint width, or sclerosis, strongly supports the diagnosis of r-axSpA. The SI joint abnormalities are typically graded from 0 to 4 to identify the nature and severity of involvement, and such grading is used to determine the degree of confidence that the changes seen reflect sacroiliitis. The assessment of sacroiliitis on plain radiographs is somewhat subjective, with lower amounts of consensus among multiple readers when changes at a particular SI joint are read as grade 2 or less, levels that indicate uncertainty regarding the presence of changes or only minimally abnormal findings. (See "Clinical manifestations of axial spondyloarthritis (ankylosing spondylitis and nonradiographic axial spondyloarthritis) in adults", section on 'Radiography'.)
A standard AP view of the pelvis has the advantage of including the hips, which should be imaged by plain radiography in patients diagnosed with axSpA and which are potential sites of destructive arthropathy in patients with axSpA. Several other radiographic views have been suggested historically, including obliques, Ferguson, prone, and cone-down SI views. Of these, only the Ferguson view is considered useful [1,48,49].
For each plain radiograph, the right and left SI joints should be evaluated separately and each assigned a grade, based upon the degree of involvement, consistent with the modified New York criteria for AS (table 1) . A patient is regarded as positive for radiographic sacroiliitis if the score is greater than or equal to grade 2 bilaterally or is greater than or equal to grade 3 unilaterally . The grades are defined as follows [50-52]:
●Grade 0: Normal (image 1)
●Grade 1: Suspicious (but not definite) changes (image 2A-B)
●Grade 2: Minimal abnormality – Small localized areas with erosions or sclerosis, without alteration in the joint width (image 2A)
●Grade 3: Unequivocal abnormality – Moderate or advanced sacroiliitis with one or more of the following: erosions, sclerosis, joint space widening, narrowing, or partial ankylosis (image 3A-B)
●Grade 4: Total ankylosis of joints (image 4A-B)
Other examples of the various grades are available in the slide library of the Assessment of Spondyloarthritis International Society.
As an alternative to numerical grading, the degree of radiological sacroiliitis is sometimes graded in clinical practice as either normal, doubtful, and obvious (or definite) .
Common conditions in which the radiographic appearance of the sacroiliac joint might mimic those of axSpA are diffuse idiopathic skeletal hyperostosis (DISH) and osteitis condensans ilii . (See 'Differential diagnosis' below.)
Abnormalities other than sacroiliitis may be seen on AP plain films of the pelvis and reflect the extent of the disease, but are of uncertain diagnostic significance and are not included in the classification criteria. These include erosions and osteitis at the ischial tuberosities, iliac crests, symphysis pubis, femoral trochanters, and hips .
Radiographic hip involvement is sometimes observed in patients with axSpA (image 5). The severity of radiographic hip involvement correlates with more severe spinal disease (image 6 and image 7) .
Radiographs of the spine are not required for the diagnosis of axSpA. AP and lateral views could be performed in patients diagnosed with r-axSpA for assessment of the severity of disease and disease progression, with a focus on the presence of syndesmophytes and bridging. In patients with nr-axSpA, radiographic abnormalities are rarely seen (less than 5 percent) .
Magnetic resonance imaging — The clear presence of bone marrow edema (BME) on MRI in subchondral bone is the only defining observation that is believed to define the presence of active sacroiliitis (in the context of other findings), but other features may be helpful diagnostically, and more information is needed. (See 'MRI of sacroiliac joints' below and 'MRI of the spine' below.)
MRI of sacroiliac joints
●Indications – In patients with obvious sacroiliitis documented by plain radiography, additional imaging, such as an MRI, is not necessary for diagnosis; however, in patients in whom plain radiographs do not reveal sacroiliitis or findings are uncertain, and when there is a high index of clinical suspicion of axSpA, abnormal findings on MRI are an important part of making the diagnosis of nr-axSpA (see 'Diagnosis' below). This is because MRI, unlike plain radiographs, can reveal inflammatory changes, fatty changes, and subtle structural abnormalities. MRI also has better interreader reliability compared with conventional radiography [57,58].
●Sequences – In addition to oblique axial images, semicoronal (coronal oblique) MRI images of the SI joints should be obtained, rather than coronal images, because of the tilt of the SI joints. Images specifically of the SI joints are strongly preferred over images of the pelvis. Only two sequences are necessary: T1-weighted, plus either short tau inversion recovery (STIR) sequences or T2-weighted sequences with fat suppression. Contrast enhancement is not necessary, unless the findings without contrast are uncertain and a high suspicion of axSpA remains. The ASAS definition of active sacroiliitis was updated in 2016 [44,59]. It includes the following:
•Active inflammatory lesions of the SI joints. These lesions should appear as "osteitis" or "bone marrow edema" (BME) on STIR or on T2-weighted images with fat suppression and be clearly present in typical locations, which include subchondral or periarticular bone marrow (image 8 and image 9). The lesions are of high intensity and give a dark signal in the T1-weighted images.
•A positive MRI, which should show at least two BME lesions on the same slice, or one lesion in the same quadrant on at least two consecutive slices. The probability of axSpA sacroiliitis is higher if there is more than one lesion. It is rare for MRI of definite sacroiliitis to show only one lesion. In addition, small lesions should be interpreted with caution.
The detection of structural abnormalities (eg, erosions, bony ankylosis, fat metaplasia, or sclerosis) alone (image 10) may still be useful in those images in which the appearance of BME is ambiguous. Structural lesions by themselves are not viewed as sufficient to meet the definition of a positive MRI for active sacroiliitis . Neither are inflammatory lesions such as synovitis, enthesitis, or capsulitis [61,62]. Of the T1-weighted images, both erosions and fat lesions have moderate diagnostic sensitivities, while the others have limited diagnostic value .
●Pitfalls – The usefulness of MRI is limited by the difficulty of interpreting these images, especially without special training, and also by MRI needing to be interpreted in the context of the degree of clinical suspicion. In addition, active SI joint inflammation is often no longer evident in patients with more advanced r-axSpA, not all patients with nr-axSpA have abnormal MRI findings, and false-positive findings may occur in individuals without axSpA . Other explanations for SI joint BME, especially mechanical stress, should be excluded when considering the presence of subchondral BME on MRI as contributing to establishing a diagnosis of axSpA. A positive MRI defined by findings of BME alone is not diagnostic of axSpA and should be interpreted in the context of other clinical and laboratory findings, as such abnormalities can occur in patients without axSpA [63-66].
As examples, the presence of BME on MRI, even of high intensity, was observed in up to 23 percent of patients with mechanical back pain and 7 percent of healthy volunteers [63,64]. A role for mechanical stress in the lesions in some individuals without SpA has also been shown [65,66]. One study found that in recreational runners and elite athletes (ice-hockey players), BME fulfilling the ASAS definition for positive findings was present in about 30 to 41 percent, with the most affected joint quadrant being the posterior lower ilium, although erosions were virtually absent in these individuals . Another study, which included 47 healthy individuals without back pain and 47 others with axSpA and positive MRI findings, found that while BME could be detected in 23 percent of the healthy individuals, none of the BME findings included deep lesions (a homogenous and unequivocal increase in signal extending at least 1 cm from the articular surface). Efforts to better define those findings that are disease-specific remain important. In general, the higher the number of quadrants of the SI joints showing BME, the less the probability of being a non-SpA lesion .
In general, the MRI should be read taking all features into consideration, as well as their extent and intensity and the possibility of false positives such as blood vessels, and the reader should generate a global index of suspicion of whether the MRI is consistent with axSpA. Interpretation of the MRI changes and determination of whether the MRI of a particular patient is positive or not should always be determined in the context of all of the clinical, laboratory, and other imaging parameters available . Among the most important differential diagnoses are infectious sacroiliitis, DISH, pelvic fractures, and an artifact of imaging termed "coil effect"  (see 'Differential diagnosis' below). We thus prefer that a rheumatologist involved in the care of a given patient participates in the assessment and interpretation of the imaging findings. A negative reading does not necessarily mean that such patients do not need follow-up.
MRI of the spine — Abnormalities of the spine on MRI are not used in the routine diagnosis of axSpA [44,59,70]. Isolated spinal involvement without sacroiliitis is seen in up to 2 percent of the patients with chronic back pain of less than three years' duration . However, MRI of the spine is most useful when pathology other than SpA is suspected. MRI of the more painful areas of the spine may be helpful in the small number of patients in whom an expert has a high level of suspicion for nr-axSpA (eg, patients with clinical features strongly suggesting SpA with normal or uncertain SI joint radiography) but in whom MRI of the SI joints is normal. MRI of the spine may also be useful in patients with severe pain in the spine of uncertain etiology and for assessment of disease activity in selected patients.
MRI of the spine should cover the part of the spine with the most severe symptoms. For evaluating spondylitis, only the sagittal plane is necessary. The MRI slices should include the entirety of the vertebral bodies being imaged and the facet joints. The spinal lesions that may be seen in patients with axSpA result in either BME observed with the STIR or the T2-weighted sequences with fat suppression, or areas of fatty deposition observed as high-intensity lesions in the T1-weighted sequences. A 2012 consensus of opinion on what spinal lesions on MRI are typical of spondylitis by the ASAS/Outcome Measures in Rheumatology (OMERACT) MRI working group was that a positive spinal MRI for inflammation can be defined as the presence of anterior or posterior spondylitis in at least three sites, as the finding of single vertebral lesions is relatively nonspecific (image 11) . Subsequent studies showed that more than three inflammatory lesions were necessary to provide sufficient specificity and that the presence on MRI of the spine of either at least five inflammatory lesions or at least five fatty lesions resulted in specificity of at least 95 percent . However, the number of lesions seen in the general population also increases with age, and this should be taken into consideration in interpreting vertebral lesions . (See 'MRI of sacroiliac joints' above.)
Imaging of enthesitis — When axSpA is suspected, ultrasound (image 12A-B) or MRI may be used to detect peripheral enthesitis, which may support the diagnosis of axSpA . In patients in whom there is clinical uncertainty regarding whether enthesitis is present, the use of ultrasonography may be of benefit for establishing its presence or absence. The combined use of B-mode and power Doppler ultrasound (PDUS) could reliably identify certain changes consistent with peripheral enthesitis in patients with axSpA with satisfactory interobserver agreement among experts; these changes include findings of hypoechogenicity, increased thickness of the tendon insertion, calcifications, enthesophytes (bone spurs), and Doppler activity [74,75]. In addition to proper general training and the availability of appropriate instrumentation, use of ultrasound for the detection of enthesitis requires specific training in the position of the probe, the setting of the Doppler, and knowledge of the anatomy of the normal blood vessels.
DIAGNOSIS — Axial spondyloarthritis (axSpA) can generally be diagnosed based upon the recognition of a pattern of clinical, laboratory, and imaging findings characteristic of axSpA (see 'Diagnostic approach' below); the likelihood of the diagnosis varies depending upon the specific findings that are present. There is no single historical feature, physical finding, laboratory test, or imaging study with sufficient specificity by itself to establish the diagnosis without the presence of additional abnormalities. Thus, the presence of a combination of features together with the exclusion of other diagnoses that may explain such symptoms or findings is necessary to arrive at an accurate diagnosis .
As an example, the diagnosis of radiographic axSpA (r-axSpA) is very likely in a patient presenting before age 45 with a >3-month history of back pain, positive testing for human leukocyte antigen (HLA)-B27, and obvious erosions and/or fusion of sacroiliac (SI) joints on plain radiographs. The diagnosis of nr-axSpA could be made in a patient with all of these features, in the absence of positive SI joint findings on plain radiography, if typical abnormalities of sacroiliitis were detected by MRI of the SI joints. Other findings can also contribute to making the diagnosis in patients who lack one, or even two, of the aforementioned features; however, the diagnosis is unlikely in the absence of all three of the typical findings (inflammatory back pain [IBP], HLA-B27, and imaging changes of sacroiliitis). An alternative diagnosis is likely in patients lacking any such features and should be excluded in patients with only a few such characteristics.
Diagnostic approach — A thorough medical history (see 'History' above), physical examination (including musculoskeletal examination) (see 'Physical examination' above), and C-reactive protein (CRP) (or, alternatively, an erythrocyte sedimentation rate [ESR]) (see 'Laboratory testing' above) should be obtained in patients suspected of axSpA based upon the presence of one or more features of disease. HLA-B27 testing should be considered in patients suspected of having axSpA, depending upon the level of suspicion of the diagnosis and the utility, availability, and cost of the test in the geographical region and practice setting; the results of such testing should be interpreted in the context of the other features of axSpA that are present. (See 'Laboratory testing' above.)
Other laboratory testing (complete blood count, renal and liver chemistries) may be useful in the evaluation for the presence of potential comorbidities, especially when further treatment is planned, but are usually not required for establishing the diagnosis.
An initial estimate of the likelihood of disease is further based upon whether features suggesting an alternative diagnosis are present; additional testing may also be obtained depending upon the differential diagnosis generated from the presenting features and other abnormalities that may be identified. (See 'Differential diagnosis' below.)
The subsequent diagnostic steps may provide guidance to the combinations of findings most typical of axSpA and unlikely to be associated with another diagnosis, although other explanations as to why a clinical, imaging, or laboratory feature is present should also be excluded (see 'Differential diagnosis' below):
●Step one – In all patients with chronic low back pain (at least three months), with onset before age 45, features characteristic of axSpA should be ascertained during medical history and physical examination:
•Enthesitis (heel pain)
•Inflammatory bowel disease
•Positive response to nonsteroidal antiinflammatory drugs (NSAIDs)
•Family history of SpA
●Step two – All patients with chronic low back pain (at least three months), with onset before age 45, and clinically suspected axSpA should have an anterior-posterior (AP) plain radiograph of the pelvis to examine the sacroiliac joints. We also recommend obtaining CRP. A diagnosis of r-axSpA can be made if, in this patient with chronic back pain, the image meets criteria for sacroiliitis (at least grade 2 bilaterally or grade 3 unilaterally) (see 'Plain radiography' above) and the patient has at least one other SpA-typical parameter.
●Step three – In patients suspected of axSpA who are not positive for sacroiliitis by plain radiography of the pelvis, we recommend MRI of the SI joints for evaluation of evidence of sacroiliitis and HLA-B27 testing. (See 'Laboratory testing' above.)
A diagnosis of nr-axSpA can be made in chronic back pain patients with several typical SpA features present and a positive MRI for sacroiliitis. Nr-axSpA can also be diagnosed in suspected axSpA patients, without abnormalities on imaging, but then HLA-B27 should be positive and again several typical SpA features must be present.
Important additional diagnostic considerations — Both under- and overdiagnosis of axSpA frequently occur, leading to under- or overtreatment with large consequences for individual patients and society. To reduce this risk, the following diagnostic considerations should be taken into account:
●The diagnosis of axSpA is unlikely and an alternative diagnosis is more likely in the absence of all three of the typical findings (IBP, HLA-B27, and imaging changes of sacroiliitis), despite the presence of other clinical features of SpA. Although the presence of numerous SpA features may raise the suspicion of axSpA, this alone is not sufficient to diagnose axSpA. In particular, positive imaging is considered important in making a diagnosis in axSpA .
●In patients with a reading of grade 2 or less for plain radiographic sacroiliitis, there may be considerable disagreement in the reading when evaluated by several separate readers. In patients with such uncertainty, an MRI may be useful in resolving whether sacroiliitis is present.
●Although sacroiliitis on MRI supports the diagnosis of nr-axSpA, it is by itself not diagnostic and should always be interpreted in the context of the patient's symptoms and other findings.
●Some clinicians may prefer to obtain HLA-B27 testing at the initial evaluation rather than waiting for the results of plain radiography. An advantage may be convenience and the avoidance of another needle stick. Disadvantages of this approach include unnecessary cost in some patients. In addition, it is important to remember that the results of such testing alone are neither diagnostic, in the case of positive testing, nor do they exclude the diagnosis, in the case of negative testing, but should be interpreted in the full clinical context. HLA-B27 testing remains most valuable when there is clinical uncertainty regarding the diagnosis of axSpA.
●Patients with an onset of pain after age 45, particularly if other features suggestive of axSpA are present, should not be excluded from evaluation, as up to 4 to 6 percent of patients with axSpA have been shown to first develop chronic back pain beyond 45 years of age .
●Selected patients may benefit from additional imaging studies, such as patients with normal SI joint imaging on plain radiography and MRI but prominent spinal pain and other features of SpA, for whom evidence of spondylitis may support a diagnosis of axSpA rather than another cause of back pain. Similarly, ultrasonography has been used by some experts to document the presence of enthesitis when there is uncertainty regarding whether this finding is present. (See 'MRI of the spine' above and 'Imaging of enthesitis' above.)
Diagnostic uncertainty remains in some patients despite the use of extensive testing. In patients in whom the diagnosis cannot be made with confidence, a definitive conclusion should be avoided, and a "wait-and-see" policy regarding the diagnosis and symptomatic treatment with regular follow-up assessments will usually be the most appropriate approach.
As an example of the uncertainty that may occur, in a 2016 study of 319 axSpA patients seen at 68 rheumatology sites in the United States, the degree of confidence in the diagnosis of each patient was assessed on a scale of 0 to 10 (10 being perfectly confident) by the patient's treating clinician . Only about 50 percent of patients diagnosed with r-axSpA and 30 percent of patients diagnosed with nr-axSpA were diagnosed at a confidence level of 10.
DIFFERENTIAL DIAGNOSIS — Conditions that cause chronic low back and spinal pain may present in a similar fashion to axial spondyloarthritis (axSpA). Certain conditions are more commonly confused with axSpA, although the wide variety of conditions associated with low back pain should be considered in patients presenting with this symptom. The evaluation of the patient with low back pain is discussed in detail separately. (See "Evaluation of low back pain in adults".)
The major entities that should be considered in the differential diagnosis for axSpA include:
●Acute or chronic mechanical nonspecific back pain and inflammatory back pain without SpA – Mechanical back pain or back strain may occur with or without associated detectable anatomic abnormalities or with degenerative disc disease and can mimic axSpA. A minority of patients with these conditions, which are very common, may have a pattern of discomfort that is typical of inflammatory back pain (IBP) (see 'History' above):
•Back strain without demonstrated anatomic change – Acute back strain or lumbago, also termed musculoligamentous lumbar sprain or strain, is the most common cause of low back pain in the general population. It may occur in young adults in the same age range as axSpA but can usually be easily distinguished from axSpA by its acuity, short duration, and frequent association with or attribution to a specific injury or physical strain. (See "Evaluation of low back pain in adults".)
•Acute or chronic mechanical low back pain with associated anatomic abnormalities – Multiple anatomic lesions have been associated with acute and chronic back pain, including disc injuries, discogenic pain syndrome, facet syndrome, radiculopathy, acute bony injuries, spondylolisthesis, spondylosis, pars articularis injuries, sacroiliac (SI) joint injuries, and lumbar spinal stenosis. These are generally able to be readily distinguished from axSpA by the absence of inflammatory characteristics in the majority of patients together with not being associated with the other features of SpA. (See 'History' above and 'Physical examination' above and "Evaluation of low back pain in adults".)
●Fibromyalgia and myalgia – Fibromyalgia may be associated with back pain, as a component of chronic widespread pain, and painful tender points may mimic some of the findings of enthesitis. However, the pattern of pain is not typical of IBP, fibromyalgia tender points can usually be distinguished from enthesitis with a thorough musculoskeletal examination, and patients with fibromyalgia lack the imaging changes or elevated acute phase response of SpA. In addition, a response to nonsteroidal antiinflammatory drugs (NSAIDs) is infrequently seen or only modest when present, unlike the good response typical of many patients with axSpA. On the other hand, fibromyalgia can coexist with axSpA. In 2020, a meta-analysis reported a 16.4 percent prevalence of fibromyalgia in patients with axSpA . (See "Clinical manifestations and diagnosis of fibromyalgia in adults".)
●Diffuse idiopathic skeletal hyperostosis – Diffuse idiopathetic skeletal hyperostosis (DISH) is commonly confused with axSpA because the radiographic appearance of the spine may resemble changes seen in SpA, with large osteophytes. However, unlike patients with axSpA, the SI joints are usually spared, the apophyseal joints are not ankylosed, exuberant osteophyte formation and flowing calcification occur, at least four contiguous vertebral bodies are typically ossified, and disc height is maintained. There may also be extraspinal involvement with hyperostosis noted in the region of the olecranon, patella, calcaneus, shoulder, and acetabulum. (See "Diffuse idiopathic skeletal hyperostosis (DISH)".)
●Vertebral compression fracture – This typically occurs in patients with low bone mass of the spine, especially with osteoporosis or trauma. Patients are typically older than those with SpA, and the pain is usually more acute and more severe than in SpA and is localized to the spine. Occasionally, the pain may become chronic, but it generally remains localized even when it does not resolve in a matter of weeks. (See "Osteoporotic thoracolumbar vertebral compression fractures: Clinical manifestations and treatment".)
●Sacroiliac joint infection – Infection of the SI joint may mimic inflammatory noninfectious sacroiliitis. However, SI joint infection is typically unilateral, severe, and often associated with other signs of infection. The inflammation response seen on MRI often spreads to involve the surrounding soft tissue. A history of chronic pain typical of axSpA is usually absent. Infection can usually be documented by imaging-guided joint aspiration and culture of the infection. (See "Septic arthritis in adults", section on 'Diagnosis'.)
●Osteitis condensans ilii – This is usually a radiological diagnosis noted incidentally or during an evaluation of possibly unrelated low back or pelvic pain [80,81]. It is sometimes confused with sacroiliitis. Radiographic findings show a typical pattern of sclerosis that is often confined to the iliac side of the SI joint but can also be found in other parts. The SI joints do not show erosions or fusion, but patients, especially women who have given birth, might have chronic back pain for mechanical reasons. Subchondral bone marrow edema (BME) may be present, especially in early disease, but can also be seen intermittently later on .
●Insufficiency bone fractures – The common symptoms of insufficiency fractures of the pelvis are acute or chronic buttock and hip pain, which may also be experienced by patients with SpA. Imaging studies can distinguish these causes. On MRI, lesions appear as BME osteitis, but T1-weighted sequences can show a hypointense fine line, which may be confirmed by CT .
●Erosive osteochondrosis and Schmorl's nodes – Erosive osteochondritis (or lumbar erosive osteochondrosis), which occurs as a consequence of degenerative disc disease, may cause back pain along with changes on imaging of the lumbar spine that resemble spondylodiscitis in patients with SpA. Erosive osteochondrosis can be distinguished from axSpA by characteristic MRI findings, the lack of other features typical of SpA, and normal imaging findings of the SI joints. Another entity that may be seen on radiography, Schmorl's nodes, is usually asymptomatic and would similarly lack other SpA features and is not associated with SI joint abnormalities .
●Familial Mediterranean fever – Patients with familial Mediterranean fever (FMF), especially those with more severe disease, may develop symptoms of back pain and imaging changes consistent with sacroiliitis [85,86], as well as symptoms and findings of peripheral arthritis and of enthesitis (Achilles tendinopathy and plantar fasciitis) at the ankle, including symptoms of calf and ankle pain with activity . Sacroiliitis may be more common in patients with FMF who are human leukocyte antigen (HLA)-B27-positive and/or M6794V-positive within some populations, but many of these latter patients are HLA-B27-negative [85,86,88,89]. These manifestations of FMF can be distinguished from other patients with SpA clinically based upon their history of FMF.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Spondyloarthritis".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Ankylosing spondylitis (The Basics)")
●Beyond the Basics topics (see "Patient education: Axial spondyloarthritis, including ankylosing spondylitis (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Nomenclature - There are two forms of spondyloarthritis (SpA) with predominantly axial involvement, collectively termed axial SpA (axSpA). axSpA is composed of radiographic axSpA (r-axSpA, also termed ankylosing spondylitis [AS]), typically with radiographic sacroiliitis on plain radiography, and axSpA without plain radiographic changes of sacroiliitis, which is termed nonradiographic axSpA (nr-axSpA). (See 'Nomenclature and classification' above.)
●Diagnosis - axSpA can generally be diagnosed based upon the recognition of a pattern of clinical, laboratory, and imaging findings characteristic of axSpA, including chronic back pain, sacroiliitis on imaging, and the presence of human leukocyte antigen (HLA)-B27 and other features; the likelihood of the diagnosis varies depending upon the specific findings that are present.
•There is no single historical feature, physical finding, laboratory test, or imaging study with sufficient specificity by itself to establish the diagnosis without the presence of additional abnormalities. (See 'Evaluation of suspected axial spondyloarthritis' above and 'History' above and 'Physical examination' above and 'Laboratory testing' above and 'Imaging studies' above and 'Diagnosis' above.)
•In all patients with chronic low back pain (at least three months) with onset before age 45, features characteristic of axSpA should be ascertained during medical history and physical examination:
-Inflammatory back pain (IBP)
-Enthesitis (heel pain)
-Inflammatory bowel disease
-Positive response to nonsteroidal antiinflammatory drugs (NSAIDs)
-Family history of SpA
•All patients with chronic back pain (at least three months) with onset before age 45 and clinically suspected axSpA should have an anterior-posterior (AP) plain radiograph of the pelvis to examine the sacroiliac (SI) joints. We also recommend obtaining C-reactive protein (CRP). A diagnosis of r-axSpA can be made in such patients if the image meets criteria for sacroiliitis (at least grade 2 bilaterally or grade 3 unilaterally) and at least one additional characteristic feature of SpA is positive in the assessment of the patient. (See 'Plain radiography' above and 'Diagnostic approach' above.)
•In patients suspected of axSpA who are not positive for sacroiliitis by plain radiography of the pelvis, we recommend MRI of the SI joints for evaluation of evidence of sacroiliitis and HLA-B27 testing (see 'Laboratory testing' above). A diagnosis of nr-axSpA can be made in chronic back pain patients with several typical SpA features present and a positive MRI for sacroiliitis. Nr-axSpA can also be diagnosed in suspected axSpA patients without abnormalities on imaging, but then HLA B27 should be positive and again several typical SpA features must be present. (see 'Diagnostic approach' above and 'Important additional diagnostic considerations' above and 'Differential diagnosis' above)
•A diagnosis of axSpA is less likely in patients who have negative plain radiography, as well as negative MRI of the SI joints and negative testing for HLA-B27.
•In patients with several characteristic findings of axSpA, the diagnosis should not be excluded solely based upon either a negative test for HLA-B27, onset of symptoms after age 45, or back pain for less than three months. Arriving at a diagnosis in some patients may require a longer period of follow-up care and monitoring, including additional imaging investigation. (See 'Important additional diagnostic considerations' above.)
●Differential diagnosis - The differential diagnosis of SpA includes acute or chronic mechanical back pain, fibromyalgia, diffuse idiopathic skeletal hyperostosis (DISH), vertebral compression fracture, SI joint infection, osteitis condensans ilii, and erosive osteochondrosis and Schmorl's nodes. These conditions and the wide variety of conditions associated with low back pain must be considered in patients presenting with suspected SpA. (See 'Differential diagnosis' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David Yu, MD, who contributed to earlier versions of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟