Polymyalgia rheumatica (PMR) is the most common inflammatory rheumatic disease of the elderly and represents the most common indication for long-term corticosteroid therapy in community-based patients (1). There are wide variations of clinical practice in the treatment of PMR, because the disease may be managed in primary or secondary care by general practitioners, rheumatologists, and nonrheumatologists (2). Researchers have debated the issues of guidelines for the diagnosis and management of PMR since the 1970s (3).

Many features of PMR predispose the unwary clinician to diagnostic errors (4). The main symptom of PMR, proximal pain and stiffness syndrome, can occur in many other illnesses. Patients often have systemic symptoms such as fever, anorexia, and weight loss, and distal musculoskeletal manifestations such as peripheral arthritis, distal swelling with pitting edema, and carpal tunnel syndrome. PMR is also associated with giant cell arteritis (GCA) in 10–20% of cases. The acute phase response used in diagnosing PMR can occur in other settings such as neoplasia, rheumatoid arthritis, and infection.

PMR and GCA are regarded as indications for the prompt initiation of corticosteroid therapy. Many clinicians use a response to corticosteroids as a defining feature of these conditions. This may encourage diagnostic error since corticosteroids are potent antiinflammatory agents that can mask symptoms of a host of serious conditions including osteoarthritis, rotator cuff problems, rheumatoid arthritis, cancer, infection, migraine, and intracranial tumors, especially if used in high doses and for protracted lengths of time.

Is PMR underdiagnosed or overdiagnosed? Studies of conditions that mimick PMR, and studies reporting revised diagnoses in long-term followup of patients with an initial PMR diagnosis would suggest overdiagnosis. The main conditions that can mimick PMR are malignancies, seronegative arthritides, and other connective tissue diseases; however, myelodysplastic syndromes, Lyme disease, bacterial endocarditis, quinidine-induced rheumatic syndromes, infection, hemangioma, C5 radiculopathy, and multiple myeloma have also been reported to mimick PMR (4). Other studies have shown that there is a lack of recognition of PMR, particularly in primary care, and suggest underdiagnosis (5).

Diagnostic criteria in PMR

  1. Top of page
  2. Diagnostic criteria in PMR
  3. A step-wise process for the diagnosis of PMR
  4. A look toward the future

There are several sets of diagnostic criteria for PMR (Table 1). Three criteria are based on clinical experience (6–8) and 1 is based on evaluating the performance of alternative criteria (5). The Bird et al criteria (3) are based on the presence of at least 3 of 7 criteria that were found to give the highest combined sensitivity and specificity. The criteria presented by Chuang et al (6), and Jones and Hazleman (8) require all criteria to be present for a diagnosis to occur, and Healey's criteria (7) require some, but not necessarily all criteria to be present for a diagnosis of PMR.

Table 1. Comparison of diagnostic criteria for polymyalgia rheumatica*
CriteriaChuang et al (1982)Healey (1984)Bird et al (1979)Jones and Hazleman (1981)
  • *

    ✓= criteria appears in criteria set; ESR = erythrocyte sedimentation rate; CRP = C-reactive protein level.

  • With borderline elevated ESR, the presence of systemic symptoms, history of giant cell arteritis, and a rapid response to low-dose corticosteroids can be used to support diagnosis.

  • Response to corticosteroids can be used to confirm the diagnosis.

I. Age✓≥50 years✓>50 years✓>65 years 
II. Bilateral pain/aching of neck, shoulders, and pelvic girdle✓Any 2✓Any✓Shoulder✓Shoulder & pelvic girdle
III. Morning stiffness✓≥30 minutes✓>1 hour✓>1 hour
IV. Duration of symptoms✓≥1 month✓≥1 month ✓≥2 months unless treated
V. Onset of symptoms  ✓<2 weeks 
VI. ESR✓>40 mm/hour✓Elevated✓≥40 mm/hour✓>30 mm/hour or CRP >6 mg/liter
VII. Depression and/or weight loss   
VIII. Bilateral upper arm tenderness   
IX. Exclusion of other diagnosis 
X. Rapid response to corticosteroids ✓≤20 mg prednisone 
DiagnosisAll above requiredCriteria I and IV, plus any 3 from II, III, IV, or XAny 3 criteria, or any 1 plus positive temporal artery biopsyAll above required


Two sets of criteria agree that PMR does not occur prior to the age of 50 years. The Bird et al criteria use a higher threshold (65 years), but will still allow a PMR diagnosis in younger patients.

Duration and mode of onset

Only the Bird et al criteria consider an acute onset of disease useful for diagnosis. The Healey and the Chuang et al criteria sets both emphasize that the myalgia needs to be persistent (>1 month), presumably to exclude all other causes of self-limited myalgia. The Jones and Hazleman criteria stipulation that the myalgia needs to be present for at least 2 months is probably no longer acceptable in terms of patient expectations and the need for prompt treatment.

Shoulder, hip, and neck pain and other clinical features

All of the criteria sets include the presence of proximal pain and stiffness, although a PMR diagnosis is still possible without these features in both the Bird et al and the Healey criteria. Three sets of criteria recognize the need to exclude other competing conditions before the diagnosis of PMR is considered. Clearly there is often confusion in the assessment of clinicians as to whether PMR is a syndrome such as back pain or a specific disease. Only the Bird et al criteria add other features such as depression, weight loss, and arm tenderness.

Elevated acute phase response and laboratory tests

All criteria include an elevated acute phase response, the erythrocyte sedimentation rate (ESR), with the C-reactive protein (CRP) level often being used as an adjunct to or instead of the ESR. There is disagreement on the cut-off levels for these measures. Use of lower levels may result in overdiagnoses, but reports suggest that CRP levels and the ESR are within normal ranges in 10–20% of cases. It is still unclear whether CRP level or the ESR is the better measure to use for diagnosis and for monitoring treatment response. The use of low circulating CD8+ T cells and elevated interleukin-6 (IL-6) levels have also been considered as a measure of acute phase response in PMR (9).

Response to corticosteroids

Both the Jones and Hazleman and the Healey criteria include a rapid response to corticosteriods. The other sets of criteria refer to corticosteroid response; however, none define what constitutes a response.

Role of imaging

There is no consensus concerning the role of imaging for suspected PMR. Magnetic resonance imaging shows that bilateral subacromial and subdeltoid bursitis (present in all patients with active PMR), glenohumeral joint synovitis, and long head biceps tenosynovitis may be useful in diagnosing PMR. Ultrasonography results shows evidence of bilateral subacromial/subdeltoid bursitis in almost all patients with PMR.

A step-wise process for the diagnosis of PMR

  1. Top of page
  2. Diagnostic criteria in PMR
  3. A step-wise process for the diagnosis of PMR
  4. A look toward the future

Should the criteria be applied as a phased process to reflect the real-life diagnostic process in suspected PMR? For purposes of classification, and to be useful in the clinic, it may be helpful to apply the criteria in the following step-wise process: evaluate for inclusion criteria, evaluate for exclusion criteria, prescribe corticosteroids and evaluate response, confirm diagnosis on followup.

Evaluate for inclusion criteria

The existing criteria sets combine individual criteria in different ways. We may need to differentiate core symptoms (such as proximal pain and stiffness) that are essential to the diagnosis from other features (such as acute phase response) that may increase the likelihood of a PMR diagnosis but may not be present in all cases. Are there conditions such as GCA and rheumatoid arthritis that can coexist with the diagnosis of PMR? There is confusion about whether GCA with polymyalgia should be labeled the same way as PMR without GCA. The signs and symptoms of PMR may also occur in patients with GCA, although there appear to be differences in immunogenetic markers (HLA–DRB1) and polymorphisms (RANTES, tumor necrosis factor a microsatellites, and IL-6 promoter) between PMR and GCA (9). There are no clinical or routine laboratory features that will allow early differentiation of polymyalgic late-onset seronegative rheumatoid arthritis from PMR (10). Long-term followup for persistent synovitis and erosive changes is often needed to distinguish these conditions. Further investigation of anti–citrullinated peptide antibody testing in PMR may help to answer the diagnostic question (11).

Evaluate for exclusion criteria

What are the mimicking conditions that cannot coexist with PMR and should rule out a PMR diagnosis?

Prescribe corticosteroids and evaluate response

What dose of corticosteroids was prescribed and what route of administration was used? What level of response is required, and over what period of time? A small response encourages diagnostic error because corticosteroids are potent antiinflammatory agents that mask symptoms from a host of serious conditions including osteoarthritis, rotator cuff problems, rheumatoid arthritis, cancer, and infection. Other unanswered questions include: how many patients with proximal pain and stiffness respond to corticosteroids even though we do not think they have PMR? Can a patient have PMR even if they do not respond to corticosteroids at a specified dose?

Confirmation of the diagnosis on followup

Many experienced clinicians feel that evaluating the symptoms and signs (e.g., persistent synovitis) and evaluating pertinent investigations (e.g., erosions on radiographs) may be important steps in the diagnostic process.

There is a trade-off between sensitivity and specificity of criteria. Should we err on the side of overdiagnosing or underdiagnosing PMR? Criteria that lack sensitivity will mean PMR, and its significant morbidity, going untreated—to date we know little about quality of life in PMR. However, a lack of specificity will lead to more incorrect diagnoses, side effects from corticosteroids, and failing to appropriately treat serious conditions such as cancer, infection, and other inflammatory arthropathies and connective tissue diseases.

We need to recognize that PMR is a disease that is treated in both primary and secondary care settings by rheumatologists, primary care physicians, and other nonrheumatologist physicians in secondary care. Should criteria reflect who makes the diagnosis and treats the patient? The choice of criteria set also needs to consider the availability and costs of laboratory tests and additional studies (e.g., imaging studies). Finally, patient acceptability must be a consideration with respect to the risks and benefits of additional tests and delays in treatment.

A look toward the future

  1. Top of page
  2. Diagnostic criteria in PMR
  3. A step-wise process for the diagnosis of PMR
  4. A look toward the future

In the face of these open questions, there is a need to develop diagnostic and classification criteria in PMR. This will be a complex process involving review of existing evidence and eliciting expert opinion using formal consensus methods, followed by further research.

The statement for reporting studies of diagnostic accuracy (STARD) was developed in response to reports on the low methodologic quality of many studies of diagnostic tests (12). If we use the STARD standards to interpret and assess the strength of the evidence in PMR, we may find a lack of adequate evidence. This is particularly difficult in PMR due to the lack of a gold-standard diagnostic test that is independent of any criteria being evaluated.

The process of addressing a new classification system for PMR began at the Third International Conference on PMR and GCA, held at St John's College, Cambridge, UK in July 2005. A multidisciplinary group including rheumatologists, general practitioners, and nonrheumatologist secondary care physicians as well as statisticians and methodologists met to identify an initial list of potentially relevant indications for diagnostic and classification criteria in PMR. This effort is in response to the need for such criteria in the scientific community in order to better study this disease, and directly in response to a call from the American College of Rheumatology for development of classification criteria in PMR. The international working group has reviewed existing literature, criteria, and a core set of consensus criteria that will be tested in a wider Delphi survey in order to develop potential draft criteria that can be evaluated in a prospective study. Future studies, including studies of imaging modalities such as musculoskeletal ultrasound, and genomic and proteomic approaches will be helpful to establish more definitive diagnostic criteria once these classification criteria are in place.


  1. Top of page
  2. Diagnostic criteria in PMR
  3. A step-wise process for the diagnosis of PMR
  4. A look toward the future
  • 1
    Walsh LJ, Wong CA, Pringle M, Tattersfield AE. Use of oral corticosteroids in the community and the prevention of secondary osteoporosis: a cross sectional study. BMJ 1996; 313: 3446.
  • 2
    Chakravarty K, Elgabani SHS, Scott DGI, Merry P. A district audit on the management of polymyalgia rheumatica and giant cell arteritis. Br J Rheumatol 1994; 33: 1526.
  • 3
    Bird HA, Esselinckx W, Dixon ASJ, Mowat AG, Wood PHN. An evaluation of criteria for polymyalgia rheumatica. Ann Rheum Dis 1979; 38: 4349.
  • 4
    Brooks RC, McGee SR. Diagnostic dilemmas in polymyalgia rheumatica. Arch Intern Med 1997; 157: 1628.
  • 5
    Blaauw AA, Schuwirth LW, van der Vleuten CP, Smits F, van der Linden S. Assessing clinical competence: recognition of case descriptions of rheumatic diseases by general practitioners. Br J Rheumatol 1995; 34: 3759.
  • 6
    Chuang TY, Hunder GG, Ilstrup DM, Kirkland LT. Polymyalgia rheumatica: a 10 year epidemiologic and clinical study. Ann Intern Med 1982; 97: 67280.
  • 7
    Healey LA. Long-term follow-up of polymyalgia rheumatica: evidence for synovitis. Semin Arthritis Rheum 1984; 13: 3228.
  • 8
    Jones JG, Hazleman BL. Prognosis and management of polymyalgia rheumatica. Ann Rheum Dis 1981; 40: 15.
  • 9
    Dasgupta B, Kalke S. Polymyalgia rheumatica. In: IsenbergD, MaddisonP, WooP, GlassD, BreedveldFC, editors. Oxford textbook of rheumatology. 3rd ed. Oxford: Oxford University Press; 2004. p. 97783.
  • 10
    Caporali R, Montecucco C, Epis O, Bubbio Pallavicini F, Maio T, Cimmino MA. Presenting features of polymyalgia rheumatica (PMR) and rheumatoid arthritis with PMR like onset: a prospective study. Ann Rheum Dis 2001; 60: 10214.
  • 11
    Lopez-Hoyos M, Ruiz de Alegria C, Blanco R, Crespo J, Pena M, Rodriguez-Valverde V, et al. Clinical utility of anti-CCP antibodies in the differential diagnosis of elderly-onset rheumatoid arthritis and polymyalgia rheumatica. Rheumatology (Oxford) 2004; 43: 6557.
  • 12
    Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Ann Intern Med 2003; 138: 404.