Polymyalgia rheumatica (PMR) is a chronic inflammatory condition that predominantly involves large joints and periarticular structures. Following rheumatoid arthritis, PMR is the second most common autoimmune rheumatic disease, with a lifetime risk of approximately 2%.1
PMR affects adults over the age of 50 years and becomes more frequent with advancing age, peaking among individuals aged 70 to 80 years. Additionally, women are affected 1.5 to 2 times more often than men.
Although the etiology and pathogenesis of PMR are not known, epidemiologic studies suggest that there are both genetic and environmental factors involved in the development of this disease.
This is further evidenced by a geographic variance in the incidence of PMR, with the highest rates seen among inhabitants of and descendants from Northern European countries.2 While PMR has been observed in people of Hispanic, Asian, and African heritage, incidence rates among these ethnicities are notably lower.
The hallmark clinical picture of PMR is characterized by pain and stiffness (Table 1). Stiffness typically lasts greater than 30 minutes and is worse after rest or inactivity. Patients often describe difficulty getting dressed or discomfort when turning in bed at night that interferes with sleep. Shoulder range of motion may be limited, causing difficulty in performing activities at or above head level.
Compared to symptoms of non-inflammatory conditions, symptoms of PMR typically are symmetrical and get better with activity. In clinical practice, it can be challenging to distinguish PMR from elderly-onset rheumatoid arthritis. Indeed, inflammatory arthritis of peripheral joints (eg, wrists, knees) can be seen in up to one-third of patients with PMR.3 However, synovitis and erosions of distal joints, including the metacarpophalangeal and proximal interphalangeal joints and the feet, are more indicative of rheumatoid arthritis than PMR.4
Weakness of the proximal muscles is not a feature of PMR and its presence should prompt evaluation for a myopathic process. Similarly, diffuse muscle tenderness is not a prominent feature and should raise suspicion for electrolyte abnormalities, metabolic disturbances, or fibromyalgia.
Constitutional symptoms of low-grade fever, malaise, fatigue, anorexia, and weight loss can occur in 40% to 50% of patients with PMR.5 These non-specific symptoms often lead clinicians to consider evaluating for other diagnoses, such as infection or malignancy.
It is well recognized that PMR is associated with giant cell arteritis (GCA), a systemic vasculitis affecting the large arteries that can cause blindness, stroke, and aortic aneurysm. Although 40% to 60% of patients with GCA have associated PMR symptoms at diagnosis,6 only 10% to 20% of patients with isolated PMR will develop symptoms of GCA.
Therefore, when evaluating patients for PMR, health care providers also should evaluate for symptoms of GCA—headache, scalp tenderness, jaw claudication, diplopia, vision loss, and upper or lower extremity claudication. If any symptoms are present, testing for GCA should be undertaken, including the consideration of temporal artery biopsy.
Because of the broad differential diagnosis and potential mimicking disease states (Table 2), clinicians evaluating patients for PMR should obtain a detailed history and conduct a comprehensive physical examination, focusing particularly on the musculoskeletal, vascular, and neurologic systems.
Laboratory parameters in PMR are not specific but typically show evidence of a systemic inflammatory state. Such abnormalities can include a mild normocytic anemia due to chronic inflammation, leukocytosis, and thrombocytosis. Inflammatory markers (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP]) frequently are elevated. The ESR, however, can be normal in 6% to 20% of patients with PMR.7,8 Therefore, CRP may be a more sensitive marker of inflammation in these patients.9
Rheumatoid factor (RF) and anticitrullinated protein antibodies (ACPA) typically are absent, and their presence should raise suspicion for rheumatoid arthritis. Additional laboratory studies that are useful in the differential diagnosis include liver transaminases, creatine kinase, calcium, sodium, potassium, magnesium, creatinine, thyroid-stimulating hormone, serum protein electrophoresis, and urinalysis with microscopy.
Several sets of diagnostic criteria for PMR have been proposed, but they have not been validated or universally accepted. Common features among these criteria have included a minimum age (50–65 years), bilateral shoulder girdle and hip girdle aching, morning stiffness, and elevated inflammatory markers.
In response to a lack of standardized criteria, the European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) recently developed a set of classification criteria based on expert consensus and evaluation of a prospective cohort of 125 patients with new-onset PMR and 169 patients with disorders mimicking PMR (Table 3).4 Use of this classification scoring algorithm in patients who are at least 50 years of age and have new-onset bilateral shoulder pain and elevated inflammatory markers demonstrated a sensitivity of 68% and a specificity of 78% for the diagnosis of PMR if at least 4 clinical criteria points were present.
In recent years, there has been an increased use of imaging modalities to evaluate patients with suspected PMR. Ultrasonography and magnetic resonance imaging (MRI) commonly have identified abnormalities in large joints and periarticular structures. Characteristic findings in patients with PMR may include bicipital tenosynovitis, subacromial/subdeltoid bursitis (Figure 1), greater trochanteric bursitis (Figure 2), and glenohumeral (Figure 3) or hip joint synovitis. Although these findings also can be seen in other forms of inflammatory arthritis, identification of these abnormalities can be helpful to differentiate PMR from other conditions.
Indeed, the addition of ultrasonography to the EULAR/ACR classification scoring algorithm improved the specificity to 81% for differentiating PMR from non-PMR patients and to 89% in discriminating PMR from other shoulder disorders.4 Compared to ultrasound, MRI can additionally show extracapsular soft-tissue edema in the shoulders and cervical interspinous bursitis in the neck, but its utility is limited by cost and availability, making ultrasound the preferred modality to evaluate musculoskeletal structures in these patients.
Fluorodeoxyglucose-positron emission tomography (FDG-PET) scanning, although not part of routine practice, also has been used to evaluate patients with active PMR. Inflammatory activity can be seen in the shoulders, hips, and interspinous processes in patients with PMR.3 Although PET does not provide detailed images of articular structures, this modality can identify subclinical large-vessel inflammation in patients without clinical signs or symptoms of GCA (Figure 4). Indeed approximately one-third of patients with isolated PMR show evidence of low-grade large arterial (aorta, subclavian, axillary) inflammation on PET scans.10
Vascular ultrasonography of the upper extremity arteries may provide a more practical and cost-effective alternative for screening patients with PMR for occult large-vessel vasculitis. Ultrasound and Doppler evaluation of the subclavian, axillary, and proximal brachial arteries can readily detect signs of vascular inflammation including homogenous wall swelling, stenosis, or vascular occlusion.11
Although temporal artery biopsies are not considered necessary in patients without concurrent symptoms of GCA, ultrasound of the temporal arteries can detect vasculitis of the temporal arteries in approximately 8% of patients presenting with isolated PMR.12 Given that, some experts recommend including ultrasonography of the temporal and upper-extremity arteries in all patients presenting with PMR to exclude concurrent GCA.13
In routine clinical practice, however, only patients with symptoms of vasculitis or those who do not respond to standard treatment would be considered for vascular imaging studies.
Because of the wide variations in clinical management of PMR and the lack of standardized recommendations for patients with this condition, an international collaborative initiative coordinated by EULAR and ACR recently established management guidelines for this condition based on available clinical evidence and expert opinion.14 The collective principles from these guidelines are outlined below.
Glucocorticoids (GC) remain the standard treatment for PMR. Treatment for PMR should be individualized for each patient and a minimum effective single daily glucocorticoid dose (typical range, 12.5-25 mg prednisone or equivalent) should be used. Patients at high risk for GC-associated side effects, such as those with pre-existing comorbidities (glaucoma, osteoporosis, diabetes, hypertension, etc) should be started at the lower end of the dosing range. However, the use of 7.5 mg or less of prednisone daily for initial dosing is discouraged. Initial treatment with more than 30 mg of prednisone daily also is not recommended for PMR unless there is high suspicion for concurrent GCA.14
The majority of patients with PMR will have a marked clinical response within 7 days and normalization of inflammatory markers within 4 weeks. Rapid resolution of symptoms with low-dose prednisone has been considered a cardinal feature of PMR, but studies have found that 29% to 45% of patients may not reach full clinical remission by 3 to 4 weeks.4,15 Thus, atypical clinical presentations and/or failure to respond to appropriate initial GC therapy should prompt consideration of alternative pathologies and specialist referral.
For patients with an appropriate response to therapy, tapering to a dosage of 10 mg per day should occur within 4 to 8 weeks after GC initiation (eg, reduction by 2.5 mg/d every 2-4 weeks until 10 mg/d is reached). Further tapering can occur by 1-mg decrements every 4 weeks until discontinuation if remission is maintained. A treatment duration of 1 to 3 years should be expected.
Approximately 50% of patients with PMR will experience at least one relapse during the course of treatment. If relapse symptoms are consistent with clinical features of PMR, then an increase to the pre-relapse dose followed by gradual reduction over 4 to 8 weeks to the dose at which relapse occurred is recommended. If relapse symptoms include features of GCA, then high-dose steroids (40-60 mg/d) should be instituted and evaluation for this condition undertaken.
Due to significant morbidity associated with prolonged exposure to GCs, adjunct immunosuppressive medications have been evaluated to assess their GC-sparing effects in PMR. While disease-modifying antirheumatic drugs (DMARDS) have been successful in the treatment of other inflammatory conditions, insufficient evidence is available to suggest that hydroxychloroquine (Plaquenil, others), azathioprine (Azasan, Imuran, others), or leflunomide (Arava, others) are beneficial in the treatment of PMR. Methotrexate (Otrexup, Rasuvo, others, 7.5-10 mg/week) has been evaluated in 3 prospective randomized studies with variable results regarding steroid-sparing effect and reduction in disease relapse risk.16-18 In clinical practice, methotrexate is not initiated routinely at disease onset but can be considered in patients at high risk for developing GC-associated adverse events, or in patients experiencing 2 or more relapses.19
Despite an expanding field of targeted biologic therapies available for the treatment of inflammatory disorders, no biologic agents are recommended for the treatment of PMR. Infliximab (Remicade, others), a tumor-necrosis factor alpha inhibitor, did not show any benefit in newly diagnosed patients with PMR in a single prospective randomized trial.20 The anti-IL-17 agent secukinumab (Cosentyx) and the anti-IL-1β agent canakinumab (Ilaris) failed to impact PMR disease activity to the same degree as prednisone in a 3-arm proof of concept study, leading to study termination (NCT01364389). Tocilizumab (Actemra), a humanized monoclonal IgG antibody against the IL-6 receptor, has shown both biologic plausibility and clinical promise in small numbers of patients and is under investigation in an open-label Phase II study (NCT01396317).
Although mortality is not increased among patients with PMR compared to the general population, exposure to GCs does increase the risk of morbidity due to GC-associated adverse events. Patients with PMR are 2 to 5 times more likely to develop vertebral and hip fractures. Given this, the standard of care should include consideration of bone protection from GC-induced osteoporosis. Vitamin D and calcium supplementation should be provided for all patients unless a contraindication exists. Bisphosphonates should be added for high-risk patients (age >65 years, prior fragility fracture, documented osteoporosis) and considered in patients with T-scores of -1.5 or lower.14
PMR is a chronic, but treatable, inflammatory syndrome that affects patients aged 50 years or older. Recent classification criteria assist in the differentiation of PMR from several possible mimics of this disease. Advanced imaging can aid in diagnosis, and ultrasonography appears to be the most suitable and cost-effective modality. Patients with PMR should be followed longitudinally because some are eventually reclassified as having rheumatoid arthritis or GCA. Disease- and treatment-related morbidity remains significant; however, mortality is not increased in patients with PMR. GCs are the mainstay of treatment, but the ongoing evaluation of biologic agents may provide alternative approaches to management in the near future.
Editor’s Note: The authors’ have also written an overview of diagnosis and treatment of polymyalgia rheumatic for your patients.