Polymyalgia rheumatica (PMR) is a relatively common chronic inflammatory condition of unknown etiology that affects elderly individuals. It is characterized by proximal myalgia of the hip and shoulder girdles with accompanying morning stiffness that lasts for more than 1 hour.[1] Approximately 15% of patients with PMR develop giant cell arteritis (GCA), and 40-50% of patients with GCA have associated PMR. Despite the similarities of age at onset and some of the clinical manifestations, the relationship between GCA and PMR is not yet clearly established.[2]
PMR is a clinical diagnosis based on the complex of presenting symptoms and the exclusion of the other potential diseases (see Presentation and Workup). Corticosteroids are considered the treatment of choice, and a rapid response to low-dose corticosteroids is considered pathognomonic. Patients who are at risk for relapse, have steroid-related adverse effects, or need prolonged steroid therapy may benefit from the addition of methotrexate or tocilizumab.[3] (See Treatment.)
Patients have an excellent prognosis. Exacerbations may occur if steroids are tapered too rapidly, however, and relapse is common.
The cause of PMR is unknown. PMR is closely linked to giant cell arteritis (GCA, temporal arteritis), although it is controversial whether GCA and PMR are two separate diseases or part of the same spectrum of disease.[4] One hypothesis is that in a genetically predisposed patient, an environmental factor, possibly a virus, causes monocyte activation, which helps determine the production of cytokines that induce manifestations characteristic of PMR and GCA. However, although several infectious agents have been investigated as possible triggers, results are inconclusive.[5]
Immunogenetic studies support a polygenic basis for GCA and PMR. Occurrence in siblings and increased prevalence in those of Northern European heritage suggest a genetic role in the pathophysiology of the disease. Although most studies confirm an association between HLA-DRB1*04 alleles and GCA, the strength of this association with PMR varies between different populations. Interleukin (IL)–1 and tumor necrosis factor–alpha (TNF-α) gene polymorphisms have weak association with GCA and PMR. In Spain, an IL-6 polymorphism was associated with the expression of PMR symptoms in GCA patients. Additionally, in this Spanish population, the RANTES polymorphism was associated with PMR and not GCA.[6]
Pathologically, GCA and PMR are similar, except that significant vascular involvement does not occur in pure PMR. Synovitis, bursitis, and tenosynovitis around the joints, especially the shoulders, hips, knees, metacarpal phalangeal joints, and wrists, are seen in PMR. Inflammation is thought to start within the synovium and bursae, with recognition of an unknown antigen by dendritic cells or macrophages.[7]
Systemic macrophage and T-cell activation are characteristic of both GCA and PMR. Patients often have an elevated IL-6 level, which is likely responsible for the systemic inflammatory response in both GCA and PMR. Most studies in PMR show that a decrease in the level of circulating IL-6 correlates with remission of clinical symptoms. Data on other circulating cytokines (eg, IL-1, IL-2, TNF-α, IL-10) are too scant to draw any conclusions. However, studies do show that interferon-gamma (IFN-γ) is expressed in nearly 70% of temporal artery biopsy samples from patients with GCA but is not detected in patients with isolated PMR, suggesting IFN-γ may be crucial to the development of GCA.[5, 6, 8]
Although PMR causes severe pain and stiffness in the proximal muscle groups, no evidence of disease is present on muscle biopsy. Muscle strength and electromyographic findings are normal. Instead, the inflammation is at the level of the synovium and bursae, with MRI studies revealing periarticular inflammation as well as bursitis in the bursae associated with both the shoulder and hip girdles.[9, 10]
Some evidence suggests the presence of cell-mediated injury to the elastic lamina in the blood vessels in the affected muscle groups. A prospective study of 35 patients with isolated PMR noted vascular (18F) fluorodeoxyglucose positron emission tomography (FDG-PET) imaging at diagnosis in 31% of patients, predominantly at the subclavian arteries, but at a much lower intensity than in GCA patients. Increased FDG uptake in the shoulders was seen in 95% of the patients, in the hips in 89%, and in the spinous processes of the cervical and lumbar vertebrae (correlating with interspinous bursitis) of 51% of the patients with isolated PMR.[11]
A study of circadian variation in PMR found that plasma concentrations of IL-6, IL-8, TNF-α, and IL-4 peaked between 4 and 8 am in both untreated patients and controls, although levels of those cytokines were higher throughout the day in patients. The peak in cytokines matched the early-morning peak of pain and stiffness in untreated patients. In addition, melatonin levels were consistently higher in patients than in controls and varied with time, peaking around 2 am, suggesting that melatonin stimulates cytokine production, which in turn accounts at least partly for PMR symptoms.[12]
The exact cause (or causes) of PMR is unknown. The disease is more common among northern Europeans, which may indicate a genetic predisposition. Other risk factors for PMR are an age of 50 years or older and the presence of GCA. PMR has been reported as a rare complication of cancer therapy with immune checkpoint inhibitors (eg, nivolumab).[13, 14]
An autoimmune process may play a role in PMR development. PMR is associated with the HLA-DR4 haplotype. A high level of IL-6 is associated with increased disease activity.
Many investigators believe that nonerosive synovitis and tenosynovitis are responsible for many symptoms of PMR.
In the United States, the average annual incidence of polymyalgia rheumatica (PMR) is 52.5 cases per 100,000 persons aged 50 years and older. The prevalence is approximately 0.5-0.7%. In a Mayo Clinic study from 2000-2014, the overall age- and sex-adjusted annual incidence of PMR was 63.9 per 100,000 population aged ≥50 years; the incidence rate was slightly higher in those years, compared with 1970-1999.[15]
Worldwide, the frequency varies by country. In Europe, the frequency decreases from north to south, with a high incidence in Scandinavia and a low incidence in Mediterranean countries. In Italy, for example, the incidence is 12.7 cases per 100,000 persons. A United Kingdom study found an overall incidence rate of 95.9 cases per 100 000 population.[16] In a systematic review of case records from a large primary care practice in the UK, the prevalence of PMR in patients age 55 years and older ranged from 0.91% to 1.53%, depending on the criteria set used for diagnosis.[17]
Whites are affected more than other ethnic groups. PMR is only occasionally reported in blacks. PMR is twice as common in women.
The incidence increases with advanced age. PMR rarely affects persons younger than 50 years. The median age at diagnosis is 72 years.[18]
Patients with polymyalgia rheumatica (PMR) were often in good health prior to disease onset, which is abrupt in about 50% of patients. In most patients, symptoms appear first in the shoulder girdle. In the remainder, the hip or neck are involved at onset. At presentation, symptoms may be unilateral but they usually become bilateral within a few weeks.
The symptoms include pain and stiffness of the shoulder and hip girdle. The stiffness may be so severe that the patient may have a great difficulty rising from a chair, turning over in bed, or raising the arms above shoulder height. Stiffness after periods of rest (gel phenomenon) as well as morning stiffness of more than 1 hour typically occurs.
Muscle weakness is not a feature of PMR. However, this can be difficult to assess in the setting of pain, especially if symptoms are protracted and untreated, resulting in disuse atrophy.
Patients may also describe distal peripheral joint swelling or, more rarely, limb edema. Carpal tunnel syndrome can occur in some patients. Most patients report systemic features as listed below.
Several diagnostic criteria for PMR exist. One set of diagnostic criteria is as follows[19] :
In 2012, the European League Against Rheumatism and the American College of Rheumatology published new provisional classification criteria for PMR in patients aged 50 or older with bilateral shoulder aching and elevated inflammatory markers. These are not diagnostic criteria, but rather are designed for enrolling patients into clinical trials of new treatments for PMR.[20] This collaborative initiative resulted in a scoring algorithm based on the following criteria:
A score of ≥4 points has a 68% sensitivity and 78% specificity for discriminating PMR from other comparison patients. There is also an additional ultrasound criteria (1 point if positive findings), which can add up to a score of ≥5 points that is associated with a 66% sensitivity and 81% specificity for PMR.[21]
Systemic findings in more than 50% of patients are as follows:
Musculoskeletal findings are as follows[22] :
PMR is a clinical diagnosis based on the complex of the presenting symptoms and exclusion of the other potential diseases. The symptoms and signs of PMR are nonspecific, and objective findings on physical examination are often lacking.
General symptoms are as follows:
Musculoskeletal findings are as follows:
In later stages, disuse muscle atrophy with proximal muscle weakness may occur. Contractures of the shoulder capsule may lead to limitation of passive and active movements.
Joint guidelines from the European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) recommend performing the following laboratory studies in all patients with polymyalgia rheumatica (PMR), both to help to exclude mimicking conditions and to establish a baseline for monitoring therapy[23] :
Additional studies to consider are as follows[23] :
If clinically indicated, tests such as the following may be considered to exclude alternative diagnoses[23] :
The ESR is a sensitive diagnostic study for PMR, but it is not specific. The ESR is frequently greater than 40 mm/hr, but it can exceed 100 mm/hr. The ESR is mildly elevated in 7-20% of patients. Occasionally, the ESR is normal; this may occur in patients with limited disease activity.[24] In these cases, the diagnosis is based on rapid positive response to low-dose oral corticosteroids (10-15 mg/day).
The CRP level is often elevated and may parallel the ESR. Longitudinal studies suggest that CRP may be a more sensitive test than ESR for the diagnosis of PMR
The CBC reveals mild normocytic, normochromic anemia in most patients. The white blood cell count may be normal or mildly elevated. Platelet counts are often increased, reflecting systemic inflammation.
Liver function tests reveal normal transaminase enzyme levels. Alkaline phosphatase may be mildly increased in approximately one third of patients. The serum albumin level may be slightly decreased.
The creatine kinase level is normal; this finding helps differentiate PMR from polymyositis and other primary myopathic disorders.
Antinuclear antibodies, complement, rheumatoid factor, and anti-CCP levels are usually normal. The serum interleukin-6 (IL-6) level is elevated and often closely parallels the inflammatory activity of the disease; however, this test is not readily available in most laboratories. Plasma fibrinogen assays are widely available, and elevation in the plasma fibrinogen level has been recommended for the diagnosis of active PMR, with subsequent decreases used for confirmation of response to treatment.[25]
In a study of serum markers related to immune cells that may be involved in PMR and giant cell arteritis (GCA), serum B-cell activating factor (BAFF) and IL-6 were most strongly associated with disease activity in both GCA and PMR patients. Serum CCL2, CCL11, IL-10, and sIL-2R were modulated in GCA patients only, while CXCL10 was modulated in PMR patients only. The study population comprised 24 newly diagnosed, untreated GCA/PMR patients; 14 corticosteroid-treated GCA/PMR patients in remission; and 13 controls.[26]
In patients who have synovitis with effusions, synovial fluid analysis reveals signs of mild inflammation, including poor mucin clotting. Synovial fluid WBC counts range between 1,300-11,000 cells/µL (median 6,000 cells/µL), with 34% polymorphonuclear leukocytes (range 12-78%).
Radiographs reveal either normal joints or evidence of osteoarthritis. Evidence of erosive arthritis should prompt evaluation for other disorders such as rheumatoid arthritis or crystalline arthritis. Magnetic resonance imaging (MRI) is not necessary for diagnosis, but MRI of the shoulder reveals subacromial, subdeltoid bursitis and glenohumeral joint synovitis in the vast majority of patients. MRI of the hands and feet demonstrates inflammation of the tendon sheaths in many patients.
In a Japanese study, MRI of the shoulder showed a significantly thicker supraspinatus tendon and more frequent severe rotator cuff tendinopathy in patients with PMR than in patients with rheumatoid arthritis or controls. In both shoulder and hip joint MRIs, effusion around the joints was greater in PMR patients, and periarticular soft tissue edema was significantly more frequent.[27]
Ultrasonography is operator-dependent but may be useful when the diagnosis is uncertain. Bursa ultrasonography may reveal an effusion within the shoulder bursae. The ultrasonography findings and those of MRI usually correlate well.[28]
Symptomatic vasculitis in cranial and extracranial vessels is rare in PMR, but a study by Kermani et al demonstrated subclinical involvement in about one third of patients using ultrasonography and positron emission tomography (PET) scanning.[29] In a study of 18F-fluorodeoxyglucose (FDG)-PET/CT, Wakura et al reported abnormal FDG accumulation at the entheses, suggesting that enthesitis may be a feature of PMR and that its presence can help differentiate.PMR from elderly-onset rheumatoid arthritis.[30]
A joint international guideline on functional FDG-PET combined with anatomical CT angiography concluded that FDG-PET imaging has high diagnostic value for the detection of large-vessel vasculitis (ie, Takayasu arteritis and GCA) or PMR and has an important role in the diagnosis of extracranial vascular involvement in these patients, although there are no definitive consensus criteria for the presence of vascular inflammation with FDG-PET in these disorders.[31]
Temporal artery biopsy (TAB) has a very low yield in patients with isolated polymyalgia rheumatica (PMR) and is therefore usually unnecessary in patients with PMR who do not have symptoms of giant cell arteritis (GCA). TAB is not indicated in patients with mild symptoms of PMR that is of recent onset or in patients who have remained stable over a long period (1 year or longer without current or previous clinical evidence of arteritis).
Patients should be monitored for symptoms or signs of GCA after treatment initiation because low-dose corticosteroids do not prevent progression of PMR to GCA. TAB should be considered if clinical signs of vasculitis develop, if clinical response is incomplete with low doses of prednisone (≤20 mg/d), and/or if the ESR or CRP remains elevated or rises despite symptom resolution on corticosteroid therapy. Low-dose corticosteroids do not appear to affect biopsy yield.
Polymyalgia rheumatica (PMR) is a chronic, self-limited disorder. Therapy is based on empiric experiences because few randomized clinical trials are available to guide treatment decisions. The therapeutic goals are to control painful myalgia, to improve muscle stiffness, and to resolve constitutional features of the disease.
Corticosteroids (ie, prednisone) are considered the treatment of choice because they often cause complete or near-complete symptom resolution and reduction of the erythrocyte sedimentation rate (ESR) to normal. However, no definite evidence demonstrates that corticosteroids (or any other therapy) alter the natural history of PMR. Before the corticosteroid era, patients with PMR occasionally experienced spontaneous improvements, and musculoskeletal symptoms were treated with nonsteroidal anti-inflammatory drugs (NSAIDs). The low-dose corticosteroids used in PMR are almost certainly ineffective in the prevention of vasculitis progression.
Joint guidelines from the European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) conditionally recommend starting corticosteroid therapy with 12.5-25 mg/day of prednisone or the equivalent.[23] A systematic examination of the peer-reviewed literature, which included 30 studies, found that remission of PMR seemed to be achieved for most patients with a starting dose of prednisone at 15 mg/day. A slow tapering of the prednisone, less than 1 mg/month, was associated with fewer relapses. Once prednisone is tapered to 10 mg/day, a slow taper by 1 mg every 2 months until treatment discontinuation was associated with optimal control of disease activity.[32]
Nevertheless, controversy remains regarding dose and duration of treatment. Dose depends on the patient's weight and severity of symptoms. Expect prompt relief of symptoms within 24-72 hours. Corticosteroid dose should be increased if symptoms are not well controlled within 1 week, and a diagnosis of giant cell arteritis may need to be pursued, especially if prednisone 20 mg/d does not control symptoms. In contrast to other rheumatic diseases, alternate-day administration of corticosteroids in PMR has largely been unsuccessful.
Tapering should be guided by clinical response to include decreased pain and stiffness, decreased morning stiffness, and decreased shoulder pain/limitation on clinical examination.[33, 34] Normalization of inflammatory markers is also helpful but should not be used as a guideline for decreasing or stopping treatment. Patient-reported outcomes including global pain, hip pain, morning stiffness, physical function, mental function, as well as inflammatory markers have been reported as the best measures of disease activity and response to treatment.
If not contraindicated, NSAIDs may provide supplemental pain relief. However, per a study of 232 patients with PMR by Gabriel and colleagues, NSAIDs are associated with considerable drug-related morbidity and thus should be used with caution.[35]
Corticosteroid-sparing agents are sometimes considered in patients with PMR to reduce corticosteroid-related adverse effects, especially in certain patient populations such as diabetic patients or in those who develop osteonecrosis.
Methotrexate has been investigated in three randomized studies in newly diagnosed PMR. One study used methotrexate at 7.5 mg/week plus 20 mg/day of prednisone and found no benefit in outcomes after 2 years of follow-up.[36] Another study used oral and intramuscular methotrexate at a higher dose of 10 mg/week added to the prednisone regimen versus prednisone regimen alone. Overall, the patients receiving methotrexate 10 mg/week plus prednisone experienced corticosteroid-sparing effects.[37, 38] A retrospective study of methotrexate in 100 patients with relapsed or steroid-resistant PMR reported benefit with doses of up to 20 mg/day.[39]
Limited but increasing data on tocilizumab, an interleukin-6 receptor antagonist, suggest that this agent is effective, safe, and well-tolerated in patients with PMR and has a robust steroid-sparing effect.[40] Early evidence also supports tocilizumab monotherapy as first-line treatment for PMR, instead of steroids.[41]
Tumor necrosis factor alpha (TNF-α) inhibitors have also been investigated as corticosteroid-sparing agents in PMR. A randomized study with infliximab revealed no benefit.[42] The only randomized trial using azathioprine (150 mg/d) during the maintenance phase of PMR showed a high frequency of adverse drug effects and a high number of patient withdrawals from the study, although a lower cumulative dose of corticosteroid at 52 weeks. At this time, the small number of completers and the high number of giant cell arteritis patients in the study make the study results difficult to interpret.[43] EULAR/ACR guidelines strongly recommend against the use of TNF-α inhibitors in PMR.[23]
Symptomatic palliation of pain with analgesic therapy alone with close monitoring may be preferable in patients with intolerable adverse effects from corticosteroids (eg, uncontrolled diabetes mellitus, severe symptomatic osteoporosis, psychosis).
Calcium and vitamin D supplementation should be initiated in all patients with PMR who are starting corticosteroid therapy.
Generally, activity restriction is unnecessary. Physical therapy is recommended for those with difficulty achieving good mobility despite adequate medical therapy.
Diagnosis and treatment involve the primary care physician and rheumatologist. In coordination with the primary care physician, the rheumatologist plays an important role in the diagnosis, treatment, and follow-up care. Ophthalmologists, pathologists, and surgeons may be consulted on an as-needed basis should concern arise about the development of giant cell arteritis.
Involvement of primary care providers is imperative, to assist with the management of comorbidities such as prophylaxis for cardiovascular disease, glucocorticoid-induced hyperglycemia, and osteoporosis. Appropriate immunizations should be administered, ideally before corticosteroid therapy is initiated.
PMR is typically treated in an outpatient setting. Determination of prognosis and of duration of treatment remain empiric and patients often need careful supervision.
Patients receiving steroids should have monthly follow-up, with regular monitoring of inflammatory markers. An isolated increase of ESR without symptoms during the course of treatment is not a valid reason to increase corticosteroid dose; however, a temporary delay in dosage reduction may be necessary. After steroid tapering, follow-up can be performed quarterly.
The risk of vertebral fractures is five times greater in women with PMR. A baseline bone mineral density study (eg, dual-energy x-ray absorptiometry [DEXA] scan) is recommended at the onset of treatment. As most patients require corticosteroids for at least 1-2 years, bisphosphonate therapy is recommended to prevent corticosteroid-induced osteoporosis. Interestingly, in an Italian, study, PMR patients who were treated with bisphosphonates were more likely to be able to discontine corticosteroids.[44]
Because relapses are more likely to occur during the initial 18 months of therapy and within 1 year of corticosteroid withdrawal, with a frequency of approximately 50%, all patients should be monitored for symptom of recurrence throughout corticosteroid tapering and for 12 months after cessation of therapy.
Relapses usually occur when the dose of prednisone is less than 5.0-7.5 mg/day or after therapy has been discontinued. If symptoms recur, the corticosteroid dose should be increased to the dose that previously controlled the symptoms. Recurrences of the disease more than 1 year after discontinuation of corticosteroid therapy has been reported.[45]
Approximately 50-75% of patients can discontinue corticosteroid therapy after 2 years of treatment. However, some patients may require low doses of corticosteroids for several years. In a population-based study from the United Kingdom, the median duration of continuous corticosteroid treatment was 15.8 months, but around 25% of patients received more than 4 years in total of corticosteroid therapy.[16]
Patients with PMR should be monitored regularly and carefully for symptoms and signs suggestive of GCA development.
The goals of therapy in polymyalgia rheumatica (PMR) are to control painful myalgia, to improve muscle stiffness, and to resolve constitutional features of the disease. Oral corticosteroids are the first line of treatment. Nonsteroidal anti-inflammatory drugs (NSAIDs) may be helpful as adjuncts to corticosteroids during tapering, or alone in mild cases; however, because they are associated with increased drug-related morbidity, they should be used with caution, especially in elderly patients. Steroid-sparing agents may be beneficial.
The interleukin-6 receptor antagonist tocilizumab is approved for use in giant cell arteritis and has demonstrated benefit for PMR in several case series and retrospective studies.[46] However, controlled trials are needed to fully establish the efficacy of tocilizumab in PMR, and it has not yet been approved for this indication by the US Food and Drug Administration.
Clinical Context: Prednisone has the capacity to dramatically reduce inflammatory manifestations. Polymyalgia rheumatica is rapidly responsive to low doses of prednisone. However, patients may require treatment for several months to several years.
Clinical Context: Corticosteroids act as potent inhibitors of inflammation. They may cause profound and varied metabolic effects, particularly in relation to salt, water, and glucose tolerance, in addition to their modification of the immune response of the body. Alternative corticosteroids may be used in equivalent dosage.
These agents cause profound and varied metabolic effects. Their exact mechanism of action in PMR is not well known, but their efficacy may stem from their general anti-inflammatory and immunomodulatory effects. In addition, corticosteroids down-regulate cytokine production.
Clinical Context: Ibuprofen is the drug of choice for patients with mild to moderate pain. It inhibits inflammatory reactions and pain by decreasing prostaglandin synthesis.
Clinical Context: Naproxen is indicated for relief of mild to moderate pain. It inhibits inflammatory reactions and pain by decreasing activity of cyclooxygenase, which results in a decrease of prostaglandin synthesis.
Clinical Context: Ketoprofen is used for relief of mild to moderate pain and inflammation. Small dosages are indicated initially in small patients, elderly patients, and patients with renal or liver disease. Doses higher than 75 mg do not increase the therapeutic effects. Administer high doses with caution, and closely observe the patient’s response.
Clinical Context: Indomethacin is used for relief of mild to moderate pain; it inhibits inflammatory reactions and pain by decreasing the activity of COX, which results in a decrease of prostaglandin synthesis.
Clinical Context: Diclofenac inhibits prostaglandin synthesis by decreasing COX activity, which, in turn, decreases formation of prostaglandin precursors.
These agents can be administered to some patients with mild symptoms; however, patients require corticosteroids for total control of symptoms. NSAIDs may be helpful in later stages of corticosteroid dosage tapering, with close monitoring for drug-related morbidity. NSAIDs generally have no effect on the ESR.
Clinical Context: Antineoplastic agent that is immunosuppressive at lower doses. Antirheumatic effects may take several weeks to become apparent. Unknown mechanism of action in treatment of inflammatory disorders; may affect immune function. Ameliorates symptoms of inflammation (eg, pain, swelling, stiffness).
Joint guidelines from the European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) recommend regular monitoring that includes clinical assessment and laboratory studies, on the following schedule[23] :
PMR usually has a limited course of several months to 5 years. Untreated patients often feel unwell and have an impaired quality of life, but generally, PMR is not associated with serious complications. Patients treated with corticosteroids are at risk for long-term complications of corticosteroid therapy.
Relapses are common and may occur in up to 25% of all treated patients. Arteritic relapse in a patient who presented exclusively with PMR is unusual.
Every patient should be considered at risk for giant cell arteritis (GCA).
Several cases of systemic amyloidosis–associated PMR have been reported. Rare cases of bilateral ocular inflammation (episcleritis, scleritis, or anterior uveitis) developing during steroid tapering have been reported.[47]
PMR is usually self-limited. With prompt diagnosis and adequate therapy, the condition has an excellent prognosis
Inform the patient about the potential benefits and risks of corticosteroids treatment and encourage the patient to participate in choosing the treatment plan.
Emphasize the importance of healthy dietary habits and ensure adequate calcium and vitamin D supplementation.
Emphasize compliance with long-term treatment plans and follow-up care in order to prevent relapses, flares, and subsequent morbidity secondary to corticosteroid therapy.
Advise patients to immediately seek medical care if symptoms recur.