A Primary Care Physician's Guide to Polymyalgia Rheumatica

    Authors: Daryl Miller, MD, Fellow, Division of Immunology & Rheumatology Department of Internal Medicine, The University of Missouri-Columbia, Columbia, MO; Susan H. Allen, MD, PhD, Medical Director, Kettering Osteoporosis Center, Kettering Medical Center, Kettering, OH; Sara E. Walker, MD, Professor, Division of Immunology & Rheumatology, Department of Internal Medicine, The University of Missouri-Columbia, Columbia, MO.

    Peer Reviewer: Gene G. Hunder, MD, Professor of Medicine, Mayo Medical School, Consultant in Internal Medicine and Rheumatology, Mayo Clinic, Rochester, MN.

Editor's Note-Polymyalgia rheumatica is associated with severe pain and stiffness that involves the proximal muscles of the shoulders and hips. The disorder primarily affects elderly Caucasians of northern European ancestry. Here, the authors provide a primary care physician's guide to polymyalgia rheumatica, including clinical features, laboratory tests, and diagnostic criteria. They also point out common problems associated with this disorder, such as giant cell arteritis and osteoporosis.


Polymyalgia rheumatica is a clinical syndrome that is characterized by muscle aching and stiffness, without weakness, that involves the neck and shoulder girdle and pelvic girdle. The disorder is largely limited to patients over the age of 50 years. Polymyalgia rheumatica is of great interest to the primary care practitioner. The diagnosis is based upon careful clinical assessment, and the symptoms are controlled rapidly and dramatically with low-dose corticosteroid treatment. Polymyalgia rheumatica can be an important clue to the existence of serious underlying illness. A classic association is giant cell arteritis.1-3 There is increasing recognition that a polymyalgia rheumatica-like syndrome can be a marker for myelodysplastic disorders4 and Lyme disease.5


The term polymyalgia rheumatica was adopted in 1959 by Barber to name a set of troublesome symptoms associated with elevation of the erythrocyte sedimentation rate (ESR).6 This syndrome had been described earlier under a number of synonyms, including "senile rheumatic gout," "periarthrosis humeroscapularis," "myalgic syndrome of the elderly with systemic reactions," and "pseudo-polyarthrite rhizomelique." Barber differentiated polymyalgia rheumatica from polymyositis and other inflammatory rheumatic diseases.6 He also noted that patients had immediate response to treatment with corticosteroids, with a good prognosis in uncomplicated disease.


There are many reports of polymyalgia rheumatica in northern Europe7,8 and the northern United States.9 In Goteborg, Sweden, 220 patients were identified in 1985-1987 with well-defined polymyalgia rheumatica and negative biopsies of the temporal artery. The average annual incidence for the entire population was 17 cases/100,000. The predilection for older individuals was illustrated by the finding that the average annual incidence for persons age 50 years and older was 50 cases/100,000.7 In the northern United States, the incidence of polymyalgia rheumatica resembled the Swedish experience. A study in Olmsted County, Minnesota (1970-1991) found that the annual age- and sex-adjusted incidence in individuals aged 50 and older was 52/100,000.9 A study conducted more recently (1987-1994) in Tromso, Norway, used negative temporal artery biopsies to exclude giant cell arteritis in 256 cases. The annual incidence in this population was 113/100,000 residents.8

Polymyalgia rheumatica clearly increases as the population ages. After the age of 80 years, the annual incidence was 78/100,000 in Goteborg7 and 97/100,000 in Olmsted County.9 Occurrence in women is about twice that in men and may be as high as 79% in women.7 The syndrome involves white individuals almost exclusively. Persons of northern European ancestry are affected more commonly than southern Europeans, and the syndrome is rare among blacks and Asians (reviewed in reference 10).


Genetic predisposition, infection, and immunological abnormalities appear to contribute to the etiology of polymyalgia rheumatica.11 The increased occurrence of polymyalgia rheumatica in people of northern European heritage, as well as aggregation in some families,12 point to hereditary factors. Polymyalgia rheumatica, giant cell arteritis, and rheumatoid arthritis are all associated with HLA DR4. Patients with polymyalgia rheumatica and giant cell arteritis share the associated sequence polymorphism encoded by the second hypervariable region of the HLA DRB1 gene. The HLA DRB1*04 allele was found in 76% of patients with polymyalgia rheumatica, and all HLA DRB1*04 alleles were represented in this group.

In contrast, rheumatoid arthritis is linked to a sequence motif in the third hypervariable region of DRB1 alleles. Expression of HLA DR4 in rheumatoid arthritis was restricted to the HLA DRB1*0401 and HLA DRB1*0404/8 alleles.13 A more recent report from the United Kingdom has, however, emphasized genetic similarity between polymyalgia rheumatica and rheumatoid arthritis. The rheumatoid arthritis shared epitope, QKRAA/QRRAA, was present in 75% of polymyalgia rheumatica patients.14

Finding antibodies directed against self antigens in patients with polymyalgia rheumatica suggests that autoimmune responses have a role in the etiology. Anticardiolipin antibodies were present in 17% of patients who presented with the uncomplicated disease.15 Another group of patients had antibodies to nuclear lamin B2 protein that reacted with the lamin-specific C terminus.16 IgG anti-endothelial cell antibodies occurred in 53% of polymyalgia rheumatica patients.17 The means whereby these antibodies might contribute to an inflammatory illness, however, have not been defined.

The theory that polymyalgia rheumatica is mediated by a systemic immune response was supported initially by reports of circulating immune complexes and depleted CD8+ (cytotoxic suppressor) T-cells.18 T-cell depletion is not found consistently, however, in polymyalgia rheumatica. Recently, numbers of these cells were reported as normal,19 and some cases had clonal expansion of CD8+ cells with a distinct J beta 2.7 gene segment usage.20

The association of fever, leukocytosis, and acute phase reactions with polymyalgia rheumatica suggest that the syndrome may be caused by infection. Enteroviral infection21 and Lyme disease5 in the elderly may present with signs and symptoms that mimic polymyalgia rheumatica. In Denmark, two epidemics of Mycoplasma pneumoniae infection were associated with peak incidence of polymyalgia rheumatica and the associated disease, giant cell arteritis. Other peaks were related to epidemics of Chlamydia pneumoniae and parvovirus B19.22 Polymyalgia rheumatica has also occurred after treatment with intravesical Bacillus Calmette-Guerin.23 The assumption that polymyalgia rheumatica is a result of infection was weakened by the report of a survey of 48 newly diagnosed patients. The prevalence of antibodies to Chlamydia pneumoniae, cytomegalovirus, enteroviruses, and respiratory syncytial virus was similar in the patients and in 22 age-matched controls. Two patients did have IgM antibodies to enteroviruses.24


Epidemiologic data suggest that an appropriate genetic background contributes to development of polymyalgia rheumatica, and an infectious agent may trigger aberrant immune responses in a susceptible individual. The disease is characterized by striking muscle stiffness and pain, and muscles may display electromyographic abnormalities (fibrillation potentials, complex repetitive discharges) that improve after treatment with corticosteroids.25 The histological appearance of skeletal muscle is relatively normal. Inflammation is not present, but Type II fiber atrophy has been reported,25 and immunoglobulin deposits were found in the perifascicular area of the perimysium.26 It is generally accepted that these changes are of little consequence, and diseased skeletal muscle is almost certainly not the cause of the painful symptoms of polymyalgia rheumatica. To underscore this fact, the muscle tissue in polymyalgia rheumatica has normal metabolic function.27

Recently, it has been appreciated that inflamed bursae and joints contribute to the striking proximal pain and stiffness that characterize this disease. A patient presenting with polymyalgia rheumatica and bilateral diffuse swelling of the hands had magnetic resonance imaging (MRI) and was shown to have synovitis of the glenohumeral joint, tenosynovitis of the extensor tendons of the hand, and marked inflammation of bursae about the shoulder. Subsequent MRI studies of 13 cases revealed that all had evidence of subacromial and subdeltoid bursitis. Three patients were studied after two months of corticosteroid treatment, and two showed resolution of periarticular disease.28 Arthroscopic biopsies of the shoulder joint have produced abnormal synovium with vascular proliferation; the tissue was infiltrated by macrophages and T-cells. HLA class II antigens and vascular cell adhesion molecule-1 (VCAM-1) were expressed strongly in cells of the synovial lining layers. Synovial vascularity and expression of HLA class II and VCAM-1 antigens were decreased following corticosteroid therapy.29,30

It has been proposed that polymyalgia rheumatica is triggered by an aberrant immune process that originates in arteries. This theory is based partially upon the observation that 10-15% of persons with polymyalgia rheumatica have giant cell arteritis proven by biopsy of the temporal artery.2 The means whereby vascular lesions could produce the symptoms of polymyalgia rheumatica are not clear. In "pure" polymyalgia rheumatica, the vessels that supply skeletal muscle and synovium do not usually have visible changes of arteritis. It is possible that giant cell arteritis affects these vessels on a subclinical level. This supposition was supported by the finding that temporal artery biopsies in polymyalgia rheumatica, which did not have visible inflammatory changes, were nevertheless abnormal. These segments of artery contained messenger RNA (mRNA) for a number of cytokines, including interleukin (IL)-2 in high levels, IL-1 beta, IL-6, and tumor necrosis factor (TNF)-alpha. Temporal artery biopsies from patients with histological evidence of giant cell arteritis also contained transcripts of these cytokines. In addition, the vessels involved in giant cell arteritis contained interferon (IFN)-gamma mRNA.31 The presence of this cytokine may be a key factor in the progression to overt arterial inflammation.2

Clinical Features

The classic symptoms of polymyalgia rheumatica are muscle pain and stiffness, and these symptoms often form the basis of the chief complaint. The shoulder girdle is commonly the first area to become symptomatic, and pain is also experienced in the neck, upper back, upper arms, lower back, and hip girdle. The pain is typically severe, aching in nature, and bilateral. Pain is accentuated by movement of the joints, but the pain is not localized to the joints. The symptoms appear to arise in muscles, tendinous attachments, and bursae.

Patients have prominent stiffness, resembling that found in rheumatoid arthritis. The stiffness is present in the morning, tends to wear off with movement, and recurs later in the day if the patient is inactive. The pain and stiffness may occur so abruptly that the patient remembers the date and hour of onset, or the onset may be gradual.

Constitutional symptoms that may be associated with polymyalgia rheumatica include malaise, fatigue, weight loss, and depression. Low-grade fever has been reported in polymyalgia rheumatica,9 but high spiking fever suggests underlying infection or giant cell arteritis.

Patients are limited functionally due to the pain and stiffness and may describe difficulties in acts that require use of the proximal musculature (brushing the hair or teeth, rising from the supine or sitting position). It should be emphasized, however, that the muscles are not truly weak.2

On physical examination, the typical patient is an elderly Caucasian with a sad and distressed facial expression. Because giant cell arteritis can occur in patients with polymyalgia rheumatica, it is a good idea to carefully palpate both temporal arteries and the branches of these vessels in the scalp. Finding normal vessels does not eliminate the possibility of arteritis. It is well recognized that involved arteries can pulsate and have no tender areas. Just the same, finding a tender segment of temporal artery can be an important clue to the presence of giant cell arteritis. A tender area is the preferred location for an arterial biopsy.

The musculoskeletal examination usually reveals normal muscle bulk, tone, and strength. Tenderness may be elicited over the paracervical muscles and the glenohumeral joints. Muscle testing is difficult because there is considerable pain with movement. If the patient has not moved the painful shoulders for a period of months, the result will likely be restriction of active and passive abduction, elevation, internal rotation, and external rotation.

Synovitis should be sought on the physical examination, because about one-fourth of patients have evidence of peripheral arthritis at the onset of disease.32 Careful palpation of the sternoclavicular joints, wrists, metacarpophalangeal joints, and knees may reveal swelling and tenderness.2

A newly recognized finding is diffuse swelling with pitting edema that can involve hands or feet.33 (See Figure 1.) The involved areas are painful, and pain is intensified when joints in the area are moved. The swollen part is occasionally warm but not erythematous. The swelling often extends beyond joint margins and occasionally follows tendons.33 Distal extremity swelling with pitting edema reflects synovial inflammation in joints and tenosynovial membranes.27

Carpal tunnel syndrome can be a "hidden" manifestation of polymyalgia rheumatica and has been reported as the presenting manifestation.34 Therefore, patients who are suspected of having this diagnosis should be questioned about nocturnal pain and paresthesias involving the thumb, index finger, and middle finger, and the examiner should note if there is atrophy of the thenar eminence, weakness in abduction or opposition of the thumb, or diminished sensation in the distribution of the median nerve.

Laboratory Tests

The key to diagnosis in polymyalgia rheumatica is an elevated ESR. This test is an indirect means of assessing concentrations of acute phase reactants, which enhance erythrocyte aggregation by binding to the red cell membrane and neutralizing the positive surface charge. In the Westergren ESR test, anticoagulated blood sits in a vertical tube. If acute phase proteins are elevated, the erythrocytes aggregate and fall rapidly. The test value is the number of mm that the red cell column falls in one hour. (See Table 1.35)

Most of the published diagnostic criteria for polymyalgia rheumatica require an ESR above 40 mm/hour,36-38 but values are often as high as 100 mm/hour or greater.2

In some instances, the ESR is not increased in patients who otherwise meet established criteria for polymyalgia rheumatica. Helfgott and Kieval found ESR of 30 mm/hour or less in 26 of 117 (22%) of patients.39 The proportion of men in this group was unusually high (58%), and anemia was less common compared to high-ESR polymyalgia rheumatica patients.39 Other authors reported that a sedimentation rate lower than 40 mm/hour was associated with a lower frequency of constitutional findings such as weight loss.40 Anemia was absent in the low-ESR patients.

Other laboratory findings are typical of systemic inflammation: increased concentrations of acute phase reactants (C-reactive protein, alpha2 globulins, fibrinogen), mild-to-moderate normocytic normochromic anemia, and elevation of the platelet count. Gamma globulins are increased in one-third of patients.37 Serum complement, which behaves as an acute phase reactant, is normal or increased.

Tests for both ESR and C-reactive protein are helpful in evaluating the atypical case. Only 1.4% of polymyalgia rheumatica patients present with normal ESR and normal C-reactive protein.32

Liver enzymes, especially alkaline phosphatase, are increased in approximately one-quarter of patients, and one-third are hypoalbuminemic.37 Increased prothrombin has been reported. Enzymes that reflect muscle inflammation, such as creatine kinase and aldolase, are normal. Tests for antinuclear antibodies and rheumatoid factor are negative, and there is no substantial evidence of renal disease.2

Elevated serum concentrations of the inflammatory cytokine, IL-6, occurred in 12 of 13 cases,41 and IL-6 levels were increased when the disease relapsed.42

Factor VIII/von Willebrand factor may be elevated in polymyalgia rheumatica. This finding is somewhat nonspecific because the factor is increased in a heterogeneous group of disorders that are associated with endothelial cell damage. The significance of high factor VIII levels in polymyalgia rheumatica is not clear; factor levels did not correlate with disease activity, the ESR value, or levels of acute phase reactants.43

Synovial fluid reflects mild inflammation. The leukocyte count is expected to be 1000-20,000 cells/mm3, with 40-50% neutrophils.

Diagnostic Criteria

In 1979, Bird reported a collaborative study in which 146 patients with polymyalgia rheumatica in Great Britain were compared with 253 other patients whose diseases resembled polymyalgia rheumatica.36 It was proposed that patients having three of seven criteria had polymyalgia rheumatica with 92% sensitivity and 80% specificity. (See Table 2.44) Other sets of published diagnostic criteria have emphasized the diagnostic importance of pain, recent onset of symptoms, elevated ESR, morning stiffness, older age (over 50 years), systemic symptoms such as weight loss and fever, elevation of acute phase reactants other than the ESR, and muscle tenderness.38

An important finding in polymyalgia rheumatica is that treatment with low doses of corticosteroids (10-20 mg prednisone a day) produces prompt relief of symptoms. This response is so predictable and dramatic that some authors consider it a diagnostic criterion.37,38,45,46

Differential Diagnosis

The diagnosis of polymyalgia rheumatica is based on clinical findings with supportive data from laboratory tests. It is a diagnosis of exclusion, in that the clinician must rule out a wide variety of diseases that can present with constitutional symptoms, proximal myalgias, and high ESR. Diseases that mimic polymyalgia rheumatica are listed in Table 3.

Elderly-onset rheumatoid arthritis, with onset at 60 years of age or older, can have an abrupt polymyalgia-like onset. The shoulders and hips are involved early, the metacarpophalangeal joints may be spared, and subcutaneous nodules appear to be less common than in younger rheumatoid arthritis patients. The ESR tends to be high with elderly onset. If the test for rheumatoid factor is negative, it may be impossible to differentiate elderly-onset rheumatoid arthritis from polymyalgia rheumatica until the patient has been observed for 6-12 months. Response to low-dose oral prednisone, 5-15 mg daily, is excellent. Some clinicians believe that efficacy of second-line remittive agents is reduced, and these drugs may be more toxic in the elderly rheumatoid arthritis patient.47

The elderly patient who presents with bilateral pitting edema of the hands and/or feet, joint pain, and an elevated ESR is a diagnostic puzzle. The possibilities that need to be considered are polymyalgia rheumatica with distal pitting edema and remitting seronegative symmetrical synovitis with pitting edema (RE3PE). Polymyalgia rheumatica can be associated with distal pitting edema that results from synovial inflammation in the joints and about tendons. This finding was reported in 19 (8%) of members of a population cohort of 245 patients with polymyalgia rheumatica.33 The soft tissue swelling usually developed gradually, involved both sides, and was present in either hands or feet or in all four extremities. Some cases had recurring episodes, but erosions and joint destruction did not occur. Clues to the correct diagnosis are a very high ESR and strongly positive tests for acute phase reactants, such as C-reactive protein. Tests for rheumatoid factor are negative. These patients have poor response to nonsteroidal anti-inflammatory drugs (NSAIDs) but do improve dramatically after treatment with modest doses of corticosteroids. The prognosis is good.33,48

RS3PE is a syndrome of elderly Caucasians, primarily men, who have the abrupt onset of proximal myalgias, symmetrical synovitis involving the wrists and small joints of the hands, and pitting edema of the hands, feet, and ankles. These patients have elevated ESR and are sero-negative for rheumatoid factor. Most are positive for the HLA-B27 antigen. They show prompt improvement with corticosteroids. The prognosis is very good; the episode occurs only once.49

Myopathies (polymyositis, drug-induced) may present with proximal muscle aching but are distinguished from polymyalgia rheumatica by true proximal muscle weakness and elevated creatine kinase. Malaise and muscle and joint pain are common presenting symptoms in systemic lupus erythematosus. Skin rash and renal involvement are two factors that can differentiate lupus from polymyalgia rheumatica. Serum complement levels are expected to be low in many cases of active lupus, whereas high levels are characteristic of polymyalgia rheumatica. The diagnosis of lupus is established by finding positive tests for antinuclear antibodies. Multiple myeloma and Waldenstrom's macroglobulinemia are characterized by paraproteinemia, and lymphomas are suspected if lymph nodes or spleen are enlarged. Very careful attention must be paid to the blood counts and peripheral blood smears in these patients, because myelodysplastic syndromes and leukemias can present with polymyalgia-like symptoms.4 The diagnosis of paraneoplastic syndrome associated with a solid tumor rests upon a thorough evaluation of the geriatric patient, with the realization that occult malignancies are classic causes of weight loss, anemia, and elevated ESR. Subacute bacterial endocarditis can reproduce many findings of polymyalgia rheumatica, including aching pain in the proximal muscles and weight loss. A patient with a significant heart murmur or peripheral embolization, especially if fever is present, should have blood cultures and an echocardiogram. Infection with enterovirus can produce polylmyalgia-like pain that should be transient.21 Lyme disease can present with musculoskeletal pain and fatigue and should be considered if the patient has been in an endemic area.5 The ESR is elevated in about one-half of Lyme disease patients, but it is rare to see elevations to the very high values that are characteristic of polymyalgia rheumatica.50 Amyloidosis is associated with severe muscle pain but can be differentiated from polymyalgia rheumatica by the presence of other findings such as skin lesions, cardiac involvement, and enlarged liver and spleen. The patient with hypothyroidism may not feel well and can present with proximal muscle weakness. In this individual, hoarseness, a slow heart rate, and typical changes in skin, hair, and the deep tendon reflexes are clues to the underlying disorder. Metastatic cancer and Paget's disease typically produce deep, distressing pain that is localized over the affected areas. The pain of metastases is worse at night. Symptomatic osteomalacia can be complicated by painful microfractures. A careful clinical assessment will determine that this pain is skeletal and not in tendons, bursae, muscles, or joints. Fibromyalgia is characterized by pain that is often described as "all over the body," fatigue upon waking, classic tender points, and a normal ESR.

Table 1. Abnormal Laboratory Tests in Polymyalgia Rheumatica

    Acute phase reactants ­­ ESR

    ­ C-reactive protein

    ­ Alpha 2 globulins

    ­ Fibrinogen

    Blood count Normochromic, normocytic anemia

    ­ Platelet count

    Liver enzymes ­ Alkaline phosphatase


    Other ¯ Albumin

    ­ Serum IL-6

    ­ Factor VIII/von Willebrand's factor

    Adapted from: Dwolatzky T, Sonnenblick M, Nesher G. Giant cell arteritis and polymyalgia rheumatica: Clues to early diagnosis. Geriatrics 1997;52:38-44; and from data published in references 41-43.

Temporal Artery Biopsy in Polymyalgia Rheumatica. Polymyalgia rheumatica can be a marker for giant cell arteritis (temporal arteritis), which is diagnosed in 10-15% of polymyalgia rheumatica patients.2 Because it is impractical to obtain a temporal artery biopsy in every patient with polymyalgia rheumatica, Rodriguez-Valverde and associates attempted to select patients who were most at risk for giant cell arteritis.51 Predictors that suggested the presence of giant cell arteritis were high spiking fevers, new onset of headache, jaw claudication, amaurosis, abnormal temporal arteries, and elevated liver enzymes.51 These features should guide the physician and the patient in the decision to obtain a biopsy.

The question has been raised that some patients with multiple risk factors almost certainly have arteritis, and a biopsy might not be necessary. Even though the diagnosis of giant cell arteritis seems assured on a clinical basis, it is recommended that a temporal artery be biopsied. A positive biopsy provides a clear basis for undertaking treatment that is potentially hazardous in the aged patient. Patients with giant cell arteritis require initial treatment with 60 mg of prednisone daily. This dose is necessary to control the inflammatory vasculitis and to prevent blindness. In contrast, the polymyalgia rheumatica patient who does not have giant cell arteritis will require a lower dose in the range of 5-20 mg daily. Occasionally, an arterial biopsy will disclose a disease that mimics giant cell arteritis. An example is amyloidosis with involvement of the temporal artery.52

In the Rodruguez-Valverde study, patients who were unlikely to have giant cell arteritis as a complication of polymyalgia rheumatica were 70 years of age or younger, without headache, and without cranial features of giant cell arteritis.51 Cranial features were defined as abnormal temporal arteries (decreased or absent pulses, arterial thickening, swelling, or tenderness), amaurosis (transient or permanent loss of vision), and jaw claudication. The risk of giant cell arteritis was very low, occurring in one of 59 patients without clinical evidence of cranial vessel involvement.51

It seems appropriate to avoid biopsy in polymyalgia rheumatica patients without cranial features of arteritis and to treat with low-dose corticosteroids.51 The patient and doctor should realize, however, that there is a small possibility of giant cell arteritis occurring in a low-risk patient who is receiving low-dose steroid treatment. Close follow-up and good communication is necessary so that fever, headache, or visual symptoms will be reported promptly.

Table 2. Diagnostic Criteria for Polymyalgia Rheumatica

    · Bilateral shoulder pain and stiffness

    · Onset of illness within two weeks

    · Initial ESR higher than 40 mm/hour

    · Morning stiffness exceeding one hour

    · Age older than 65 years

    · Depression and/or loss of weight

    · Bilateral upper arm tenderness

    Any three criteria discriminated between polymyalgia rheumatica and other conditions with sensitivity of 92% and specificity of 80%.36

    Adapted from Cimmino MA, Salvarani C. Polymyalgia rheumatica and giant cell arteritis. Balliere's Clin Rheumatol 1995;9:515-527 and based upon data in Bird HA, et al. An evaluation of criteria for polymyalgia rheumatica. Ann Rheum Dis 1979;38:434-439.

Therapy of Polymyalgia Rheumatica

The original series of patients reported by Barber included two patients who responded to phenylbutazone and three patients who had spontaneous remission.6 No patients received corticosteroids. Some patients have responded to NSAIDs. Chuang described 96 patients with polymyalgia rheumatica who were identified in a 10-year period in Olmsted County.37 Thirty were treated with aspirin, and nine received other NSAIDs. Duration of disease was shorter (median, 8 months), and there were fewer relapses compared to patients treated with corticosteroids. The good outcomes were thought to result from milder disease in these patients. They had higher pre-treatment hemoglobin and lower pre-treatment ESR values, compared to the patients who received corticosteroids.37

There may be a role for NSAID treatment in selected patients with mild polymyalgia rheumatica, but many clinicians believe that low-dose prednisone is the treatment of choice. NSAID treatment often fails. In about two-thirds of individuals who receive aspirin or other NSAIDs initially, it is necessary to switch to corticosteroids in order to adequately control pain and stiffness.37 Furthermore, low-dose prednisone provides immediate relief. Within 24-48 hours of starting this treatment, patients experience dramatic improvement in myalgic and constitutional symptoms.

Table 3. Differential Diagnosis: Diseases that can Mimic Polymyalgia Rheumatica

    · Rheumatoid arthritis and RS3PE syndrome

    · Myopathies (inflammatory, drug-induced)

    · Systemic lupus erythematosus

    · Hematopoietic malignancies

    Multiple myeloma, macroglobulinemia of Wandenstrom, myelodysplastic syndromes, lymphoma

    · Paraneoplastic syndromes

    Renal cell carcinoma

    Adenocarcinomas (uterus, stomach, pancreas, colon, prostate, unknown origin)

    · Infections (subacute bacterial endocarditis, viral infection, Lyme disease)

    · Amyloidosis

    · Hypothyroidism

    · Bone pain (metastatic cancer, metabolic bone disease)

    · Fibromyalgia

    Adapted from Brooks RC, McGee SR. Diagnostic dilemmas in polymyalgia rheumatica. Arch Intern Med 1997;157:162-168 and from data published in references 5, 47, 49, and 50.

A starting dose of 5-20 mg of prednisone a day is usually effective. Occasionally, a dose of 25 mg is required. Behn examined the prospective course of treated polymyalgia rheumatica.53 The corticosteroid dose was regulated with a goal of suppressing symptoms and keeping the ESR below 30 mm per hour. Starting treatment with prednisolone, 5-10 mg per day, gave satisfactory control in 94% of the patients with uncomplicated disease. Other investigators reported relapse in 65% of patients treated with the 10 mg dose. Relapses occurred in only 10% of those who received 20 mg daily.54

Patients should remain on the starting dose of prednisone for a minimum of four weeks. The rate of dose taper is guided by the clinical response. Resolution of symptoms is more important than reduction in the ESR. For those taking 10 mg or less of prednisone a day, a reduction of the daily dose in 1 mg increments at four-week intervals is acceptable. If the dose is greater than 10 mg, it can be decreased by 2.5 mg every four weeks until the patient is taking 10 mg per day. Subsequent reductions should be in 1 mg increments. The duration of treatment is generally two years, and prednisone can be tapered and discontinued in most patients.

Patients whose disease is not controlled, or who have significant corticosteroid side effects, should be considered for treatment with a steroid-sparing drug. In one series, 65% of polymyalgia rheumatica patients treated with an average daily prednisone dose of 9.5 mg had at least one adverse event. The risks of diabetes mellitus, vertebral fractures, and hip fractures were 2-5 times greater compared to controls.55 Methotrexate treatment reduced the amount of prednisone taken over a one-year period in one group of patients,56 but other investigators concluded that it had no steroid-sparing effect.57 Dapsone is another possible choice for adjunct treatment.58

Corticosteroid-Induced Osteoporosis

Glucocorticoids inhibit calcium absorption, increase urinary calcium excretion, decrease bone formation, and increase bone resorption. The greatest loss of bone occurs during the first 6-12 months of therapy and is predominantly due to increased bone resorption. Trabecular bone is affected more than cortical bone, and the trabecular volume may decrease by 20%.59

A major challenge facing clinicians prescribing long-term corticosteroid therapy is to monitor and, if possible, minimize the effects on bone density. Most health care providers recognize that postmenopausal women, especially those with preexisting osteoporosis, are at increased risk. It is important to recognize that men are equally at increased risk. Cumulative dose and duration of therapy are the most important risk factors. The incidence of fracture increases with higher cumulative doses and may be as high as 50% in individuals receiving 30 g of prednisone. The most common fracture sites are the vertebrae, ribs, and pelvis.60 In a cohort of 43 patients with temporal arteritis treated initially with high-dose prednisone, one-third experienced fractures within the first 12 months of therapy.61 Low-dose prednisone also places patients at risk. A dose of 8 mg per day is associated with an annualized bone loss of 3%.

When treatment with prednisone is started, the patient should be evaluated for modifiable risk factors and other medical illnesses that may be adversely affected by corticosteroid therapy. Laboratory evaluation should include a 25-hydroxyvitamin D level. Bone mineral density testing should be obtained as soon as possible after glucocorticoid therapy has been initiated. The Scientific Advisory Board of the National Osteoporosis Foundation recommends that bone density testing should be performed on any individual prescribed long-term steroids of at least 7.5 mg prednisone per day for longer than one month. Bone mineral density in the patient should be compared to the peak bone mass in young adults of the same gender (T score).

Table 4 outlines a rational management strategy for glucocorticoid-induced osteoporosis. The American College of Rheumatology has recommended supplementation with high doses of vitamin D (50,000 IU, 3 times per week) or calcitriol (0.5 mcg per day). The potential complication of hypercalcemia with these doses is a major concern. Others have suggested that a weekly dose of 50,000 IU of vitamin D for the first several months of prednisone therapy is an appropriate dose.62 The 25-hydroxyvitamin D level should be maintained within the upper limits of normal. The daily elemental calcium intake should be 1500 mg in divided doses. Thiazide diuretics may be used to decrease hypercalciuria.

Anti-resorptive therapy should be prescribed at the onset of corticosteroid therapy, when the prednisone dose is highest and the greatest bone loss occurs. Hormone replacement therapy should be prescribed in all postmenopausal women, regardless of the bone mineral density result, if there are no contraindications. Hypogonadal men should receive parenteral or transdermal testosterone replacement. Another anti-resorptive therapy such as calcitonin or a bisphosphonate such as etidronate or alendronate should be prescribed if hormone replacement is refused or contraindicated. Parenteral salmon calcitonin (100 IU, 3 times a week) showed no benefit compared to the control group in a two-year intervention trial in 48 patients with newly diagnosed polymyalgia rheumatica, temporal arteritis, and other vasculitides treated with corticosteroids.63 However, the nasal spray preparation of salmon calcitonin (200 IU per day) did prevent vertebral bone loss after one year in patients with polymyalgia rheumatica.64

Table 4. Management of Corticosteroid-Induced Osteoporosis


    · Eliminate adverse health habits-smoking and alcohol excess

    · Refer to physical therapist for exercise training including a weight-bearing exercise program, back extension and isometric abdominal exercises, proximal muscle strengthening exercises, and fall prevention strategies

    · Dietary intervention: Daily elemental calcium intake of 1500 mg and at least 400-800 IU of vitamin D; 2-3 g sodium restricted diet; caloric restriction; diet high in protein and potassium; American Diabetes Association dietary recommendations for patients with diabetes mellitus


    · Vitamin D, 50,000 IU a week for three months if the 25-hydroxyvitamin D level is below the upper limit of normal. Then re-evaluate with another 25-hydroxyvitamin D level.

    · Thiazide diuretic to reduce hypercalciuria

    · Hormone replacement therapy in postmenopausal women and testosterone therapy in men, if there are no contraindications

    · Anti-resorptive therapy: calcitonin, bisphosphonate

Intermittent cyclical etidronate (400 mg per day for 2 weeks followed by 11 weeks with no drug) has been shown to prevent glucocorticoid-induced bone loss in patients with polymyalgia rheumatica and giant cell arteritis.65 Alendronate (10 mg per day) shows potential usefulness in preventing and treating corticosteroid-induced osteoporosis.66

Bone mineral density testing should be repeated in six months to assess efficacy of therapy. A change in anti-resorptive treatment or additional anti-resorptive treatment should be considered when the bone loss is greater than 5%.


Polymyalgia rheumatica is associated with severe pain and stiffness that involves the proximal muscles of the shoulders and hips. Elevated sedimentation rate and elevated acute phase reactions are hallmarks of this disease. It is primarily a disorder of elderly Caucasians of northern European ancestry. The genetic background of affected individuals has similarities to that of patients with giant cell arteritis and rheumatoid arthritis. Recently, anticardiolipin antibodies, antibodies to nuclear lamin B2 protein, and anti-endothelial cell antibodies have been linked to polymyalgia rheumatica. Earlier reports of depleted CD8+ T-cells are now open to question. Skeletal muscle, the apparent site of pain in this disease, is relatively normal, but intense inflammation has been identified in bursae, joints, and tenosynovial areas. Giant cell arteritis is associated with polymyalgia rheumatica. It is noteworthy that the temporal arteries of polymyalgia rheumatica patients contain high levels of IL-2 mRNA, whereas arteries involved in giant cell arteritis express IFN-gamma mRNA. The physical examination in polymyalgia rheumatica is relatively normal, but the clinician should look carefully for evidence of temporal artery involvement, inflamed joints, carpal tunnel syndrome, and diffuse swelling with pitting edema that can involve hands or feet. The differential diagnosis includes rheumatoid arthritis, myopathies, systemic lupus erythematosus, and a number of malignancies. Infections (bacterial endocarditis, enteroviral infection, Lyme disease) and amyloidosis can present with a polymyalgia rheumatica-like picture. Ten to 15% of patients with polymyalgia rheumatica have giant cell arteritis, and temporal artery biopsies should be considered in those with high fever, jaw claudication, amaurosis, and/or abnormal temporal arteries. Most patients with uncomplicated polymyalgia rheumatica respond dramatically to low-dose prednisone. Because steroid-treated patients are at risk for osteoporosis and fractures, anti-osteoporosis measures should be instituted at the beginning of therapy.


    1. Hunder GG, Disney TF, Ward LE. Polymyalgia rheumatica. Mayo Clin Proc 1969;44:849-875.

    2. Hunder GG. Giant cell arteritis and polymyalgia rheumatica. Med Clin N Am 1997;81:195-219.

    3. Swannell AJ. Polymyalgia rheumatica and temporal arteritis: Diagnosis and management. BMJ 1997;314:1329-1332.

    4. Berthelot JM, et al. Joint manifestations in myelodysplastic syndromes. A report of three cases presenting as polymyalgia rheumatica. Rev Rhum Engl Ed 1997;64:95-100.

    5. Schwartzberg M, Weber CA, Musico J. Lyme borreliosis presenting as a polymyalgia rheumatica-like syndrome. Brit J Rheumatol 1995;34:392-398.

    6. Barber HS. Myalgic syndrome with constitutional effects. Polymyalgia rheumatica. Ann Rheum Dis 1957;16:230-237.

    7. Schaufelberger C, Bengtsson BA, Andersson R. Epidemiology and mortality in 220 patients with polymyalgia rheumatica. Brit J Rheumatol 1995;34:261-264.

    8. Tore Gran J, Myklebust G. The incidence of polymyalgia rheumatica and temporal arteritis in the county of Aust Agder, south Norway: A prospective study 1987-94. J Rheumatol 1997;24:1739-1743.

    9. Salvarani C, et al. Epidemiology of polymyalgia rheumatica in Olmsted County, Minnesota, 1970-1991. Arthritis Rheum 1995;38:369-373.

    10. Hazleman BL. Polymyalgia Rheumatica and Giant Cell Arteritis. In: Klippel JH, Dieppe PA (eds): Rheumatology. St. Louis: Mosby, 1994:6.18.1-6.18.8.

    11. Cimmino MA. Genetic and environmental factors in polymyalgia rheumatica. Ann Rheum Dis 1997;56:576-577.

    12. Zauber P, Zhang L, Berman E. Familial occurrence of temporal arteritis. J Rheumatol 1997;24:611-612 (letter).

    13. Weyand CM, et al. HLA-DRB1 alleles in polymyalgia rheumatica, giant cell arteritis, and rheumatoid arthritis. Arthritis Rheum 1994;37:514-520.

    14. Haworth S, et al. Polymyalgia rheumatica is associated with both HLA-DRB1*0401 and DRB1*0404. Brit J Rheumatol 1996;35:632-635.

    15. Chakravarty K, et al. A longitudinal study of anticardiolipin antibody in polymyalgia rheumatica and giant cell arteritis. J Rheumatol 1995;22:1694-1697.

    16. Brito J, et al. Autoantibodies to human nuclear lamin B2 protein. Epitope specificity in different autoimmune diseases. J Immunol 1994;153:2268-2277.

    17. Le Tonqueze M, et al. The relationship of anti-endothelial cell antibodies to anti-phospholipid antibodies in patients with giant cell arteritis and/or polymyalgia rheumatica. Autoimmunity 1995;20:59-66.

    18. Benlahrache C, et al. Decrease of the OKT8 positive T cell subset in polymyalgia rheumatica. Lack of correlation with disease activity. Arthritis Rheum 1983;26:1472-1480.

    19. Uddhammar A, et al. Peripheral blood lymphocyte subsets in polymyalgia rheumatica. Clin Rheumatol 1995;14:62-67.

    20. Martinez-Taboada VM, Goronzy JJ, Weyland CM. Clonally expanded CD8 T cells in patients with polymyalgia rheumatica and giant cell arteritis. Clin Immuno Immunopathol 1996;79:263-270.

    21. Stevens RJ, Hughes RA. Polymyalgic presentation of enterovirus infection: A cause of diagnostic confusion. Ann Rheum Dis 1996;55:147-148 (letter).

    22. Olling P, Olsson AT, Elling H. Synchronous variations of the incidence of temporal arteritis and polymyalgia rheumatica in different regions of Denmark; association with epidemics of mycoplasma pneumoniae infection. J Rheumatol 1996;23:112-119.

    23. Genereau T, et al. Polymyalgia rheumatica with temporal arteritis following intravesical Calmette-Guerin bacillus immunotherapy for bladder cancer. Clin Exp Rheumatol 1996;14:110 (letter).

    24. Uddhammar A, et al. Antibodies against Chlamydia pneumoniae, cytomegalovirus, enteroviruses and respiratory syncytial virus in patients with polymyalgia rheumatica. Clin Exp Rheumatol 1997;15;299-302.

    25. Bromberg MB, Donofrio PD, Segal BM. Steroid-responsive electromyographic abnormalities in polymyalgia rheumatica. Muscle Nerve 1990;13:138-141.

    26. Shintani S, et al. Immunofluorescence study of immune complexes in polymyalgia rheumatica. J Neurol Sci 1995;128:103-106.

    27. Mattei JP, et al. P-31 magnetic resonance spectroscopy demonstrates unaltered muscle energy utilization in polymyalgia rheumatica. Arthritis Rheum 1997;40:1817-1822.

    28. Salvarani C, et al. Proximal bursitis in active polymyalgia rheumatica. Ann Intern Med 1997;127:27-31.

    29. Meliconi R, et al. Leukocyte infiltraiton in synovial tissue from the shoulder of patients with polymyalgia rheumatica. Quantitative analysis and influence of corticosteroid treatment. Arthritis Rheum 1996;39:1199-1207.

    30. Meliconi R, et al. Synovial expression of cell adhesion molecules in polymyalgia rheumatica. Clin Exp Immunol 1997; 107:494-500.

    31. Weyand CM, et al. Tissue cytokine patterns in patients with polymyalgia rheumatica and giant cell arteritis. Ann Intern Med 1994;121:484-491.

    32. Myklebust G, Gran JT. A prospective study of 287 patients with polymyalgia rheumatica and temporal arteritis: Clinical and laboratory manifestations at onset of disease and at the time of diagnosis. Brit J Rheumatol 1996;35:1161-1168.

    33. Salvarani C, Gabriel S, Hunder GG. Distal extremity swelling with pitting edema in polymyalgia rheumatica. Report of nineteen cases. Arthritis Rheum 1996;39:73-80.

    34. Herrera B, et al. Carpal tunnel syndrome heralding polymyalgia rheumatica. Scand J Rheumatol 1997;26:222-224.

    35. Dwolatzky T, Sonnenblick M, Nesher G. Giant cell arteritis and polymyalgia rheumatica: Clues to early diagnosis. Geriatrics 1997;52:38-44.

    36. Bird HA, et al. An evaluation of criteria for polymyalgia rheumatica. Ann Rheum Dis 1979;38:434-439.

    37. Chuang T, et al. Polymyalgia rheumatica. A 10 year epidemiologic and clinical study. Ann Intern Med 1982;97:672-680.

    38. Brooks RC, McGee SR. Diagnostic dilemmas in polymyalgia rheumatica. Arch Intern Med 1997;157:162-168.

    39. Helfgott Sm, Kieval RI. Polymyalgia rheumatica in patients with a normal erythrocyte sedimentation rate. Arthritis Rheum 1996;39:304-307.

    40. Gonzalez-Gay MA, et al. Polymyalgia rheumatica without significantly increased erythrocyte sedimentation rate. Arch Intern Med 1997;157:317-320.

    41. Roche NE, et al. Correlation of interleukin-6 production and disease activity in polymyalgia rheumatica and giant cell arteritis. Arthritis Rheum 1993;36:1286-1294.

    42. Caplanne D, Le Parc J, Alexandre J. Interleukin-6 in clinical relapses of polymyalgia rheumatica and giant cell arteritis. Ann Rheum Dis 1996;55:403-404.

    43. Olsson A, Elling P, Elling H. Serological and immunohistochemical determination of von Willebrand factor antigen in serum and biopsy specimens from patients with arteritis temporalis and polymyalgia rheumatica. Clin Exp Rheumatol 1990;8:55-58.

    44. Cimmino MA, Salvarani C. Polymyalgia rheumatica and giant cell arteritis. Balliere's Clin Rheumatol 1995;9:515-527.

    45. Jones JG, Hazleman BL. Prognosis and management of polymyalgia rheumatica. Ann Rheum Dis 1981;40:1-5.

    46. Healey LA. Long-term follow-up of polymyalgia rheumatica: Evidence for synovitis. Semin Arthritis Rheum 1984;13:322-328.

    47. van Schaardenburg D, Breedveld FC. Elderly-onset rheumatoid arthritis. Semin Arthritis Rheum 1994;23:367-378.

    48. Chaoquat D. Peripheral joint involvement in polymyalgia rheumatica. Rev Rhum [Engl Ed] 1997;64:184-188.

    49. Russell EB, et al. Remitting, seronegative, symmetrical synovitis with pitting edema-13 additional cases. J Rheumatol 1990; 17:633-639.

    50. Paparone PW. Polymyalgia rheumatica or Lyme disease? How to avoid misdiagnosis in older patients. Postgrad Med 1995; 97:161-170.

    51. Rodriguez-Valverde V, et al. Risk factors and predictive models of giant cell arteritis in polymyalgia rheumatica. Am J Med 1997;102:331-336.

    52. Salvarani, et al. Primary systemic amyloidosis presenting as giant cell arteritis and polymyalgia rheumatica. Arthritis Rheum 1994;37:1621-1626.

    53. Behn AR, Perera T, Myles A. PMR and corticosteroids: How much for how long? Ann Rheum Dis 1983;48:658-651.

    54. Kyle V, Hazleman BL. Treatment of PMR and GCA: I. Steroid regimens in the first two months. Ann Rheum Dis 1989;48:658-661.

    55. Gabriel SE, et al. Adverse outcomes of antiinflammatory therapy among patients with polymyalgia rheumatica. Arthritis Rheum 1997;40:1873-1878.

    56. Ferraccioli G, et al. Methotrexate in polymyalgia rheumatica: Preliminary results of an open, randomized study. J Rheumatol 1996;23:624-628.

    57. Feinberg HL, et al. The use of methotrexate in polymyalgia rheumatica. J Rheumatol 1996;23:1550-1552.

    58. Doury P, et al. The use of dapsone in the treatment of giant cell arteritis and polymyalgia rheumatica. Arthritis Rheum 1983; 26,689-690 (letter).

    59. Canalis E. Mechanism of glucocorticoid action on bone: Implications of glucocorticoid-induced osteoporosis. J Clin Endocrinol Metab 1996;81:3441-3447.

    60. Wolinsky-Freidland M. Drug-induced metabolic bone disease. Endocrinol Metab Clin N Am 1995;24:395-420.

    61. Nesher G, et al. Analysis of steroid related complications and mortality in temporal arteritis: A 15-year survey of 43 patients. J Rheumatol 1994;21:1283-1286.

    62. Lukert B. Glucocorticoid-induced osteoporosis. In: Marcus R, et al (eds). Osteoporosis. San Diego: Academic Press; 1996;801-802.

    63. Healey, et al. A randomized controlled trial of salmon calcitonin to prevent bone loss in corticosteroid-treated temporal arteritis and polymyalgia rheumatica. Calcif Tissue Int 1996;58:73-80.

    64. Adachi JD, et al. Salmon calcitonin nasal spray in the prevention of corticosteroid-induced bone loss. Br J Rheumatol 1994; 33:348-350.

    65. Mulder H, Struys A. Intermittent cyclical etidronate in the prevention of corticosteroid-induced bone loss. Br J Rheumatol 1994;33:348-350.

    66. Saag K, et al. Alendronate for the management of glucocorticoid-induced ostroporosis: Results of the multicenter U.S. Study. Arthritis Rheum 1997;40:S134 (abstract).