INTRODUCTION — Polymyalgia rheumatica (PMR) is an inflammatory rheumatic condition characterized clinically by aching and morning stiffness in the shoulders, hip girdle, and neck [1]. It can be associated with giant cell (temporal) arteritis (GCA), and the two disorders may represent different manifestations of a shared disease process.
PMR is two to three times more common than GCA. PMR occurs in about 50 percent of patients with GCA, while approximately 15 to 30 percent of patients with PMR eventually develop GCA [1,2]. Some patients have manifestations of both disorders occurring at different times.
The treatment of PMR will be reviewed here. Additional issues in PMR as well as the clinical manifestations, diagnosis, and treatment of GCA are discussed separately. (See "Clinical manifestations and diagnosis of polymyalgia rheumatica" and "Clinical manifestations of giant cell (temporal) arteritis" and "Diagnosis of giant cell (temporal) arteritis" and "Treatment of giant cell (temporal) arteritis".)
OVERALL APPROACH — PMR is characterized by a prompt response to glucocorticoids, usually in low to moderate doses [1-3]. The initial dose of prednisone needed to alleviate musculoskeletal symptoms in PMR is lower than that required to control the vascular inflammation associated with GCA. (See "Treatment of giant cell (temporal) arteritis".)
General guidelines — We recommend treatment with glucocorticoids as initial therapy in patients diagnosed with PMR. The primary goal of therapy is the alleviation of symptoms. Therapy has not been shown to clearly improve prognosis or prevent progression to GCA. (See "Clinical manifestations and diagnosis of polymyalgia rheumatica".)
The beneficial effect of glucocorticoids in PMR has been established by clinical experience and observational studies comparing glucocorticoid doses and preparations [1,2,4,5]. We are not aware of any controlled trials comparing prednisone or prednisolone with placebo or other single agents that are not glucocorticoids. However, patients with PMR typically present with symptoms lasting months, and the vast majority of such patients experience a dramatic improvement in symptoms within days of starting glucocorticoids. There is a significant amount of variability among these studies, including the randomized trials. A 2009 systematic review of PMR therapy described a high level of heterogeneity in diagnostic criteria, treatment protocols, and outcome measures among the 30 reports meeting inclusion criteria for the review (13 randomized trials and 17 observational studies) [5].
Despite decades of use of glucocorticoids in patients with PMR and a general consensus around the approach to therapy, there is no universally accepted approach to initial glucocorticoid dosing, maintenance therapy, or dose reduction. Our approach is generally consistent with other published treatment guidelines, such as those of the British Society for Rheumatology [6]. Medications other than glucocorticoids have not been proven effective in PMR or GCA; benefit has not been consistently shown in studies using methotrexate or tumor necrosis factor inhibitors. (See 'Methotrexate' below and 'TNF inhibitors' below.)
The initial goal of therapy is to rapidly achieve symptomatic control using a relatively low dose of glucocorticoids. After a period of quiescence, we begin to slowly taper the glucocorticoid dose. Most patients require treatment for one to two years, and relapses are common with tapering. A higher ESR, larger initial doses of prednisone, and rapid tapering may be associated with earlier relapse [7].
Baseline laboratory testing — In addition to testing done at time of diagnosis, including a CBC, ESR, CRP and other tests performed based upon the presentation and differential diagnosis, a blood glucose, urinalysis, blood urea nitrogen, creatinine, aminotransferases, alkaline phosphatase, and calcium should be obtained before initiating therapy.
Appropriate measures should be taken to prevent glucocorticoid-induced osteoporosis, which may include a baseline bone density measurement. Treatment should not be delayed while obtaining this study. (See "Prevention and treatment of glucocorticoid-induced osteoporosis" and 'Side effects of glucocorticoids' below.)
INITIAL GLUCOCORTICOID THERAPY — The optimal glucocorticoid dosing regimen is not well defined.
In most patients, we suggest a starting prednisone dose of 15 mg/day (or equivalent) given orally as a single daily dose. Depending upon the patient's weight, the severity of symptoms, and comorbidities such as diabetes mellitus, severe hypertension, or heart failure, which may be adversely affected by glucocorticoids, the starting prednisone dose may range between 10 and 20 mg/day. As an example, in smaller patients with mild symptoms and brittle diabetes, a starting dose of 10 mg/day may be tried. In an otherwise healthy patient, with severe manifestations of PMR including constitutional symptoms, an initial dose of 20 mg/day may be appropriate.
Patients usually respond quickly and often note improvement after the first dose. Symptoms are generally substantially better (50 to 70 percent reduction in pain and stiffness) within three days in patients with PMR started on prednisone.
If symptoms are not well controlled within one week of initiating therapy, the prednisone dose should be increased by 5 mg, to a dose of 20 mg/day. Until substantial improvement is achieved, the dose may be increased by 5 mg/day increments each week up to a maximum of 30 mg/day. A few patients with PMR alone may require an increase to this level [1].
Occasional patients find that a single daily dose of prednisone does not provide relief from evening or night-time pain or stiffness. Use of a divided (twice daily) dose is effective in some patients, while others require an increase in total daily dose.
This approach is consistent with the findings of a 2009 systematic review that analyzed outcomes in association with varying initial prednisone doses [5]. In the studies reviewed, starting prednisone doses greater than 10 mg/day were associated with fewer relapses and shorter duration of therapy than lower doses; starting doses of 15 mg/day or less were associated with lower cumulative doses of prednisone than higher starting doses; and starting doses greater than 15 mg/day were associated with more glucocorticoid-related adverse effects.
Alternative initial therapies — No medication has been proven equivalent to glucocorticoids in the initial management of PMR or GCA, although several have been studied as glucocorticoid-sparing agents. We do not use these medications as substitutes for initial glucocorticoid therapy in patients with PMR. (See 'Glucocorticoid-sparing therapies' below.)
RESISTANT DISEASE — Practically all patients with PMR alone will respond to 15 to 20 mg/day of prednisone. Underlying GCA or another illness must be considered in patients who require higher doses of prednisone [8]. (See "Clinical manifestations and diagnosis of polymyalgia rheumatica", section on 'Differential diagnosis'.)
MAINTENANCE PHASE AND DOSE REDUCTION — We suggest that the glucocorticoid dose that suppresses symptoms be maintained for two to four weeks after the aching and stiffness have resolved.
There is no consensus concerning the optimal tapering regimen. We generally aim to reduce the dose in small decrements every two to four weeks as tolerated to find the minimum amount that is needed to maintain suppression of symptoms. In patients receiving over 15 mg/day, we suggest lowering the dose in 5 mg/day decrements every two to four weeks. In patients on 15 mg/day, we suggest lowering the dose in 2.5 mg decrements every two to four weeks. Once the dose has fallen to 10 mg/day, we suggest reducing the glucocorticoid dose no faster than 1 mg per month [1,3]. With this approach, patients without a relapse in symptoms will have been treated for approximately one year.
MONITORING RESPONSE TO THERAPY — We closely monitor the clinical response to glucocorticoid therapy. This principally involves screening for the presence and/or recurrence of symptoms of PMR or GCA. Patients should be encouraged to promptly contact their clinician if such symptoms occur.
Elevations in ESR and serum CRP generally revert to normal with low doses of prednisone in association with symptomatic improvement [9,10]. However, continued and/or recurrent high levels may suggest alternative or additional diagnoses, such as malignancy or GCA.
Initial measurement of the ESR and CRP should be performed in all patients. Subsequently, we generally obtain a complete blood count (CBC), ESR, and CRP after two months of treatment, then continue to measure the CBC, ESR and CRP approximately every three months during glucocorticoid therapy.
Results of ESR and CRP testing should be interpreted in the context of the patient's typical values prior to the onset of illness, when available, and the patient's age. Normal values for the ESR increase with age, and values for the CRP vary from laboratory to laboratory [11,12]. Additionally, factors other than inflammation may result in (usually modest) increases above normal in both the ESR and, to a lesser degree, the CRP. (See "Acute phase reactants".)
Interleukin (IL)-6 has also been shown to correlate well with disease activity [10]. Although IL-6 measurement may be useful in assessment and prognosis, assays are not available in most laboratories and a role for IL-6 measurement in routine clinical care has not been established.
RELAPSE
Symptomatic patients — Relapse is defined by an increase in aching and morning stiffness due to the disease. This may occur in as many as 25 to 50 percent of patients [7,13]. Recrudescence of symptoms can occur during the maintenance phase or during dose reduction, particularly when the glucocorticoid dose is tapered too rapidly. It may also occur in patients who are no longer taking glucocorticoids.
The optimal approach to the management of clinical relapse is not clear. Management varies based upon whether relapse occurs in those no longer taking glucocorticoids and/or undergoing a glucocorticoid taper, along with other factors.
Our approach to the treatment of patients with recurrent symptoms is as follows:
It is important to distinguish recurrent symptoms due to PMR from the onset of diffuse aching that may occasionally be seen as a result of the prednisone taper alone. With rest, the morning stiffness and aching from PMR-related inflammation typically worsen, while symptoms of glucocorticoid-withdrawal usually improve. (See "Glucocorticoid withdrawal", section on 'Recommended tapering regimen'.)
Abnormal testing without symptoms — Elevations in the ESR and/or CRP may occur in the completely asymptomatic patient. Our approach to the management of such patients is as follows:
Alternative approaches — Although not consistent with our approach, some experts suggest that first and second relapses may also be treated by a single intramuscular injection of methylprednisolone (40 to 120 mg) [2].
SIDE EFFECTS OF GLUCOCORTICOIDS — Treatment of patients with PMR with glucocorticoids with or without NSAIDs is often associated with side effects. Prevention of and monitoring for such side effects, including osteoporosis, glucose intolerance, and hypertension, is an important component of a comprehensive management program for PMR.
(See "Major side effects of systemic glucocorticoids".)
Because most patients with PMR require glucocorticoid treatment for more than several months, we recommend that standard guidelines for prevention of glucocorticoid-induced osteoporosis be followed, including an assessment of bone mineral density at or near the time prednisone is begun, treatment with calcium and vitamin D supplementation, and when indicated by treatment guidelines, prophylactic use of bisphosphonates. These issues are discussed in detail elsewhere. (See "Prevention and treatment of glucocorticoid-induced osteoporosis".)
In addition to the complete blood count and acute phase reactants used for monitoring of disease activity, blood glucose, and when indicated, hemoglobin A1c levels, should be obtained to monitor for the development of diabetes mellitus. Patients on glucocorticoids should also have their blood pressure monitored regularly, especially early in therapy when doses are higher and in patients with known hypertension. (See "Screening for diabetes mellitus".)
GLUCOCORTICOID-SPARING THERAPIES — No medication has been consistently proven effective as a glucocorticoid-sparing agent when administered in combination with glucocorticoids. Inconsistent results have been reported in clinical studies of methotrexate and tumor necrosis factor inhibitors.
In the infrequent case where they may be tried in combination with glucocorticoids, such as a patient with diabetes who is difficult to manage due to glucocorticoid therapy, patients should be advised of the limitations of available data regarding the efficacy of potentially glucocorticoid-sparing therapies. If such an agent were required, our preference would be to use methotrexate.
Methotrexate — Among patients with PMR who are at increased risk for glucocorticoid-induced side effects, the addition of methotrexate may be glucocorticoid-sparing, as suggested by some, but not all, studies [14-16]. However, methotrexate is seldom indicated in patients with PMR without GCA, since adverse effects are relatively mild with the low doses of prednisone required to adequately control the disease.
TNF inhibitors — Limited data with TNF inhibitors show mixed results, but it is not clear whether these differences are due to the agent employed or to study design.
Infliximab — The use of prednisone plus infliximab (3 mg/kg) at weeks 0, 2, 6, 14, and 22 was not more effective than prednisone plus placebo in a randomized, double-blind trial of 51 patients with newly diagnosed PMR [17]. The proportion of patients in the two groups who were relapse-free or recurrence-free was similar at 22 weeks (55 versus 54 percent with placebo) and at 52 weeks (30 versus 37 percent).
Etanercept — Limited studies have suggested that etanercept may have some benefit in patients with PMR, but evidence that it can substitute for glucocorticoids is lacking, particularly for those with early disease. Only modest, but statistically significant, benefit of etanercept (24 percent reduction in PMR activity score at day 14 of treatment) was shown in a trial that randomly assigned 20 glucocorticoid-naïve patients to initial treatment with either etanercept (25 mg administered subcutaneously twice weekly) or placebo [18]. The PMR activity score did not change significantly in patients who received placebo injections.
Use of etanercept for PMR was also the subject of a small uncontrolled and unblinded pilot study of six patients with PMR who could not reduce their prednisone dose below 7.5 to 10 mg per day or who had developed glucocorticoid side effects [19]. All patients responded to etanercept (25 mg twice weekly for 24 weeks), and all maintained the response for three months after etanercept was stopped.
Given the limited size and quality of these studies, further trials of etanercept will be needed to determine the magnitude of benefit compared with glucocorticoids and whether responses may differ based upon the duration of disease. We limit the use of etanercept to those patients in whom methotrexate has been ineffective and who are either unable to tolerate the glucocorticoid dose required to control disease or who cannot be tapered below 7.5 mg of prednisone daily.
OTHER THERAPIES
Physical therapy — Physical therapy is suggested for patients who have difficulty regaining good mobility, including full range of motion in affected joints, despite control of the inflammatory process [2].
Nonsteroidal antiinflammatory drugs — Nonsteroidal antiinflammatory drugs (NSAIDs) have been tried as glucocorticoid-sparing agents, but can be associated with drug-related morbidity [2,20]. We occasionally administer NSAIDs in patients on very low doses of glucocorticoids who also require analgesic or antiinflammatory therapy for other disorders, such as osteoarthritis or rotator cuff tendonitis. If NSAIDs are used concurrently with glucocorticoids, appropriate gastroprotective measures should be employed. (See "NSAIDs (including aspirin): Primary prevention of gastroduodenal toxicity".)
PROGNOSIS — In most patients, PMR runs a self-limited course over months to years and glucocorticoid therapy can eventually be discontinued [21,22]. Many patients, however, require treatment for up to two to three years. Some patients require long-term therapy with stable doses less than 5 mg/day. About 10 percent of patients will relapse within 10 years following completion of their initial treatment course [2,7].
There is no evidence of increased mortality associated with PMR itself [23,24]. As a result, every effort must be made to control symptoms with a minimum of glucocorticoid side effects.
Ongoing monitoring should be performed for signs suggestive of GCA. In a large series described above, 4.4 percent of patients with pure PMR had a positive temporal artery biopsy [25]. These patients are usually not detected unless they develop manifestations of vasculitis, which presumably can also occur over time in some patients with initially normal temporal arteries.
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)
SUMMARY AND RECOMMENDATIONS
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.