Polymyalgia Rheumatica (PMR) is an inflammatory rheumatic disease that commonly affects individuals over 50 years of age, characterised by pain and morning stiffness of the shoulder and pelvic girdle. PMR can present as ‘isolated’ form or may be associated with giant cell arteritis. The progress of imaging techniques has helped in understanding different clinical patterns: subclinical vasculitis can occur in at least one-third of PMR patients, causing ischaemic complications. It is considered a polygenic disease and environmental factors may play a role in its pathogenesis, such as viral or bacterial triggers, both in the ‘wide’ form or assembled with adjuvants in vaccines. The response to steroid therapy is generally dramatic and side effects may occur, as well as the development of glucocorticoid resistance. The optimisation of therapy may require steroid-sparing agents as well as modified-release prednisone as ‘nighttime’ replacement therapy.
Polymyalgia rheumatica (PMR) is an inflammatory condition characterised by pain and morning stiffness of the shoulder and pelvic girdle alongside evidence of an synovitis of proximal joints and extra-articular synovial structures that commonly affects individuals over 50 years of age. It can occur in ‘isolated’ form or be related to giant cell arteritis (GCA).
The first description of PMR dates back to 1888, when Bruce defined a proximal muscular syndrome as ‘senile rheumatic gout’ in five elderly patients .
Until 1950s case reports appeared under varying names, such as ‘periarticular fibrositis’, ‘peri-extra articular rheumatism’, ‘periarthrosis humeroscapularis’, ‘myalgic syndrome of the aged’ and ‘pseudo-polyarthrite rhizomielique’.
The term PMR was coined in 1957 by the British physician Barber, who treated patients at a spa in Buxton .
The definition ‘anarthritic rheumatoid disease’ was used in 1963 by Bagratuni to describe a group of patients with articular manifestations similar to rheumatoid arthritis (RA) but without the development of articular erosions . The concept of PMR as a systemic articular disease was furthered by Healey, who described variant forms of PMR that he called ‘benign synovitis’ .
In the 1960s, reports by Paulley and Hughes and, subsequently, by Alestig and Barr proposed a relationship between PMR/GCA . In 1968, Hamrin and colleagues presented six cases of ‘polymyalgia arteritica’, in which clinical manifestations were associated to histopathologic signs of GCA in the superficial temporal arteries of three patients and, in a further three patients, in the aorta and its branch .
Hamrin also sustained that PMR should be considered as ‘a common disease caused by an arteritic process’ occurring in the aorta and large arteries, and GCA as ‘an episodic and unusual phenomenon in the course of PMR’ . Nowadays, modern panoramic imaging techniques give us the possibility to detect some occult vasculitic processes in asymptomatic PMR patients with important implications for treatment and prognosis.
Moreover, in 1985, McCarthy identified a remitting seronegative symmetric synovitis with pitting oedema (RS3PE) as an overlapping syndrome between PMR and RA .
Diagnostic criteria and epidemiology
The diagnostic criteria of PMR are not universally defined. The first evaluation of criteria for PMR was proposed by Bird in 1979 : mean age of the onset at 65 years; clinical criteria included the presence of bilateral shoulder pain or stiffness and bilateral upper arm tenderness with an onset duration no longer than 2 weeks. In 1982, Chuang and colleagues defined new clinical criteria and shifted the age of onset to 50 years . Finally, 2 years later, Healey added the evaluation of the rapidity of response to prednisone to Chuang’s criteria, as previously established by Jones and Hazleman ( Table 1 ).
| Bird, 1979 [ ] | Chuang et al., 1982 [ ] | Healey, 1984 [ ] | |
|---|---|---|---|
| Age onset (years) | >65 | >50 | >50 |
| ESR (mm/h) | >40 | >40 | >40 |
| Peculiar clinical manifestations | Pain/stiffness involving shoulders bilaterally | Bilateral aching/stiffness ≥ 1 month involving at least 2 areas among: neck/torso, shoulders/proximal regions of arms, hip/proximal aspects of the thighs | Pain ≥ 1 month in at least 2 out of 3 specific areas: neck, shoulders, pelvic girdle |
| Duration of morning stiffness (hours) | >1 | – | >1 |
| Differential diagnosis | – | Other diagnosis have to be excluded, except GCA | Diagnosis of exclusion |
| Response to steroids | – | – | Rapid response to prednisone (≤20 mg /day) |
| Constitutional symptoms | Weight loss/depression | – | – |
| Other symptoms | Bilateral upper arm tenderness | – | – |
| Onset duration | <2 weeks | – | – |
| No. of criteria needed | ≥3 (sensitivity of 92% and specificity of 80%) | All | All |
The absence of univocal diagnostic criteria justified the lack of epidemiological studies about PMR and its correlation with GCA. We know that PMR is more frequent than GCA, it is more common in women, its exordium before 50 years of age is considered rare and the mortality rate does not vary from that of the general population of the same age and sex. The mean age of diagnosis is 72.8 years .
However, PMR and GCA are closely related syndromes: around 16–21% of the patients with PMR have underlying GCA, whereas 40–60% of biopsy-proven GCA patients have PMR manifestations. The highest incidence rates have been reported in Scandinavian countries or in populations of Scandinavian descent, with a progressive decrease of incidence in southern Europe countries .
The incidence increases with age but, differently from GCA, appears to be stable over the years, as observed in a population-based study evaluating the average annual age- and sex-adjusted incidence of PMR per 100 000 population over 50 years of age from 1970 through 1991 in Olmsted County, Minnesota : the annual incidence was 52.5, similar to what was observed in Goteborg, Sweden between 1985 and 1987 (50/100 000) and to what was found in Ribe Country, Denmark (68.3/100 000) from 1982 to 1985 .
By contrast, the annual incidence for patients with ‘pure’ PMR was 13.5/100 000 in Lugo, Spain , and 12.7/100 000 in Reggio Emilia, Italy , confirming the hypothesis of a north-to-south gradient. In a study from the UK, the age-adjusted incidence rate of PMR was 8.4/10 000 person-years, rising from 6.9/10 000 person-years in 1990 to 9.3/10 000 in 2001, signalling a possible variation in the incidence of the disease not only for GCA but also for PMR . The prevalence has been estimated at 1 case per 133 population according to Olmsted County analysis.
Recently, the prevalence of PMR has been analysed using population-based administrative data, taking into consideration errors associated with case ascertainment approaches when using these databases in the United Kingdom . This study suggests a higher prevalence of PMR in rural versus urban regions.
The development of universally accepted diagnostic criteria, perhaps including new biomarkers specifically related to PMR, would be useful to clarify not only the clinical aspects but also the epidemiological pattern of this disease in the near future.
Aetiopathogenesis
The aetiopathogenesis of PMR remains unknown, although the knowledge about the mechanisms of the inflammation has made significant steps forward, highly supporting the suggestion of the essential role of both environmental and genetic factors.
Genetic factors
The first survey on familial aggregation of GCA and PMR goes back to 1974 and, since then, scientific literature has accumulated data pointing to a genetic predisposition. By contrast, in a recent review on familial aggregation in PRM/GCA, Liozon and colleagues found no familial cases in their personal experience among 128 patients diagnosed with ‘pure’ PMR . Despite the presence of some conflicting data, PMR is considered a polygenic disease.
The role of the HLA–DRB1 locus in genetic patterns of PMR has been widely investigated, both in the ‘pure’ form and in association with GCA. In 1998 the association between isolated PMR and HLA-DRB1*13/14 (and previously DR6) was proved in a series of 86 patients affected, whereas the frequency of HLA-DRB1*0401 and *0404 alleles was only marginally increased. In general, the distribution of DRB1*04 is more frequent in GCA than in PMR ; nevertheless, relapses of PMR in isolated PMR patients were more frequently observed in those carrying HLA-DRB1*04 alleles, particularly *0401 . Moreover, the association of both HLA-DRB1*0401 and the ICAM-1 codon 241 GG’s homozygosity was significantly associated with increased risk of relapse in a series of isolated PMR patients from northwestern Spain .
The environmental factors
There is evidence that environmental factors, possibly viral, could trigger the disease’s onset in up to one-fourth of cases . Epidemiological studies about the seasonal distribution of the incidence of PMR are not consistent in individuating a regular cyclical pattern of yearly incidence.
In a work on the seasonal pattern of PMR the statistical analysis by χ2 testing showed a significantly increased frequency of onset in winter, while rhythm analysis identified a significant peak of occurrence in January; although the data corroborated the hypothesis of a precipitating environmental factor, yet, direct evidence of active viral or bacterial infections in temporal artery biopsies or in the synovial tissue is lacking. Nevertheless, the procalcitonin levels, as an early marker of bacterial infection, were normal in all GCA and PMR patients, suggesting no role of bacterial triggers for GCA or PMR .
To detect the possible role of influenza vaccination as a trigger of PMR/GCA, causing the so-called ‘ASIA’ syndrome (i.e., autoimmune/inflammatory syndrome induced by adjuvants), we reported a systematic analysis of our case series . As many as 10/20 cases of previously healthy subjects developed GCA/PMR within 2–3 months after influenza vaccination during an observation period of 6 years ( Table 2 ). A number of similar cases in the literature supported the real possibility that influenza vaccination may trigger GCA/PMR in healthy subjects, because of a possible genetic predisposition, with a latency period of up to 3 months. For this reason, we recommend a systematic research of previous vaccinations in patients with recent onset of GCA/PMR.
| Observational Period: 2005–2010 Patient No | Sex, Age (yrs) a | Disease | Vaccination | Time interval b (wk/mo) | HLA Typing |
|---|---|---|---|---|---|
| 1 | F, 80 | GCA/PMR c | INF-V | 3 mo | NA |
| 2 | F, 64 | PMR | INF-V | 1 mo | DR11, DR15, DQ1, DQ7 |
| 3 | F, 78 | GCA/PMR c | INF-V | 3 wk | NA |
| 4 | F, 67 | GCA c | INF-V | 1 mo | NA |
| 5 | F, 78 | GCA c | INF-V | 2 mo | NA |
| 6 | F, 71 | PMR | INF-V | 2 mo | NA |
| PMR Relapse d | INF-V (2 yrs later) | 2 wk | |||
| 7 | F, 80 | GCA c | INF-V | 2 mo | DR4, DR8, DR53, DQ4, DQ8 |
| 8 | F, 73 | GCA c | INF-V | 3 mo | NA |
| 9 | F, 70 | GCA c | INF-V | 3 mo | NA |
| 10 | F, 74 | GCA c | INF-V | 1 mo | NA |
a Age (years) at the time of onset of the disease.
b Time interval between vaccine administration and the onset of the disease (weeks/months). INF-V, influenza vaccination. NA, not available.
d Case n°6 had a relapse two years after the clinical remission, when another seasonal INF-V was administered.
The role of cytokines
The knowledge about the role of circulating cytokines in PMR and GCA is considered pivotal, to provide important insights into the comprehension of the pathogenetic mechanisms of both diseases and their differences in clinical manifestations, as well as for the implications of treatment strategies. Multiple circulating cytokines and their polymorphisms have been studied: interleukin (IL)-1 and the IL-1 receptor antagonist (IL-1ra) family, IL-2, IL-10, interferon (IFN)γ, IL-4, tumour necrosis factor (TNF)α, IL-6; nowadays the role of circulating IL-6 over other cytokines appears dominant. The hyper-production of IL-6 in PMR and GCA has been widely confirmed and it has been suggested to use IL-6 levels as a marker for monitoring patients during treatment . Furthermore, over-expression of IL-10 as a suppressor of inflammation in patients with PMR supports the hypothesis of a ‘protective’ role for the development of arteritis, in which a genetic down-regulation of IL-10 synthesis has been found.
The role of endocrinosenescence as a possible pathogenetic factor in PMR has also been investigated. It has been supposed that hypothalamic–pituitary–adrenal axis response could be inadequate in relation to the pro-inflammatory status and increased levels of IL-6, but evidence that these abnormalities are cause and not consequence of chronic inflammation is still lacking , as well as data regarding circadian variations of TNF-α and IL-6 secretions in PMR.
Indeed, literature about the role of innate immunity in PMR is limited to the expression and function of human toll-like receptors (TLRs). As many as 70 patients affected by PMR and GCA at different stages of the disease were investigated; a significantly increased expression of TLR7 in circulating T, B and mononuclear cells of patients with PMR and GCA was found, whereas it appeared reversible to normal levels after corticosteroid therapy; despite this evidence, circulating monocytes from patients with active disease showed a significantly lower in vitro response for the production of pro-inflammatory cytokines. It remains unclear if these results represent the consequence of the marked inflammatory process that occurs in these syndromes . Also, no association with TLR4 expression of B-cells, T-cells or monocytes or a distinct phenotype of TLR4 response was found in PMR and elderly onset rheumatoid arthritis (EORA) patients .
Histopathology
In a specimen of synovial membranes of the proximal joints and periarticular structures, a synovitis with a predominance of macrophage and CD4+ lymphocytes can be observed; the production of inflammatory cytokines by the macrophages is responsible for systemic manifestations of PMR. Furthermore, the absence of B-cells and γδ T-cells differentiates the synovitis pattern of PMR from that of RA. In 1994, Weyand et al. demonstrated that in the PMR artery walls adventitial dendritic cells are in an activated state; both IL-1 and IL-6 are detectable in temporal arteries but, differently from GCA, T-cells producing IFN-γ are not recruited into vascular tissue and there is no histological evidence of arteritis in PMR . In any event, the subclinical vasculitic process in affected patients is suggested by findings of positron emission tomography (PET) with fluoro-18-deoxyglucose (FDG) in detecting the vessels involvement in PMR . It is possible that subclinical vasculitis may be present in extra-cranial arteries of PMR patients and that IFN-γ production may be crucial to the development of overt arteritis.
Aetiopathogenesis
The aetiopathogenesis of PMR remains unknown, although the knowledge about the mechanisms of the inflammation has made significant steps forward, highly supporting the suggestion of the essential role of both environmental and genetic factors.
Genetic factors
The first survey on familial aggregation of GCA and PMR goes back to 1974 and, since then, scientific literature has accumulated data pointing to a genetic predisposition. By contrast, in a recent review on familial aggregation in PRM/GCA, Liozon and colleagues found no familial cases in their personal experience among 128 patients diagnosed with ‘pure’ PMR . Despite the presence of some conflicting data, PMR is considered a polygenic disease.
The role of the HLA–DRB1 locus in genetic patterns of PMR has been widely investigated, both in the ‘pure’ form and in association with GCA. In 1998 the association between isolated PMR and HLA-DRB1*13/14 (and previously DR6) was proved in a series of 86 patients affected, whereas the frequency of HLA-DRB1*0401 and *0404 alleles was only marginally increased. In general, the distribution of DRB1*04 is more frequent in GCA than in PMR ; nevertheless, relapses of PMR in isolated PMR patients were more frequently observed in those carrying HLA-DRB1*04 alleles, particularly *0401 . Moreover, the association of both HLA-DRB1*0401 and the ICAM-1 codon 241 GG’s homozygosity was significantly associated with increased risk of relapse in a series of isolated PMR patients from northwestern Spain .
The environmental factors
There is evidence that environmental factors, possibly viral, could trigger the disease’s onset in up to one-fourth of cases . Epidemiological studies about the seasonal distribution of the incidence of PMR are not consistent in individuating a regular cyclical pattern of yearly incidence.
In a work on the seasonal pattern of PMR the statistical analysis by χ2 testing showed a significantly increased frequency of onset in winter, while rhythm analysis identified a significant peak of occurrence in January; although the data corroborated the hypothesis of a precipitating environmental factor, yet, direct evidence of active viral or bacterial infections in temporal artery biopsies or in the synovial tissue is lacking. Nevertheless, the procalcitonin levels, as an early marker of bacterial infection, were normal in all GCA and PMR patients, suggesting no role of bacterial triggers for GCA or PMR .
To detect the possible role of influenza vaccination as a trigger of PMR/GCA, causing the so-called ‘ASIA’ syndrome (i.e., autoimmune/inflammatory syndrome induced by adjuvants), we reported a systematic analysis of our case series . As many as 10/20 cases of previously healthy subjects developed GCA/PMR within 2–3 months after influenza vaccination during an observation period of 6 years ( Table 2 ). A number of similar cases in the literature supported the real possibility that influenza vaccination may trigger GCA/PMR in healthy subjects, because of a possible genetic predisposition, with a latency period of up to 3 months. For this reason, we recommend a systematic research of previous vaccinations in patients with recent onset of GCA/PMR.
| Observational Period: 2005–2010 Patient No | Sex, Age (yrs) a | Disease | Vaccination | Time interval b (wk/mo) | HLA Typing |
|---|---|---|---|---|---|
| 1 | F, 80 | GCA/PMR c | INF-V | 3 mo | NA |
| 2 | F, 64 | PMR | INF-V | 1 mo | DR11, DR15, DQ1, DQ7 |
| 3 | F, 78 | GCA/PMR c | INF-V | 3 wk | NA |
| 4 | F, 67 | GCA c | INF-V | 1 mo | NA |
| 5 | F, 78 | GCA c | INF-V | 2 mo | NA |
| 6 | F, 71 | PMR | INF-V | 2 mo | NA |
| PMR Relapse d | INF-V (2 yrs later) | 2 wk | |||
| 7 | F, 80 | GCA c | INF-V | 2 mo | DR4, DR8, DR53, DQ4, DQ8 |
| 8 | F, 73 | GCA c | INF-V | 3 mo | NA |
| 9 | F, 70 | GCA c | INF-V | 3 mo | NA |
| 10 | F, 74 | GCA c | INF-V | 1 mo | NA |
a Age (years) at the time of onset of the disease.
b Time interval between vaccine administration and the onset of the disease (weeks/months). INF-V, influenza vaccination. NA, not available.
d Case n°6 had a relapse two years after the clinical remission, when another seasonal INF-V was administered.
The role of cytokines
The knowledge about the role of circulating cytokines in PMR and GCA is considered pivotal, to provide important insights into the comprehension of the pathogenetic mechanisms of both diseases and their differences in clinical manifestations, as well as for the implications of treatment strategies. Multiple circulating cytokines and their polymorphisms have been studied: interleukin (IL)-1 and the IL-1 receptor antagonist (IL-1ra) family, IL-2, IL-10, interferon (IFN)γ, IL-4, tumour necrosis factor (TNF)α, IL-6; nowadays the role of circulating IL-6 over other cytokines appears dominant. The hyper-production of IL-6 in PMR and GCA has been widely confirmed and it has been suggested to use IL-6 levels as a marker for monitoring patients during treatment . Furthermore, over-expression of IL-10 as a suppressor of inflammation in patients with PMR supports the hypothesis of a ‘protective’ role for the development of arteritis, in which a genetic down-regulation of IL-10 synthesis has been found.
The role of endocrinosenescence as a possible pathogenetic factor in PMR has also been investigated. It has been supposed that hypothalamic–pituitary–adrenal axis response could be inadequate in relation to the pro-inflammatory status and increased levels of IL-6, but evidence that these abnormalities are cause and not consequence of chronic inflammation is still lacking , as well as data regarding circadian variations of TNF-α and IL-6 secretions in PMR.
Indeed, literature about the role of innate immunity in PMR is limited to the expression and function of human toll-like receptors (TLRs). As many as 70 patients affected by PMR and GCA at different stages of the disease were investigated; a significantly increased expression of TLR7 in circulating T, B and mononuclear cells of patients with PMR and GCA was found, whereas it appeared reversible to normal levels after corticosteroid therapy; despite this evidence, circulating monocytes from patients with active disease showed a significantly lower in vitro response for the production of pro-inflammatory cytokines. It remains unclear if these results represent the consequence of the marked inflammatory process that occurs in these syndromes . Also, no association with TLR4 expression of B-cells, T-cells or monocytes or a distinct phenotype of TLR4 response was found in PMR and elderly onset rheumatoid arthritis (EORA) patients .
Histopathology
In a specimen of synovial membranes of the proximal joints and periarticular structures, a synovitis with a predominance of macrophage and CD4+ lymphocytes can be observed; the production of inflammatory cytokines by the macrophages is responsible for systemic manifestations of PMR. Furthermore, the absence of B-cells and γδ T-cells differentiates the synovitis pattern of PMR from that of RA. In 1994, Weyand et al. demonstrated that in the PMR artery walls adventitial dendritic cells are in an activated state; both IL-1 and IL-6 are detectable in temporal arteries but, differently from GCA, T-cells producing IFN-γ are not recruited into vascular tissue and there is no histological evidence of arteritis in PMR . In any event, the subclinical vasculitic process in affected patients is suggested by findings of positron emission tomography (PET) with fluoro-18-deoxyglucose (FDG) in detecting the vessels involvement in PMR . It is possible that subclinical vasculitis may be present in extra-cranial arteries of PMR patients and that IFN-γ production may be crucial to the development of overt arteritis.
Related posts:
Mixed connective tissue disease: An overview of clinical manifestations, diagnosis and treatment
Undifferentiated CTD: A wide spectrum of autoimmune diseases
Idiopathic inflammatory myopathies
Neurologic and other systemic manifestations in FMF: Published and own experience
Psoriatic arthritis
Tocilizumab – A novel therapy for non-organ-specific autoimmune diseases
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
