Giant cell arteritis: epidemiological clues to its pathogenesis and an update on its treatment
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Rheumatology 2003;42:413–421
doi:10.1093/rheumatology/keg116, available online at www.rheumatology.oupjournals.org
Review
Giant cell arteritis: epidemiological clues to its
pathogenesis and an update on its treatment
E. Nordborg and C. Nordborg1
Giant cell arteritis (GCA) is a chronic systemic vasculitis with a marked female
predominance and restriction to old age. The disease process distinctly targets
large and medium sized arteries, preferentially the aorta and its extracranial
branches. Morphological observations indicate that the age and sex distribution of
GCA is related to the occurrence of degenerative changes in the arterial wall. GCA
is not a truly infectious vasculitis. However, an infection might be a triggering
factor. Different centres report an increase in GCA incidence, but annual
fluctuations have not been shown to be statistically significant. However, significant
seasonal variations have been observed by several groups. The mortality is not
increased in adequately treated patients. Although, alternative steroid-sparing
agents have been proposed, corticosteroids are still the first treatment choice.
Introduction adventitia was proposed to be the centre of the immune
response, with the vasa vasorum being the port of
Giant cell (temporal) arteritis (GCA) is a chronic, sys- entrance of the antigen-presenting cells. The adventitial
temic vasculitis, with a distinct tropism for large and
macrophages and T lymphocytes produce high levels of
medium-sized arteries with well-developed elastic mem-
cytokines, thereby promoting further inflammatory
branes. The inflammatory process preferentially involves
the aorta and its extracranial branches, of which the reaction, but not tissue destruction. The macrophages of
external carotid artery with its superficial temporal divi- the media, on the other hand, produce metalloprotein-
sion, in particular, is affected. Although major advances ases and oxygen radicals, leading to the disintegration
have been made in recent years in genetics, molecular of elastic laminae and further injury of the vessel wall.
biology and the description of the vessel wall morpho- The tissue cytokine patterns of the temporal artery were
logy, the aetiology and pathogenesis of GCA are still correlated with clinical phenotypes of the disease. Thus,
incompletely understood. A single cause or aetiological high levels of the cytokine interferon-c (IFN-c) corre-
agent has not as yet been identified. lated with cranial symptoms, whereas patients with sys-
The epidemiology of GCA suggests striking diffe- temic symptoms only, displayed low levels of this cytokine
rences in disease risk among ethnic groups, with the w9x. Growth factors, such as platelet-derived growth
highest incidence rates measured in Scandinavia and in factor (PDGF) and vascular endothelial growth factor
subjects of Northern European descent, irrespective of (VEGF) are both amply expressed in the inflammatory
their place of residence w1–5x. infiltrate, stimulating intimal hyperplasia. Interestingly,
Genetic factors appear to be of importance for the these factors were also produced by the multinucleated
development of this disease, with a predominance of giant cells, which are, therefore, not only removers of
certain variants of HLA-DR4 allele expression w6, 7x. debris but also secretory.
Histologically, the inflammatory reaction is granulo- Thus, the vascular pathology in GCA is the result of
matous with highly activated macrophages and T lym- immunological injury to the vessel wall, as well as
phocytes, of which CD4+ T cells are in the majority. stromal response within the arterial wall w10x. Moreover,
Despite its name, giant cells are not a prerequisite for the significant media atrophy and calcification of the internal
diagnosis. The local activation of CD4+ T cells in the elastic membrane (IEM) appear to be prerequisites
outer layers (adventitia) of the vessel wall is suggestive of for its occurrence w11x.
an antigen-driven disease w8x. The possible antigen might The hallmarks of GCA are the systemic inflamma-
be of external origin, but it may also be autologous. The tion and the inflammatory infiltrate of the vessel wall,
Departments of Rheumatology and 1Pathology, Sahlgrenska University Hospital, SE-413 45, Göteborg, Sweden.
Submitted 20 September 2001; revised version accepted 4 September 2002.
Correspondence to: E. Nordborg. E-mail: nordborg@swipnet.se
413 ß 2003 British Society for Rheumatology414 E. Nordborg and C. Nordborg
resulting in luminal narrowing and end-organ ischaemia. been clarified. Recently, it was demonstrated that elderly
The most feared acute complications include blindness people who failed to produce specific Abs following
and infarcts of various vascular territories, whereas the influenza vaccination showed a predominance of CD8+
development of mural weakness, resulting in aortic CD282 T cells. The aetiology of the CD282 T-cell
dissection, has been considered a late manifestation w12x. recruitment is not fully understood in healthy indivi-
Modern history of giant cell arteritis runs along two duals, but there is evidence that these cells might be a
paths; that of temporal arteritis (TA) and that of response to continued antigenic stimulation. An increased
polymyalgia rheumatica (PMR). There is still a con- number of autoreactive CD8+ CD282 T cells leads to
troversy over whether PMR and GCA (TA) are linked the production of large amounts of IFN-c, which might
entities, and specifically if PMR is a vasculitis. Several trigger an imbalance in the production of T helper 1
contributions over the last decade have indicated a simi- (TH1) and TH2 cytokines, and a polarization towards the
lar pathogenetic process in the two conditions. Analyses TH1 effector type with higher age. Furthermore, in the
of the temporal artery biopsies have shown similar elderly there is a reduced production of adrenal and
patterns of T-cell and macrophage-derived cytokines gonadal steroids, resulting in less inhibitory effects on
winterleukin (IL)-1b, IL-6, transforming growth factor-b pro-inflammatory cytokine production. Changes have
(TGF-b), IL-2x in biopsy-negative PMR patients as well also been reported in the activity and reactivity of the
as in patients with biopsy-proven GCA, but not in age- hypothalamus–pituitary–adrenal axis w16x. It has been
matched controls. However, IFN-c was not found in speculated that an imbalance in the production of
PMR patients, but only in the TA patients, indicating pro- and anti-inflammatory cytokines could reduce the
that IFN-c might be crucial for the development of an protection against infections in elderly people and
overt granulomatous process w9x. These data indicate also increase the risk of developing other age-related
that PMR patients have a subclinical vasculitis and disorders such as Alzheimer’s disease and atherosclerosis.
therefore PMR has been regarded as a ‘forme fruste’ with Exactly how the ageing of the immune system affects
minor vascular involvement w9x. According to a recent immune reactivity in a patient with GCA has not been
study, using positron emission tomography (PET) with investigated.
fluoro-18-deoxyglycose (18F-glycose), there was a sig-
nificantly increased vascular uptake in the large thoracic Gender
arteries (aorta, subclavian and carotid arteries) also in There is a clear female predominance in GCA, with 2- to
PMR patients without clinical or morphological evidence 4-fold more women than men w3–5, 14, 19, 20x. This
of inflammation in temporal arteries w13x. This inves- difference appears to be more marked in the northern
tigation gave further support to the contention that parts of Europe. Only in one Spanish study were males
PMR and TA are two different expressions of the same reported to be predominant in GCA w21x. Evidence was
underlying disorder and that the large arteries, and not recently presented indicating that other types of primary
merely the temporal arteries, may be affected in both vasculitides display an inverse gender preference. An
conditions. Therefore, in this review GCA is used to epidemiological review of Wegener’s granulomatosis,
signify all different clinical manifestations of this disease, Churg–Strauss syndrome and polyarteritis nodosa in the
also including biopsy-negative PMR. United Kingdom, Spain and Norway showed that, in all
This review will focus on potential epidemiological areas and in all disease categories, the incidence was
clues and issues regarding the aetiopathogenesis of GCA. higher in men than women, with a peak incidence at the
age of 65–74 yr w22x. These vasculitides affect medium-
sized and small vessels, in contrast to GCA, which
Risk factors affects large and medium-sized arteries. Consequently,
although all these disorders display a peak incidence in
Age elderly people, there appear to be different gender-
GCA is markedly age-restricted. Essentially, no cases related factors behind the initial immune stimulation in
younger than 50 yr of age have been identified and the the two disease groups.
likelihood of being diagnosed with this disease increases
continuously with age. GCA is about 20 times more Genetic factors
common among people in their 9th decade compared The possibility of a genetic influence on GCA suscepti-
with people aged between 50 and 60 yr w14x, which may bility was initially supported by reports of cases among
indicate that its pathogenesis is related to the ageing of first-degree relatives. Several studies have shown an
the arterial wall w14x. Immunosenescence, the term used association of GCA incidence, and risk of visual com-
for changes in the immune system with ageing, implies a plications, with the HLA-DRB1*-04 alleles w23–25x.
decline in immunocompetence, resulting in an increased However, in isolated PMR, without GCA symptoms,
risk of infections and autoimmuneuinflammatory dis- the HLA class II expression varied from one population
orders. Furthermore, a decreased antibody (Ab) produc- to another w26x. Genetic polymorphisms with regard
tion and a shortened duration of protective immunity to the expression of tumour necrosis factor (TNF),
following immunization are characteristic features in the intercellular adhesion molecule (ICAM-1), regulated
elderly w15x. Despite much research in this field, basic on activation, normal T-cell expressed and secreted
mechanisms of age-related immune dysfunction have not (RANTES) and interleukin receptor antagonistGiant cell arteritis: pathogenesis and treatment 415
(IL-1Ra) were also shown to influence the susceptibility state during pregnancy protects the artery wall w32x.
for GCA and PMR, irrespective of DRB1 type w26x. Oestrogen is involved in a wide variety of different
mechanisms which, theoretically, may be related to
GCA and arterial degeneration GCA. It is thus known to preserve a normal vessel
Atherosclerosis affecting large and medium-sized arteries wall by stimulating as well as inhibiting the growth of
accounts for at least half of all deaths in the Western vascular smooth-muscle cells w33–37x and there is
world. There is growing evidence that its pathogenesis is evidence that it influences the immune system w38x. One
driven by inflammatory mechanisms similar to those in recent study showed that mononuclear and giant cells in
rheumatic disorders (RA) w27x. Geographically, the mor- GCA display the cytoplasmic accumulation of oestrogen
tality and morbidity of ischaemic heart disease is con- receptor-a (ER-a). Cytoplasmic ER-a was also seen in
siderably lower in the Mediterranean areas compared media smooth muscle in GCA and in non-GCA controls.
with countries in Northern Europe. There is a clear male The nucleotide sequence analysis of the ER-a gene
predominance in ischaemic heart disease in all European revealed no differences between GCA patients and
countries w28x and this gender difference persists through- controls w39, 40x. Whether the reduction in circulating
out life. Apart from gender, the size of the affected oestrogen in post-menopausal women plays a role in the
vessels, the age of the patients and the geographical development of the asymmetrical loss of smooth-muscle
distribution are similar to those of GCA. However, cells in the temporal arteries and IEM calcification,
morphologically, there is no reason to believe that athero- which appear to be a prerequisite for the disorder,
sclerosis is a risk factor or is an initiator of GCA. requires further investigation.
Atherosclerosis is seldom seen in temporal arteries.
Instead, in the general population, a morphologically Environmental factors
distinct type of arterial degeneration is encountered, Since fever, high erythrocyte sedimentation rate (ESR)
indicating media atrophy and minute calcification of the and general illness are some of its clinical symptoms,
internal elastic membrane (IEM) w29x. The age and sex GCA has repeatedly been suggested to be infectious.
distribution of these calcifications is similar to that of Clustering in disease incidence, as well as rhythmic
GCA, i.e. they are more common in women and they annual and seasonal fluctuations in incidence, may
increase with age w20x. Using light and electron micro- indicate the involvement of epidemic infections or other
scopy, the inflammatory process of GCA appears to exogenous aetiological factors. A study from Olmsted
start with the formation of foreign-body giant cells, County described a cyclic pattern in the annual incidence
which attack the IEM calcifications. The presence of during the period 1950–1991, with peaks every 7th yr,
this degenerative lesion in the general population might each period lasting for about 3 yr w19x. Due to the rela-
thus explain the age and sex distribution of the disorder. tively small number of cases (n=125), the statistical
However, it remains to be elucidated why only a minor- power was low. The cyclic annual fluctuation was not
ity of the population react against their IEM calcifica- confirmed in a large series (n=665) of biopsy-positive
tions. It may be speculated that infections might play a GCA in Göteborg in 1976–1995. However, there was a
role by activating the immune system or, more directly, significant seasonal variation with peaks in late winter
by inducing the formation of the foreign-body giant cells and autumn w41x. Other studies report peaks in clinical
w11, 29x. onset during the summer or winteruspringtime w42–45x.
A few epidemiological studies have addressed the A Danish study reported a clustering of GCA in relation
question of whether there might be an association to two epidemics of Mycoplasma pneumoniae and two
between degenerative vascular disease and GCA. In a major outbreaks of human parvovirus B19 w46, 47x.
population-based case–control study, Machado et al. w30x Interestingly, human parvovirus B19, which infects endo-
reported that smoking was a statistically significant risk thelial cells, has been reported in connection with other
factor for GCA wodds ratio (OR) 2.3, 95% confidence vasculitides, such as Wegener’s granulomatosis and
interval (CI)=1.3–4.1x, but manifestations of athero- cerebral vasculitis in children w48, 49x.
sclerosis, such as hypertension, angina, myocardial Serological studies, including influenza A and B,
infarction, peripheral vascular disease or stroke, were mumps, adeno-, rota- and enterovirus, Q-fever, lepto-
not significantly related to GCA. However, the limited spirosis, M. pneumoniae and Chlamydia, most of them
number of included cases reduced the power of that performed on small series of patients, revealed no
study. A French multicentre, prospective, case–control difference in seroprevalence between GCA cases and
study confirmed the association between smoking and controls w50x.
GCA w31x. The seroprevalence was also similar in cases and
controls regarding viruses known to induce giant cells in
Female sex hormones humans, i.e. the herpes virus group, Epstein–Barr virus,
Since GCA is a disorder among elderly females, the measles, respiratory syncytial virus and parainfluenza
question might be raised of whether sex hormones are types 1, 2 and 3 w50x. One French study suggested that
involved in its pathogenesis. Interestingly, a recent newly diagnosed cases of GCA were more likely to be
epidemiological study revealed that the number of preg- positive for IgM anti-human parainfluenza virus type 1,
nancies was lower among GCA patients than among compared with randomly selected sex- and age-matched
controls. It was suggested that the hyperoestrogenic controls from the general population w51x. Varicella416 E. Nordborg and C. Nordborg
zoster virus (VZV) produces multinucleated giant cells in 1973–1975, 1977–1986 and 1976–1995 w3, 14, 41x
acutely infected tissue w52x and is commonly reactivated demonstrated a statistically significant increase with an
in elderly people. A study of biopsy-positive temporal average annual incidence of biopsy-positive GCA of
arteries, using immunohistochemical and polymerase 16.8, 18.3 and 22.2u100 000, respectively. The relation-
chain reaction (PCR) techniques, did not reveal VZV ship between positive and negative biopsies was constant
antigen or DNA in any of the subjects w53x. during this period w41x. A similar trend was suggested in
Several studies associate Chlamydia pneumoniae Olmsted County, Minnesota w2x. Such observations might
(TWAR) organisms with the atheromatous plaque, reflect genuine changes in morbidity, but they could also
although their pathogenetic role is unclear w54–58x. The be related to greater awareness of the disease or better
role of C. pneumoniae in GCA was recently addressed. identification of cases. Other independent causes that
Wagner et al. w59x, detected C. pneumoniae, using might exert an influence are differences in the catchment
immunocytochemistry and PCR techniques, in the area of the population studied or an increase in the age
majority of their GCA patients but in none of the con- of the general population. However, our recent statistical
trols. In contrast, Haugeberg et al. w60x, using the same investigation showed that the increase in GCA incidence
techniques on a southern Norwegian population, were was not significantly influenced by the increasing age
not able to detect C. pneumoniae in any of 20 patients of the Göteborg population (Nordborg et al., in
with biopsy-proven GCA. preparation). Furthermore, the study design might differ
Chlamydia pneumoniae is a common infection world- with a shift from retrospective to prospective studies.
wide w61x and there is a clear sex distribution showing a The Göteborg studies were all retrospective in design,
considerably higher seroprevalence in men compared involving the same catchment area.
with women in older age groups w61, 62x. However, low As is indicated by the large number of temporal artery
rates were found in Denmark and Norway w62x. biopsies performed, physicians are definitely aware of
Therefore, C. pneumoniae infection does not relate statisti- this disease. On the other hand, taking the data from
cally to the high prevalence of GCA in Scandinavian Östberg w70x into consideration, GCA may be under-
women. diagnosed during life. She found a prevalence of 1.7% in
To summarize, there is no convincing evidence today her post-mortem study, including more than 20 000
that GCA is a truly infectious vasculitis, although it may cases, using the following histological criteria: ‘intimal
be speculated that environmental factors such as infec- thickening with narrowing of the arterial lumen, ingrowth
tions might trigger the immune system in a susceptible of capillaries in media anduor intima, destruction of the
host, as suggested by Russo et al. w63x, who, in a clinical elastic membrane with accumulation of mononuclear
retrospective study, found a correlation between various round cells anduor giant cells’.
infections and the onset of GCA. Recently, a geographical gradient of frequencies in GCA
was reported with a statistically verified increase in
frequency by latitude north w22, 71x. The geoepidemiology
Occurrence of other primary systemic vasculitides, such as Wegener’s
granulomatosis, Churg–Strauss syndrome, polyarteritis
Incidence rates nodosa and microscopic polyangitis, was recently com-
The incidence of GCA varies greatly in different geo- pared in different European regions w22x; there are
graphical areas. It has repeatedly been shown that the differences between geographical areas also in the inci-
disease predominately affects subjects of Northern dence of small vessel vasculitides. Wegener’s granulo-
European descent, in particular those of Nordic heritage, matosis was thus found to be more common in the UK
irrespective of their place of residence w5–13x, with than in Spain, with a trend towards even higher incidence
estimates of about 20 cases annually per 100 000 persons rates in Tromsö (northern Norway), indicating that envi-
older than 50 yr of age. The incidence rates are lower in ronmental and genetic factors might also be important in
Southern Europe w64, 65x. Only a few cases are reported the aetiopathogenesis of these types of vasculitis.
in Israel and in black populations w66, 67x, while in Asian
countries GCA is distinctly infrequent.
The highest figures worldwide were documented from Mortality
Southern Norway w68x using the 1990 American College
of Rheumatology criteria w69x. The annual incidence rate GCA may be life-threatening. The aorta and its main
was 32.8u100 000 in individuals older than 50 yr, and branches (subclavian and carotid arteries) are the main
29.1u100 000 for biopsy-proven GCA, thus confirming targets for the arteritic process, but coronary arteries
the high incidence data reported by Gran et al. w5x. and cerebral arteries are sometimes involved. However,
Whether such figures reflect a high endemic clustering only few fatal cases are reported in the literature. The
of GCA in a distinct region or are instead the result of fact that GCA may cause cerebral and myocar-
a continuous increase in incidence rates with time is dial infarction as well as aortic aneurysms is not general
not known. knowledge. Fewer fatal GCA cases are probably
Clearly, the incidence of GCA has increased in recent detected today, due to decreasing autopsy rates and to
years, indicating time trends in morbidity rates. From the fact that post-mortem histological examination of
Göteborg, Sweden, three previous studies from the periods the arteries is not routinely performed.Giant cell arteritis: pathogenesis and treatment 417
Aortic lesions, which are mainly located in the thora- impairment. Among those who were treated within 24 h
cic segment, are asymptomatic in most patients and after loss of sight, an improvement was achieved in more
may therefore be overlooked. Furthermore, their mani- than 50%. In contrast, only 6% of the patients improved
festations evolve slowly and an aortic dissection may when treatment was delayed for more than 24 h w80x.
only be overt years after terminated treatment. In a retro- Schmidt et al. w81x emphasized the importance of an
spective study, Evans et al. w12x calculated an increased early diagnosis and prompt corticosteroid treatment.
risk of thoracic aortic aneurysms of 17.3 (95% CI 7.9–33) They reported six cases with severe vascular complica-
compared with the general population; the correspon- tions (bilateral blindness, cerebral strokes) on which
ding risk of an abdominal aneurysm was 2.4 (95% CI 0.8– corticosteroid treatment had no effect. The patient delay
5.5). In Sweden, aortic aneurysms were detected in 6u90 between first symptoms (PMR, jaw claudication, head-
patients after a median follow-up period of 11.3 yr w72x. ache, amaurosis fugax) and the vascular complication
The prevalence of myocardial infarction in GCA is was on average 7 weeks. When a vascular catastrophy is
unknown. The available literature is sparse and limited manifest, the corticosteroid therapy, whatever dose chosen,
to case reports. Cerebral infarction has been reported in may prevent another vascular incident but does not
about 7% of consecutive patients with biopsy-proven reverse the symptoms of the first accident.
GCA w73x. The preventive effect of corticosteroid treat- The optimal initial dose regimen of oral prednisolone,
ment on mortality has been documented in many long- which is the drug most frequently used, has been dis-
term follow-up studies w72, 74, 75x. Although one study cussed. According to prospective large series applying
reported increased mortality early in the disease course, acceptable diagnostic criteria, and using predefined
during the first 4 months after the diagnosis of biopsy- treatment protocols, 20–60 mg of prednisolone (mostly
proven GCA, the death rates were subsequently normal- 40–60 mg) was shown to be an appropriate starting
ized w76x. Schaufelberger et al. w77x noticed an increased dosage in about 90% of cases. Alternate-day adminis-
mortality during the first 2 yr of treatment in patients tration regimens of corticosteroids have not proved
with biopsy-negative PMR, but not thereafter (personal effective in GCA w82–84x and this treatment does not
communication). In the latter two studies the patients reduce the development of steroid-induced osteoporosis
were treated by many different physicians without special w85x. Treatment with intramuscular methylprednisolone
knowledge of GCA in contrast to the long follow-up was reported to result in a more beneficial side-effect
study from Göteborg w72x, where only doctors with a profile in patients with pure PMR in one study w86x,
special interest in the disease were involved.
although these data were not confirmed in a recent,
The risk of fatal complications in GCA stresses the
multicentre, prospective study of biopsy-proven GCA
importance of early diagnosis and of follow-up, also
followed up for 1 yr w87x. However, in the latter study,
after clinical remission.
steroids were administered as intravenous pulses, with
different kinetics, which might have influenced the result.
Occurrence of malignancies in GCA Due to the well-known comorbidities related to long-
Malignancy may be associated with polymyalgia term corticosteroid treatment w88x and the fact that
rheumatica-like symptoms w78x. However, cancer-related GCA has a remarkably high tendency to relapse during
symptoms differ from those of classic PMR; symptoms tapering, alternative corticosteroid-sparing therapy has
may appear before the age of 50 yr, they may be asym- been requested. Several approaches, including combined
metrical and there may be joint pain. Incomplete or therapy with cytotoxic (azathioprine, methotrexate and
delayed relief from corticosteroid treatment is also cyclosporin) agents, have been suggested. To date, the
characteristic. Several papers have addressed the poten- vast majority of studies in this field have not led to any
tial association between GCA and the incidence of malig- firm conclusions, since they are too restricted in number
nancy. One large study found that malignancies were 2.3 w89, 90x, short term w91–93x and biased by high drop-out
times more common in GCA patients compared with rates anduor they were uncontrolled w89–91x. So far,
the general population. Considering the long interval methotrexate is the drug that holds the best promise. The
between the diagnosis of malignancy and GCA, the results of a randomized, double-blind, 24-month, placebo-
relationship appeared weak w79x. So far, there is no proof controlled trial from Spain, published last year, revealed
of a causal association between GCAuPMR and neoplasm. a significant reduction in the rate of relapses and the
cumulative mean dose of corticosteroids was lower in the
methotrexate arm compared with the placebo group w94x.
Treatment Surprisingly, neither the rate nor the severity of adverse
GCA is a highly corticosteroid-responsive disorder. A events differed between the two groups. The effect on
majority of the patients experience an excellent thera- bone mineral density or fracture rate was not evaluated
peutic effect. Moreover, this treatment has a preventive and no information was provided about whether
effect on vascular complications, as has clearly been patients with more aggressive disease did better on this
shown in many long-term follow-up studies w72, 75x. treatment. A recent, multicentre, placebo-controlled trial
The time before the initiation of corticosteroid treat- w95x showed no benefit of methotrexate treatment in
ment was the single most important factor predicting terms of the number of relapses, treatment failures,
outcome in patients presenting with symptoms of visual cumulative steroid doses or treatment toxicities. However,418 E. Nordborg and C. Nordborg the number of patients included was small, which reduced be influenced, despite high doses of corticosteroid treat- the power of this study. ment. These observations provide an explanation for the New drugs such as TNF blockers have rapidly promptness of the therapeutic effect seen in glucocorti- emerged as efficient treatments in rheumatoid arthritis costeroid treatment, but they also indicate why patients w96x. Regarding their use in GCA, the experience is have to be treated for a long time. Persistent disease is limited to case reports, albeit with some encouraging evidenced by active histological lesions, as well as by the results. Cantini et al. w97x reported a complete response fact that patients develop aortic aneurysms even years in three of four patients with long-standing active giant after they were considered to be in remission w12x. cell arteritis, still requiring high doses of prednisolone after more than 42 months. All patients had developed Osteoporosis serious corticosteroid-related side-effects. After two infu- The age of the GCA population puts them at great risk sions with infliximab (3 mgukg), three of the patients for corticosteroid side-effects, of which osteoporosis displayed a clinical and humoral remission. The remis- leads to the most severe sequelae. In contrast to increased sion sustained after a third infusion and during a resorption, resulting from the disease process, the follow-up time of 6 months, despite withdrawal of the steroid-induced loss of bone is due to decreased for- corticosteroids. One patient, who did not respond to mation of bone. However, studies of corticosteroid- therapy, was withdrawn after the second infusion, in related osteoporosis morbidity in GCA have produced accordance with the protocol. The therapy was well contradictory results. Pearce et al. w101x reported that tolerated by all patients. No side-effects were reported. even small daily doses of 6 mg of prednisolone might Interestingly, the same group recently reported a simi- result in bone loss, initially in trabecular bone, such as lar good response with normalization of clinical and the spine, followed by cortical bone. On the other hand, serological activity after three infusions with infliximab in a recent Norwegian study the authors focused on bone in three of four patients with persistent PMR, without mineral density of the radius, spine and hip, measured by cranial symptoms w98x, whereas one patient had a partial dual-energy X-ray absorptiometry (DEXA) in GCA effect. In the good responders corticosteroid therapy was (PMR and TA). Patients currently or previously treated terminated and in the partial responder it was reduced with prednisolone were compared with 30 newly by 50%. The remission was sustained at the control 1 yr diagnosed cases of GCA (PMR or TA), examined after the first infusion. Infliximab was well tolerated; there prior to the start of corticosteroid therapy, and with were no side-effects. These open pilot studies suggest 70 healthy controls. For current users the mean daily that TNF-a blockade may have a steroid-sparing effect dose was 6.5 mg of prednisolone, the mean cumulative in patients with corticosteroid-resistant GCA. dose 7.7 g, and for previous users 5.6 mg and 6.6 g, In a critical review of treatment studies, it was respectively. No statistically significant difference was suggested that only about 10–15% of patients have found between any of the four groups w102x. A British a ‘corticosteroid-resistant’ disease, defined as a daily prospective study over 2 yr in patients with PMR w103x requirement of >15–20 mg of prednisolone more than showed increased levels of cross-links (resorption mar- 2 months after the start of therapy w99x. These patients kers) before the start of corticosteroids, and a significant require a safe and effective additional agent. Future decrease at 6 months. Bone mineral density (BMD) trials should focus on appropriately defined large at 1 yr correlated with ESR at baseline, but not with cohorts of patients, ideally biopsy-proven patients, who the cumulative dose of corticosteroids. The last study need >15 mg of prednisolone as chronic maintenance stresses the impact of the systemic inflammation per se treatment. The rate of remission in the long run is on bone mass. Thus, the morbidity of bone loss in GCA another important issue which must be focused on is multifactorial. If BMD is within the normal range, the increasingly in new treatment modalities. bone-sparing agents calcium and vitamin D should be The characteristic prompt relief of symptoms and given generously to patients at the start of treatment preventive effect on vascular complications after the ini- with corticosteroids. Bisphosphonates could be insti- tiation of corticosteroid therapy in the vast majority of tuted if BMD shows decreased bone density on the patients indicates the unique role of steroid-mediated DEXA scan. In patients for whom corticosteroids are an immunosuppression in GCA. Corticosteroids have been unavoidable necessity, always try to use the lowest regarded as the cornerstone in the therapy of GCA and possible dose to suppress the inflammation, at initiation today there is little evidence, if any, that they can or and during maintenance therapy. should be replaced. On the other hand, despite good clinical improvement in the systemic signs of the disease, Conclusions the inflammatory infiltrate persists for a long time in the vessel wall, which further emphasizes the need for GCA is suggested to be an antigen-driven disease, where optimized therapy in GCA. the antigen might be exogenous or endogenous. A single Recently, the biological action of corticosteroids was cause or aetiological antigen has not as yet been found. elucidated, using temporal arteries with biopsy-proven There is no proof that GCA is a truly infectious disease GCA explanted into immunodeficient SCID mice w100x. and, on morphological and epidemiological grounds, NFkB-dependent cytokines (IL-2, IL-1b, IL-6) were there is no reason to believe that atherosclerosis initiates suppressed, whereas TGF-b and IFN-c did not appear to or is a risk factor for GCA. For the future, efforts should
Giant cell arteritis: pathogenesis and treatment 419
be made to identify further relevant environmental, age- 16. Deuschle M, Gotthardt U, Schweiger U et al. With aging
related, genetic and, in particular, gender-specific risk in humans the activity of the hypothalamus–pituitary–
factors. adrenal system increases and its diurnal amplitude flattens.
Corticosteroids remain the cornerstone of drug treat- Life Sci 1997;61:2239–46.
17. Straub RH, Konecna L, Hrach S et al. Serum dehydro-
ment of GCA. Future large trials, divided into clinically epiandrosterone (DHEA) and DHEA sulfate are negatively
important subgroups, may provide information about correlated with serum interleukin-6 (IL-6), and DHEA
the optimal therapy for each patient with GCA. inhibits IL-6 secretion from mononuclear cells in man
Other approaches with new drugs, such as TNF-a in vitro: possible link between endocrinosenescence and
blockades, might hold promise but have not as yet been immunosenescence. J Clin Endocrinol Metab 1998;
tested in controlled trials in GCA. 83:2012–7.
18. Straub RH, Miller LE, Scholmerich J, Zietz B. Cytokines
and hormones as possible links between endocrinosen-
References escence and immunoscenescence. J Neuroimmunol 2000;
109:10–15.
1. Fauchald, Rygvold O, Öystese B. Temporal arteritis and 19. Salvarani C, Gabriel SE, O’Fallon WM, Hunder GG.
polymyalgia rheumatica. Ann Intern Med 1972;77:845–52. The incidence of giant cell arteritis in Olmsted County,
2. Machado EB, Michet CJ, Ballard DJ et al. Trends in Minnesota: apparent fluctuations in a cyclic pattern. Ann
incidence and clinical presentation of temporal arteritis in Intern Med 1995;123:192–4.
Olmsted County, Minnesota, 1950–1985. Arthritis Rheum 20. Nordborg C, Nordborg E, Petursdottir V, Fyhr IM.
1988;31:745–9. Calcification of the internal elastic membrane in temporal
3. Nordborg E, Bengtsson BA. Epidemiology of biopsy- arteries; its relation to age and gender. Clin Exp Rheumatol
proven giant cell arteritis (GCA). J Intern Med 1990; 2001;19:565–8.
227:233–6. 21. Gonzalez-Gay MA, Alonso MD, Aguero JJ, Bal M,
4. Baldursson O, Steinsson K, Björnsson J, Lie JT. Giant cell Fernandez CB, Sanchez AA. Temporal arteritis in a
arteritis in Iceland. An epidemiologic and histopathologic northwestern area of Spain: study of 57 biopsy proven
analysis. Arthritis Rheum 1994;37:1007–12. patients. J Rheumatol 1992;19:277–80.
5. Gran JT, Myklebust G. The incidence of polymyalgia 22. Watts RA, Gonzalez-Gay M, Garcia-Porrua C, Lane S,
rheumatica and temporal arteritis in the county of Aust Bentham G, Scott DGI. Geoepidemiology of systemic
Agder, south Norway: A prospective study 1987–94. vasculitis. Ann Rheum Dis 2001;60:170–2.
J Rheumatol 1997;24:1739–43. 23. Weyand CM, Hicock KC, Hunder GG, Goronzy JJ. The
6. Richardson JE, Gladman DD, Fam A, Keystone EC. HLADRB1 locus as a genetic component in giant cell
HLA-DR4 in giant cell arteritis: association with poly- arteritis: mapping of a disease-linked sequence motif to the
myalgia rheumatica syndrome. Arthritis Rheum 1987; antigen binding site of the HLA-DR molecule. J Clin
30:1293–7. Invest 1992;90:2355–6.
7. Weyand CM, Hunder NN, Hicok KC, Hunder GG, 24. Dababneh A, Gonzalez-Gay MA, Garcia-Porrua C, Hajeer
Goronzy JJ. HLA-DRB1 alleles in polymyalgia rheuma- A, Thomson W, Ollier W. Giant cell arteritis and poly-
tica, giant cell arteritis, and rheumatoid arthritis. Arthritis myalgia rheumatica can be differentiated by distinct patterns
Rheum 1994;37:514–20. of HLA class II association. J Rheumatol 1998;25:2140–5.
8. Andersson R, Jonsson R, Tarkowski A, Bengtsson BA, 25. Gonzalez-Gay MA, Garcia-Porrua C, Llorca J et al. Visual
Malmvall BE. T cell subsets and expression of immuno- manifestations of giant cell arteritis: trends and clinical
logical activation markers in the arterial walls of spectrum in 161 patients. Medicine (Baltimore) 2000;
patients with giant cell arteritis. Ann Rheum Dis 1987; 79:283–92.
46:915–23. 26. Gonzalez-Gay MA. Genetic epidemiology in giant cell arteri-
9. Wagner AD, Bjornsson J, Bartley GB, Goronzy JJ, tis and polymyalgia rheumatica. Arthritis Res 2001;3:154–7.
Weyand CM. Interferon-gamma-producing T-cells in giant 27. Wållberg-Jonson S, Johansson H, Ohman ML, Rantapää-
cell vasculitis represent a minority of tissue-infiltrating cells Dahlqvist S. Extent of inflammation predicts cardio-
and are located distant from the site of pathology. Am J vascular disease and overall mortality in seropositive
Pathol 1996;148:1925–33. rheumatoid arthritis. A retrospective cohort study from
10. Weyand CM, Goronzy JJ. Arterial wall injury in giant cell disease onset. J Rheumatol 1999;26:2562–71.
arteritis. Arthritis Rheum 1999;42:844–53. 28. Sans S, Kesteloot H, Kromhout D. The burden of
11. Nordborg E, Bengtsson BÅ, Nordborg C. Temporal artery cardiovascular diseases mortality in Europe. Task force
morphology and morphometry in giant cell arteritis. of the European Society of Cardiology on cardiovascular
APMIS 1991;99:1013–23. mortality and morbidity. Statistics in Europe. Eur Heart J
12. Evans ME, O’Fallon M, Hunder GG. Increased incidence 1997;18:1231–48.
of aneurysm and dissection in giant cell (temporal) 29. Nordborg C, Nordborg E, Petursdottir V. The patho-
arteritis. A population-based study. Ann Intern Med genesis of giant cell arteritis: morphological aspects. Clin
1995;122:502–7. Exp Rheumatol 2000;18(Suppl. 20):18–21.
13. Levine SM, Hellmann DB. Giant cell arteritis. Curr Opin 30. Machado EBV, Gabriel SE, Beard CM, Michet CJ,
Rheumatol 2002;14:3–10. O’Fallon WM, Ballard DJ. A population-based case–control
14. Bengtsson BA, Malmvall BE. The epidemiology of giant study of temporal arteritis: Evidence for an association
cell arteritis including temporal arteritis and polymyalgia between temporal arteritis and degenerative vascular
rheumatica. Incidences of different clinical presentations disease? Int J Epidemiol 1989;18:836–41.
and eye complications. Arthritis Rheum 1981;24:899–904. 31. Duhaut P, Pinede L, Demolombe-Rague S et al. Giant cell
15. Grubeck-Loebenstein B, Wick G. The aging of the immune arteritis and cardiovascular risk factors. Arthritis Rheum
system. Adv Immunol 2002;80:243–84. 1998;41:1960–5.420 E. Nordborg and C. Nordborg
32. Duhaut P, Pinede L, Demolombe-Rague S et al. Giant cell 50. Duhaut P, Bosshard S, Dumontet C. Giant cell arteritis
arteritis and polymyalgia rheumatica: Are pregnancies a and polymyalgia rheumatica: Role of viral infections. Clin
protective factor? A prospective, multicentre case–control Exp Rheumatol 2000;(Suppl. 20):22–3.
study. Rheumatology 1999;38:118–23. 51. Duhaut P, Bosshard S, Calvet A et al. Giant cell arteri-
33. Cheng LP, Kuwahara M, Jacobsson J, Foegh ML. tis, polymyalgia rheumatica and viral hypotheses: A multi-
Inhibition of myointimal hyperplasia and macrophage center, prospective case–control study. Groupe de Recherche
infiltration by estradiol in aorta allografts. Transplantation sur l’Arterite a Cellules Geantes. J Rheumatol 1999;26:361–9.
1991;52:967–72. 52. Gilden DH, Kleinschmidt-De Masters BK, Wellish M,
34. Foegh ML, Asotra S, Howell MH, Ramwell PW. Estradiol Hedley-Whyte ET, Rentier B, Mahalingam R. Varicella
inhibition of arterial neointimal hyperplasia after balloon zoster virus, a cause of waxing and waning vasculitis.
injury. J Vasc Surg 1994;19:722–6. NEJM case 5–1995 revisited. Neurology 1996;47:1441–6.
35. Karas RH, Patterson BL, Mendelsohn ME. Human 53. Nordborg C, Nordborg E, Petursdottir V et al. Search for
vascular smooth muscle cells contain functional estrogen varicella zoster virus in giant cell arteritis. Ann Neurol
receptor. Circulation 1994;89:1943–50. 1998;44:413–4.
36. Keyes LE, Moore LG, Walchak SJ, Dempsey EC. 54. Kou CC, Jackson LA, Campbell LA, Grayston JT.
Pregnancy-stimulated growth of vascular smooth muscle Chlamydia pneumoniae (TWAR). Review. Clin Microbiol
cells: importance of protein kinase C-dependent synergy Rev 1995;8:451–61.
between estrogen and platelet-derived growth factor. J Cell 55. Kuo CC, Shor A, Campbell LA, Fukushi H, Patton DL,
Physiol 1996;166:22–32. Grayston JT. Demonstration of Chlamydia pneumoniae in
37. Losordo DW, Kearney M, Kim EA, Jekanowski J, atherosclerotic lesions of coronary arteries. J Infect Dis
Isner JM. Variable expression of estrogen receptor in nor- 1993;167:841–9.
mal and atherosclerotic coronary arteries of premenopausal 56. Kuo CC, Gown AM, Benditt EP, Grayston JT. Detection
women. Circulation 1994;89:1501–10. of Chlamydia pneumoniae in aortic lesions of atherosclero-
38. Cutolo M, Sulli A, Seriolo B, Accardo S, Masi AT. sis by immunocytochemical staining. Arterioscler Thromb
Estrogens, the immune response and autoimmunity. Clin 1993;13:1501–4.
Exp Rheumatol 1995;13:217–26. 57. Shor A, Kuo CC, Patton DL. Detection of Chlamydia
39. Petursdottir V, Nordborg E, Moraghebi N, Persson M, pneumoniae in the coronary artery atheroma plaques. S Afr
Nordborg C. Estrogen receptors in giant cell arteritis. An Med J 1992;82:158–61.
immunocytochemical, Western blot and RT-PCR study. 58. Kuo CC, Grayston JT, Campbell LA, Goo YA, Wissler
Clin Exp Rheumatol 1999;17:671–7. RW, Benditt E. Chlamydia pneumoniae (TWAR) in coro-
40. Petursdottir V, Moslemi A-R, Persson M, Nordborg E, nary arteries of young adults (15–35 years old). Proc Natl
Nordborg C. Estrogen receptor a in giant cell arteritis: a Acad Sci USA 1995;92:6911–4.
molecular genetic study. Clin Exp Rheumatol 2001; 59. Wagner AD, Gerard HC, Fresemann T et al. Detection of
19:297–302. Chlamydia pneumoniae in giant cell vasculitis and correlation
41. Petursdottir V, Johansson H, Nordborg E, Nordborg C. with the topographic arrangement of tissue-infiltrating
The epidemiology of biopsy-positive giant cell arteritis: dendritic cells. Arthritis Rheum 2000;43:1543–51.
special reference to cyclic fluctuations. Rheumatology 60. Haugeberg G, Bie R, Nordbo SA. Chlamydia pneumoniae
1999;38:1208–12. is not detected in temporal artery biopsies from patients
42. Sonnenblick M, Nesher G, Friedlander Y, Rubinow A. with temporal arteritis. Scand J Rheumatol 2000;
Giant cell arteritis in Jerusalem: A 12-year epidemiological 29:127–8.
study. Br J Rheumatol 1994;33:938–41. 61. Wang SP, Grayston JT. Population prevalence antibody to
43. Cimmino MA, Caporali R, Montecucco CM, Rovida S, Chlamydia pneumoniae, strain TWAR. In: Bowic WR,
Baratelli E, Broggini M. A seasonal pattern in the Caldwell HD, Jones RP et al., eds. Chlamydial infections.
onset of polymyalgia rheumatica. Ann Rheum Dis 1990; Cambridge: Cambridge University Press, 1990.
49:521. 62. Grayston JT. Infections caused by Chlamydia pneumoniae,
44. Kinmont PDC, McCallum DI. The etiology and course of strain TWAR. Clin Infect Dis 1992;15:757–63.
giant cell arteritis. The possible role of light sensitivity. Br J 63. Russo MG, Waxman J, Abdoh AA, Serebro LH.
Dermatol 1965;77:193–202. Correlation between infection and the onset of giant
45. Jonasson F, Cullen JF, Elton RA. Temporal arteritis. A cell (temporal) arteritis syndrome. A trigger mechanism?
14-year epidemiological, clinical and prognostic study. Arthritis Rheum 1995;38:374–80.
Scott Med J 1979;24:111–7. 64. Barrier J, Pion P, Massari R, Peltier P, Rojouan J,
46. Elling P, Olsson AT, Elling H. Synchronous variations of Grolleau JY. Epidemiological approach to Horton’s
incidence of temporal arteritis and polymyalgia rheumatica disease in the department of Loire-Atlantique. 110 cases
in different regions of Denmark; associations with epi- in 10 years (1970–1979). Rev Med Interne 1982;3:13–20.
demics of Mycoplasma pneumoniae infection. J Rheumatol 65. Salvarani C, Macchioni PL, Tartoni P et al. Polymyalgia
1996;23:112–9. rheumatica and giant cell arteritis: A 5-year epidemiologic
47. Elling H, Olsson AT, Elling P. Human Parvovirus and and clinical study in Reggio Emilia, Italy. Clin Exp
giant cell arteritis: A selective arteritic impact? Clin Exp Rheumatol 1987;5:205–15.
Rheumatol 2000;18(Suppl. 20):12–4. 66. Friedman G, Friedman B, Benbassat J. Epidemiology of
48. Finkel TH, Torok TJ, Ferguson PJ et al. Chronic temporal arteritis in Israel. Isr J Med Sci 1982;18:241–4.
parvovirus B19 infection and systemic necrotising 67. Smith CA, Fidler WJ, Pinals RS. The epidemiology of
vasculitis: opportunistic infection or aetiological agent? giant cell arteritis. Report of a ten-year study in Shelby
Lancet 1994;343:1255–8. County, Tennessee. Arthritis Rheum 1983;26:1214–9.
49. Nikkari S, Mertsola J, Korvenranta H, Vainiopaa R, 68. Haugeberg G, Paulsen PQ, Bie RB. Temporal arteritis in
Toivanen P. Wegener’s granulomatosis and Parvovirus Vest Agder County in Southern Norway: Incidence and
B19 infection. Arthritis Rheum 1994;37:1707–8. clinical findings. J Rheumatol 2000;27:2624–7.Giant cell arteritis: pathogenesis and treatment 421
69. Hunder GG, Bloch DA, Michel BA et al. The American methylprednisolone in the initial treatment of simple
College of Rheumatology 1990 criteria for the classification forms of giant cell arteritis: a one year follow-up study of
of giant cell arteritis. Arthritis Rheum 1990;33:1122–8. 164 patients. J Rheumatol 2000;27:1484–91.
70. Östberg G. On arteritis with special reference to poly- 88. Gabriel SE, Sunku J, Salvarani C, O’Fallon WM,
myalgia arteritica. Acta Pathol Microbiol Scand 1973; Hunder GG. Adverse outcomes of anti-inflammatory
Suppl. 237:1–59. therapy among patients with polymyalgia rheumatica.
71. Cimmino MA. Genetical and environmental factors in Arthritis Rheum 1997;40:1873–8.
polymyalgia rheumatica. Ann Rheum Dis 1997;56:576–7. 89. De Silva M, Hazleman BL. Azathioprine in giant cell
72. Andersson R, Malmvall BE, Bengtsson BÅ. Long-term arteritisupolymyalgia rheumatica: A double blind study.
survival in giant cell arteritis including temporal arteritis Ann Rheum Dis 1986;45:136–8.
and polymyalgia rheumatica. A follow-up study of 90 90. Krall PA, Mazanec DJ, Wilke WS. Methotrexate for
patients treated with corticosteroids. Acta Med Scand corticosteroid resistant polymyalgia rheumatica and giant
1986;220:361–4. cell arteritis. Cleve Clin J Med 1989;56:253–7.
73. Caselli RJ, Hunder GG, Whisnant JP. Neurologic disease 91. Van der Veen MJ, Dinant HJ, Van Booma-Franckfort C,
in biopsy-proven giant cell (temporal) artertis. Neurology Van Albada-Kuipers GA, Bijlsma JWJ. Can methotrexate
1988;38:352–9. be used as a steroid sparing agent of polymyalgia
74. Säve-Söderbergh J, Malmvall BE, Andersson R, Bengtsson rheumatica and giant cell arteritis? Ann Rheum Dis
BÅ. Giant cell arteritis as a cause of death—report of nine 1996;55:218–23.
cases. J Am Med Assoc 1986;255:493–6. 92. Hernandez-Garcia C, Soriano C, Morado C et al.
75. Matteson EL, Gold KN, Bloch DA, Hunder GG. Methotrexate treatment in the management of giant cell
Long-term survival of patients with giant cell arteritis in arteritis. Scand J Rheumatol 1994;23:295–8.
the American College of Rheumatology giant cell arteritis 93. Schaufelberger C, Andersson R, Nordborg E. No additive
classification criteria cohort. Am J Med 1996;100:193–6. effect of cyclosporin A compared with glucocorticoid
76. Nordborg E, Bengtsson BÅ. Death rates and causes of treatment alone in giant cell arteritis. Results of an open,
death in 284 consecutive patients with giant cell arteritis controlled, randomized trial. Br J Rheumatol 1998;
confirmed by biopsy. Br Med J 1989;299:549–50. 37:464–5.
77. Schaufelberger C, Bengtsson BÅ, Andersson R. Epidemiology 94. Jover JA, Hernandez-Garcia C, Morado IC, Vargas E,
and mortality in 220 patients with polymyalgia rheumatica. Banares A, Fernandez-Gutierrez B. Combined treatment
Br J Rheumatol 1995;34:261–4. of giant-cell arteritis with methotrexate and prednisolone.
78. Naschitz JE, Slobodin G, Yeshurun D, Rozenbaum M, A randomized, double-blind, placebo-controlled trial.
Rosner I. A polymyalgia rheumatica-like syndrome as Ann Intern Med 2001;134:106–14.
presentation of metastatic cancer. Clin Rheumatol 1996; 95. Hoffman GS, Cid MC, Hellmann DB et al. A multicenter,
2:305–8. randomized, double-blind, placebo-controlled trial of
79. Haga HJ, Eide GE, Brun J, Johansen A, Langmark F. adjuvant methotrexate treatment for giant cell arteritis.
Cancer in association with polymyalgia rheumatica and Arthritis Rheum 2002;46:1309–18.
temporal arteritis. J Rheumatol 1993;20:1335–9. 96. Feldman M, Maini RN. Anti-TNF-a therapy of rheuma-
80. Gonzalez-Gay MA, Blanco R, Rodriguez-Valverde V et al. toid arthritis: what have we learned? Annu Rev Immunol
Permanent visual loss and cerebrovascular accidents in 2001;19:163–96.
giant cell arteritis: Predictors and response to treatment. 97. Cantini F, Niccoli L, Salvarani C, Padula A, Olivieri I.
Arthritis Rheum 1998;41:1497–504. Treatment of longstanding active giant cell arteritis with
81. Schmidt D, Vaith P, Hetzel A. Prevention of serious infliximab: report of four cases. Arthritis Rheum 2001;
ophthalmic and cerebral complications in temporal 44:2933–5.
arteritis. Clin Exp Rheumatol 2000;18(Suppl. 20):61–3. 98. Salvarani C, Cantini F, Olivieri I et al. Treatment of
82. Hunder GG, Sheps SG, Allen GL, Joice JW. Daily and longstanding active polymyalgia rheumatica with infli-
alternate-day corticosteroid regimens in treatment of giant ximab: report of four cases. Ann Rheum Dis 2002;61
cell arteritis: comparison in a prospective study. Ann Intern (Suppl. 1): 375–6.
Med 1975;82:613–8. 99. Wilke WS, Hoffman GS. Treatment of corticosteroid-
83. Bengtsson BÅ, Malmvall BE. An alternate-day cortico- resistant giant cell arteritis. Rheum Dis Clin North Am
steroid regimen in maintenance therapy of giant cell 1995;21:59–71.
arteritis. Acta Med Scand 1981;209:347–50. 100. BrackA, Rittner HL, Younge BR, Kaltschmidt C,
84. Kyle V, Hazleman BL. Treatment of polymyalgia rheu- Weyand CM, Goronzy JJ. Glucocorticoid-mediated
matica and giant cell arteritis II. Relationship between repression of cytokine gene transcription in human
steroid dose and steroid associated side effects. Ann arteritis–SCID chimeras. J Clin Invest 1997;99:2842–50.
Rheum Dis 1989;48:662–6. 101. Pearce G, Ryan PFJ, Delmas PD et al. The deleterious
85. Ruegsegger P, Medici TC, Anliker M. Corticosteroid- effects of low-dose corticosteroids on bone density in
induced bone loss. A longitudinal study of alternate-day patients with polymyalgia rheumatica. Br J Rheumatol
therapy in patients with bronchial asthma using quantita- 1998;37:292–9.
tive computed tomography. Eur J Clin Pharmacol 1983; 102. Haugeberg G, Myklebust G, Dovland H, Mikkelsen B,
25:615–20. Gran JT. No permanent reduction in bone mineral
86. Dasgupta B, Dolan AL, Panayi GS, Fernandes L. density during treatment of polymyalgia rheumatica and
An initially double-blinded controlled 96 week trial of temporal arteritis using low doses of corticosteroids.
depot methylprednisolone against oral prednisolone in Scand J Rheumatol 2000;29:163.
the treatment of PMR. Br J Rheumatol 1998;37:189–95. 103. Dolan AL, Moniz C, Dasgupta B et al. Effects of inflam-
87. Chevalet P, Barrier JH, Pottier P, Magadur-Joly G, mation and treatment on bone turnover and bone mass
Pottier MA, Hamidou M. A randomized, multicenter, in polymyalgia rheumatica. Arthritis Rheum 1997;
controlled trial using intravenous pulses of 40:2022–9.You can also read
