Internal Medicine Journal 2002; 32: 315–319
Audit of the management of suspected giant cell arteritis in a large teaching hospital
N. DALBETH,1 N. LYNCH,1 L. McLEAN,2 F. McQUEEN1,2 and J. ZWI1
1Auckland Healthcare and 2Department of Molecular Medicine, University of Auckland, Auckland, New ZealandAbstract Results: The mean waiting time for biopsy for all
Background: The diagnosis of giant cell arteritis
patients was 5.6 days (range 0–42 days). This time
(GCA) is often confirmed by an early temporal artery
varied from 9.3 days for rheumatology patients to
(TA) biopsy of adequate length. Treatment of this
days for ophthalmology patients (P = 0.003).
condition with high-dose corticosteroids may be asso-
Only 44/117 (37.6%) specimens measured more than
ciated with significant morbidity, including osteo-
10 mm. For GCA patients, the median initial oral
prednisone dose was 60 mg/day. Osteoporosis proph-ylaxis was prescribed in 24/37 (65%) GCA patients,
Aim: To audit current management of patients with
suspected GCA at Auckland Healthcare, a largeteaching hospital. Conclusions: There is significant variation in the man-agement of GCA within our institution. This audit
Methods: We performed a retrospective chart review
has highlighted several areas where improvement
of all TA biopsies from January 1996 to June 2000. A
could be made, particularly in streamlining the
total of 117 biopsies from 111 patients was audited.
process of obtaining TA biopsy and in promoting the
Of these patients, 37/111 (33%) had a final clinical
use of osteoporosis prophylaxis. (Intern Med J 2002;
diagnosis of GCA (GCA patients). The areas of inter-
est for audit were waiting time for TA biopsy, lengthof sample, initial corticosteroid therapy and osteo-
Key words: audit, corticosteroids, giant cell arteritis,
osteoporosis, temporal artery biopsy.
temporal arteries. Although clinical findings areintegral to the diagnosis of GCA, temporal artery
Giant cell arteritis (GCA) is a common form of vas-
(TA) biopsy remains an important confirmatory test.2
culitis affecting elderly patients. Clinical features
Typical histological findings of active GCA include
include arteritic symptoms such as headache, scalp
transmural inflammation, disruption of the internal
tenderness, jaw claudication and visual disturbance,
elastic lamina, intimal fibrosis and oedema, and giant
and systemic symptoms including malaise, weight loss
cells.3 The risk of false-negative biopsy owing to ‘skip
and fatigue. Symptoms of polymyalgia rheumatica
lesions’ is well recognized.4 Management with high-
(PMR) may occur concurrently.1 Clinical examina-
dose corticosteroids reduces complications of disease,
tion may demonstrate tender, thickened or pulseless
particularly loss of vision.5 However, side-effectsrelated to therapy can be significant.6 Although thereis controversy about some aspects of GCA manage-ment, good clinical practice includes a waiting time
Correspondence: Nicola Dalbeth, MRC Human Immunology Unit, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital,
for biopsy of less than 2 weeks, specimen length of
Oxford OX3 9DU, UK. Email: nicola.dalbeth@imm.ox.ac.uk
more than 10 mm and osteoporosis prophylaxis for all
Received 14 May 2001; accepted 9 November 2001.
This retrospective study audited the management of
five had findings consistent with healed GCA, and
suspected GCA at a large teaching hospital facility. In
seven had ‘age-related changes’ or normal biopsies.
particular, we assessed the waiting times for biopsy,
The seven patients with negative biopsies had other-
length of biopsy specimen, initial corticosteroid dose
wise typical clinical features of GCA.
and use of routine osteoporosis prophylaxis.
Of the 11 patients with TA biopsies reported as
consistent with healed GCA, only five patients wereconsidered by the clinical team as having a final diag-
We audited all TA biopsies taken at Auckland
nosis of GCA. The remaining six patients were
Healthcare (Auckland Hospital and Greenlane
treated for conditions other than GCA (including
Hospital) over a 4.5-year period between January
polymyalgia rheumatica alone, non-specific poly-
1996 and June 2000. This institution is a 1036-bed
arthritis, atherosclerotic stroke, cluster headache,
teaching hospital associated with the University of
migraine and chronic pain syndrome).
Auckland, and has a population catchment of386 000 people. The biopsies were identified from a
computerized database in the Department ofPathology, and confirmed by cross-referencing data-
Patients with a final diagnosis of GCA had a mean age
bases in the Departments of Rheumatology and
of 73.2 years (range 54–89 years). The majority of
General Surgery. A retrospective patient chart review
these patients were female (70%) and 95% were of
was performed for each procedure. Results were
European descent (the remaining 5% was of Samoan
analysed using Student t-tests and contingency tables
descent). There were only 7/37 (19%) patients who
on Prism 3.0 (Graph Pad, San Diego, CA, USA).
had symptoms for less than 1 week before presenta-tion. The most common symptoms were headache
During this time, 122 biopsies were taken from 116
(89%) and visual disturbance (64%). Associated
patients (mean 27 biopsies/year). Bilateral biopsies
PMR symptoms (either at the time of diagnosis or
were performed in six patients (three sequential and
previously) were reported in 63%, and symptoms of
three simultaneous). Medical records were not avail-
jaw claudication occurred in 43%. The mean erythro-
able for five patients, so that a total of 117 biopsies
cyte sedimentation rate (ESR) for this group was
(114 episodes) in 111 patients was audited.
77.3 mm/h (range 19–133 mm/h). There were 3/37(8.1%) GCA patients with ESR less than 30 mm/h.
A number of different specialties requested TA biop-
Information regarding waiting time was available for
sies, most commonly Rheumatology and Ophthal-
104/114 procedures. There were 11/104 (10.6%)
mology. There were 82/114 (71.9%) procedures done
patients who waited for 2 weeks or more for biopsy.
by the General Surgery service, 30/114 (26.3%) by
The mean waiting time for TA biopsy was 5.6 days.
the Ophthalmology service, and 2/114 (1.8%) by the
Rheumatology patients had a mean waiting time of
Neurosurgery service. Most of the biopsies requested
9.3 days (range 0–42 days). This compares with a
by Rheumatology and other medical services were
mean waiting time of 2.6 days (range 0–9 days) for
taken by the General Surgical service, while all
biopsies requested by Ophthalmology were taken by
0–23 days) for General Medical patients (P = 0.003
for Ophthalmology vs Rheumatology). (Fig. 1)
Of the TA biopsies taken, histopathologists reported
Waiting times also varied depending on the surgical
25/117 (21.4%) as being consistent with active GCA,
service performing the TA biopsy, with a mean time
11/117 (9.4%) as being consistent with healed GCA,
of 2.9 days (range 0–10 days) for Ophthalmology
and 81/117 (69.2%) as normal or ‘age-related
changes’ (including arteriosclerosis).
There were 37/111 (33%) patients termed ‘GCA
Mean waiting times for biopsy varied from 2.8 days
patients’ for the purposes of this analysis. This group
for those with positive biopsy results, 6.4 days for
was defined as having a clinical diagnosis of GCA and
those with negative biopsy results, and 12.36 days for
was treated long term for this condition. Of these 37
those with healed GCA on biopsy (P = 0.03, positive
patients, 25 had active GCA reported on TA biopsy,
Internal Medicine Journal 2002; 32: 315–319
Management of giant cell arteritis
Waiting time for temporal artery biopsy for
referring specialties. P = 0.003, Ophthalmology vs
Osteoporosis prophylaxis use depending on
treating specialty. Other physicians include General Physicians, Neurologists and Geriatricians. P = 0.023, Ophthalmology vs Rheumatology. GCA, giant cell arteritis.
We assessed the length of TA biopsy measured by the
Oral prednisone doses were not significantly higher in
histopathologist after formalin fixation (which may be
GCA patients with visual symptoms compared with
associated with some shrinkage). Very few biopsy
measurements were recorded in the clinical recordsby the surgical team.
Follow-up information regarding complications ofcorticosteroids was available for 32/37 GCA patients.
Pathology measurements showed that 44/117
There were 10/32 patients (31.2%) who had at least
(37.6%) specimens measured more than 10 mm. The
one complication that could be attributed to high-
median biopsy length was 10 mm. There were 15/117
dose corticosteroids (including newly diagnosed
(12.8%) biopsies that measured 5 mm or less, and
diabetes mellitus, or worsening diabetic control,
only 6/117 (5.1%) biopsies measured more than
osteoporotic fractures, proximal myopathy and
20 mm. There was no difference in length between
positive and negative TA biopsy samples.
For GCA patients, the use of osteoporosis prophy-
Information regarding initial prednisone dosage was
laxis (other than calcium supplementation) was also
available for 36/37 GCA patients. For these patients,
variable, with only 24/37 (65%) receiving osteo-
the median initial prednisone dose was 60 mg/day
porosis prophylaxis. For patients treated with
(range 10–80 mg/day). Only 10/36 (27.8%) patients
osteoporosis prophylaxis, 20/24 (83.3%) received
were treated with less than 60 mg/day prednisone.
Pulsed intravenous methylprednisolone was used in4/36 (11.1%) GCA patients (two patients treated by
The use of osteoporosis prophylaxis other than
the Neurology service, two patients treated by the
calcium supplementation in GCA patients varied
Ophthalmology service, all with new onset of visual
depending on treating specialty, with 12/15 (80%)
symptoms). One patient was prescribed steroid-
Rheumatology-treated patients and 2/8 (25%)
sparing medication (methotrexate) during the follow-
Ophthalmology-treated patients receiving such
therapy (P = 0.023) (Fig. 2).
Initial prednisone doses varied depending on the
treating specialty. The mean initial dose for Rheum-atology GCA patients was 50
There is significant variation in some aspects of the
initial management of suspected GCA within our
large teaching hospital. This study has raised partic-
ular issues of interest, including excessive waiting
Internal Medicine Journal 2002; 32: 315–319
times for some biopsies, short specimen length and
Patients with GCA are at high risk of corticosteroid-
inadequate osteoporosis prophylaxis.
induced osteoporosis, due to advanced age and alsoto high prednisone doses administered for long
The importance of an early TA biopsy of adequate
periods. It has been demonstrated that significant
length is generally accepted, although the minimum
bone loss occurs in the first year of corticosteroid
acceptable length remains controversial. The
use12 and that prophylactic bisphosphonates13,14 can
presence of the patchy distribution of pathological
prevent treatment-related osteoporotic fractures.
lesions has led to a recognition that longer specimens
Our study reinforces the need for an education
should be taken.4 It has been suggested that it is best
programme to promote greater awareness of early
to sample an area where clinical disease is suspected,
but, failing this, at least 1 cm and preferably 2 cm ofartery should be taken.7 Almost two-thirds of the
In summary, this audit has demonstrated shortfalls in
biopsies taken from the patients in our study
ideal clinical care and the need for a more coordi-
measured 1 cm or less, suggesting that we may be
nated approach to the management of suspected
GCA at Auckland Healthcare. Such variations in themanagement of GCA may well occur in other large
The upper limit of acceptable waiting time is also
teaching hospitals. Development of evidence-based
disputed. An early study8 showed that positive biopsy
clinical guidelines and closer liaison between
rates fell from 82 to 20% after 1 week of therapy. In
subspecialty services are likely to improve the care of
this study, the mean biopsy length was short, with a
mean of 7 mm (similar to our mean sample length). More recently, a study from the Mayo clinic9 showed
no difference in positive biopsy rates after 2 weeks ofcorticosteroid therapy. However, the mean biopsy
The authors would like to thank Professor Charles
length in this study was much greater at 36 mm (with
McGhee, Department of Ophthalmology, University
one biopsy measuring 12.5 cm) and, arguably, this
of Auckland, and Dr Peter Gow, Rheumatologist,
may have offset the longer waiting time. Our study
Middlemore Hospital, Auckland for their useful
did not show that specimen length had a significant
effect on biopsy results, although the sample size maybe insufficient to detect such a difference. The findingof shorter waiting times in those with positive biop-sies is interesting and may suggest the importance
of obtaining early biopsies to avoid false-negativeresults. However, a possible confounding factor may
1 Hazleman B. Polymyalgia rheumatica and giant cell arteritis.
have greater perceived urgency for those patients with
In: Klippel JH, Dieppe PA (eds). Rheumatology. London:
more classical features of disease. The variations in
2 Hall S, Lie JT, Kurland LT et al. The therapeutic impact of
waiting time between services at our institution may
temporal artery biopsy. Lancet 1983; 2: 1217–20.
in part be explained by the prompt action of the
3 Ashton-Key M, Gallagher PJ. Surgical pathology of cranial
Ophthalmology service in taking biopsies on their
arteritis and polymyalgia rheumatica. Baillieres Clin
4 Klein RG, Campbell RJ, Hunder GG, Carney JA. Skip lesions
in temporal arteritis. Mayo Clin Proc 1976; 51: 504–10.
Generally, higher doses of corticosteroids were used
5 Kyle V. Treatment of polymyalgia rheumatica/giant cell
as initial therapy in our group of GCA patients, with
arteritis. Baillieres Clin Rheumatol 1991; 5: 485–92.
considerable variation in dosage. A dose of 40 mg/day
6 Kyle V, Hazleman BL. Treatment of polymyalgia rheumatica
and giant cell arteritis. II. Relation between steroid dose and
quate.10 Higher corticosteroid doses may be indicated
steroid associated side effects. Ann Rheum Dis 1989; 48: 662–6.
for patients with visual symptoms. Prescribing differ-
7 Anon. Temporal artery biopsy [editorial]. Lancet 1983; 1: 396.
ences between ophthalmologists and rheumatologists
8 Allison MC, Gallagher PJ. Temporal artery biopsy and
are well recognized.11 The high rate of corticosteroid-
corticosteroid treatment. Ann Rheum Dis 1984; 43: 416–17.
associated side-effects in our patients is similar to
9 Achbar AA, Lie JT, Hunder GG et al. How does previous
previous reported rates for GCA and PMR patients.6
corticosteroid treatment affect the biopsy findings in giant cell (temporal) arteritis? Ann Intern Med 1994; 120: 987–92.
A retrospective chart review such as our study may
10 Kyle V, Hazleman BL. Treatment of polymyalgia rheumatica
underestimate the true incidence of therapy-related
and giant cell arteritis. I. Steroid regimens in the first two
months. Ann Rheum Dis 1989; 48: 658–61. Internal Medicine Journal 2002; 32: 315–319
Management of giant cell arteritis
11 Swannell AJ. Polymyalgia rheumatica and temporal
13 Adachi JD, Benson WG, Brown J et al. Intermittent etidronate
arteritis: diagnosis and management. Br Med J 1997; 314:
therapy to prevent corticosteroid-induced osteoporosis. N Engl
12 Reid IR, Heap SW. Determinants of vertebral mineral density
14 Saag KG, Emkey R, Schnitzer TJ et al. Alendronate for the
in patients receiving long-term glucocorticoid therapy. Arch
prevention and treatment of glucocorticoid-induced
osteoporosis. N Engl J Med 1998; 339: 292–9. Internal Medicine Journal 2002; 32: 315–319
NATALIA MARTÍN CRUZ JUAN HERNANGÓMEZ BARAHONA JUAN MANUEL DE LA FUENTE SABATÉ El ámbito institucional en el que se circunscribe la regulación delos medicamentos en España y los cambios que en él se han producidohan influido necesariamente en el proceso de autorización de nuevosmentos genéricos y precios de referenciaen 1996. Los resultados de estas actuacio-rios creados para
Condições Gerais de Venda (RIBALNET, LDA) Encomendas Todas as encomendas deverão ser feitas através do departamento comercial, via fax ou e-mail. A RIBALNET pode dispensar o cliente deste procedimento e aceitar a encomenda via telefone. Deverá igualmente mencionar se pretende levantar a mercadoria constante na encomenda, caso contrario, a RIBALNET expedi-la-á por meios