Influence of Previous Corticosteroid Therapy on Temporal Artery Biopsy Yield in Giant Cell Arteritis

Gta

*I

I

B

d

RH

0d

Influence of Previous Corticosteroid Therapyon Temporal Artery Biopsy Yield in Giant Cell Arteritis

Javier Narváez, MD, PhD,* Berta Bernad, MD,†

Daniel Roig-Vilaseca, MD,* Carmen García-Gómez, MD,*Carmen Gómez-Vaquero, MD, PhD,* Xavier Juanola, MD, PhD,*Jesús Rodriguez-Moreno, MD,* Joan M. Nolla, MD, PhD,‡ and

José Valverde, MD, PhD§

Objective: To determine the impact of prior corticosteroid treatment on temporal artery biopsy(TAB) yield to establish the diagnosis of giant cell arteritis (GCA).Methods: Retrospective study of a consecutive cohort of 78 patients clinically diagnosed andmanaged as GCA, who received corticosteroids before TAB.Results: Among the 78 patients, TAB was positive in 57 (73%) and negative in 21 (27%). Nosignificant differences in the length of the specimen were found between the positive and negativebiopsies.

We grouped patients according to treatment duration before TAB. In those with newly diag-nosed GCA treated with high-dose steroid therapy, the biopsy results were positive in 78% (35/45)of patients treated for less than 2 weeks, in 65% of those treated for 2 to 4 weeks (13/20), and in40% of those treated for more than 4 weeks (2/5). We also observed 8 patients that developedGCA on a background of a prior history of polymyalgia rheumatica (PMR); in this group biopsywas positive in 88% of the cases, after a median duration of treatment of 180 � 172 days and anaverage daily dose of 7.1 � 1.4 mg/d.Conclusion: The performance of TAB should not delay the prompt institution of steroid therapyon diagnosis of GCA, since the diagnostic yield of TAB seems valuable within 4 weeks of startinghigh-dose steroid treatment. In patients that developed GCA on a background of a prior history ofPMR, a late TAB is also generally informative despite long-term treatment with low doses ofcorticosteroids.© 2007 Elsevier Inc. All rights reserved. Semin Arthritis Rheum 37:13-19Keywords: giant cell (temporal) arteritis, temporal artery biopsy yield, corticosteroid treatment

itabdcI1ttf

nbm

iant cell arteritis (GCA) is a common vasculitis ofunknown origin that usually involves large- andmedium-sized vessels, with a predisposition for

he cranial arteries. It almost exclusively affects individu-ls over the age of 50. Prompt diagnosis and treatment are

Staff Physician, Department of Rheumatology, Hospital Universitario de Bellvitge–DIBELL, Barcelona, Spain.

†Resident, Department of Rheumatology, Hospital Universitario de Bellvitge–DIBELL, Barcelona, Spain.

‡Chief of Section, Department of Rheumatology, Hospital Universitario deellvitge–IDIBELL, Barcelona, Spain.§Head of Department, Department of Rheumatology, Hospital Universitario

e Bellvitge–IDIBELL, Barcelona, Spain.Address reprint requests to: Dr. Francisco Javier Narváez Garcia, Department of

theumatology (Planta 10-2), Hospital Universitario de Bellvitge, Feixa Llarga s/n,ospitalet de Llobregat, Barcelona 08907, Spain. E-mail: [email protected].

049-0172/07/$-see front matter © 2007 Elsevier Inc. All rights reserved.oi:10.1016/j.semarthrit.2006.12.005

mportant to prevent serious vascular ischemic complica-ions, particularly blindness. Considering that clinicalnd laboratory markers lack specificity, temporal arteryiopsy (TAB) is considered the gold standard test for theiagnosis of GCA (1-3). TAB is a minor procedure thatan yield important results for the management of GCA.t has a sensitivity of 24 to 90% and a specificity of 81 to00% (4-16). A positive biopsy also helps to prevent fu-ure doubts regarding the accuracy of the diagnosis, par-icularly when the patient suffers corticosteroid side ef-ects or fails to respond promptly to therapy.

However, a negative TAB does not rule out the diag-osis of GCA; the likelihood of a false-negative TAB maye influenced to some extent by the length of the speci-en, the presence of skip lesions, pathological sectioning

echniques, and the duration of steroid therapy before

13

btaC(t

crmTpmhp(ttprp(Cstacfid

sTap

P

WnwSmp

ncot(htu(aatc

pfp

wphtct

chtdd(dtd(iiommmwf

S

CtcSMwttni

R

S

TpAcpmtbte

14 Influence of previous corticosteroid therapy

iopsy (7,16-26). As a result, TAB is not always affirma-ive, and in some series of patients clinically diagnosednd managed as GCA according to the 1990 Americanollege of Rheumatology (ACR) classification criteria

11), negative biopsies have been reported in 42 to 61% ofhe patients (4,6,12,14,15,27).

Controversy exists about the extent of influence priororticosteroid treatment has on TAB findings. One studyeported that histological evidence of arteritis is usuallyasked after 1 week of corticosteroid treatment (21).hese investigators questioned the value of the TAB inatients who received high doses of corticosteroids forore than 1 week before biopsy. In contrast, other studies

ave demonstrated that histological evidence of arteritisersists despite 2 weeks of corticosteroid treatment12,23). Similarly, Liozon and coworkers (22) looked athe relationship existing between previous corticosteroidreatment and the results of TAB, including both classicalathological results and immunohistochemical stainingesults, and observed that the immunological process isoorly controlled by a short course of high-dose steroidslasting from 1 day to 1 month). More recently, Ray-haudhuri and colleagues (24) found, in a prospective

tudy performed in 11 patients, that diagnostic his-opathological findings can be demonstrated even 4 weeksfter the initiation of steroids. Although this data are en-ouraging, due to the small sample size, additional con-rmation is needed before definitive conclusions can berawn.To further ascertain the impact of previous high-dose

teroid treatment on the likelihood of obtaining a positiveAB in patients with GCA, we have reviewed the clinical

nd histological findings in a consecutive cohort of 78atients who received corticosteroids before TAB.

ATIENTS AND METHODS

e conducted a retrospective study of all patients diag-osed with GCA by the Department of Rheumatologyithin the Hospital Universitario de Bellvitge (Barcelona,pain) from January 1986 until December 2004. We pri-arily determined, from medical records, whether each

atient had received steroid treatment before TAB.During this time, a total of 121 patients were diag-

osed with GCA; 78 patients (64%) who received corti-osteroids before biopsy were eligible to be included inur study. The diagnosis of GCA was made according tohe 1990 ACR criteria for the classification of GCA (11):1) age at disease onset �50 years; (2) sudden onset ofeadache; (3) temporal artery abnormality on examina-ion (decreased pulses unrelated to arteriosclerosis, nod-les, thickening, swelling, or tenderness to palpation),4) erythrocyte sedimentation rate (ESR) � 50 mm/h;nd (5) a TAB specimen showing vasculitis mainly char-cterized by mononuclear cell infiltration or granuloma-ous inflammation, usually with multinucleated giant
ells. Patients were diagnosed with GCA if they had a m
ositive TAB or, in cases with negative biopsy, if theyulfilled the remaining 4 criteria and had a prompt andersistent response to corticosteroid treatment.Biopsy of the more clinically suspicious temporal artery

as performed in all patients. Gross biopsy sectioning andathologic interpretation were consistently performed in ourospital according to a standard protocol. The pathologistshat examined the biopsy specimens were blinded to thelinical data and previous corticosteroid treatment informa-ion.

The dose and duration of corticosteroid treatment re-eived before TAB and detailed clinical, laboratory, andistopathological data at the time of diagnosis were ex-racted from clinical records according to a specificallyesigned protocol. We calculated the approximate totalose and the average daily dose of prednisone receivedaverage daily dose � total dose of prednisone/number ofays of treatment). Biopsy results were classified as nega-ive (no evidence of arteritis) or positive (histologic evi-ence of arteritis). In agreement with previous studies24,25), we further sorted positive biopsy results accord-ng to active or healed arteritis. Both are characterized bynterruption of the internal elastic laminae, with or with-ut intimal thickening, and thickening and scarring of theedia and adventitia. Active arteritis displays markedononuclear cell infiltration or granulomatous inflam-ation, with or without multinucleated giant cells,hereas healed arteritis is only associated with occasional

oci of lymphocytes, with or without intimal fibrosis.

tatistical Analysis

ontinuous data were described as mean � standard devia-ion (SD) and categorical variables were presented as per-entages. Comparisons between groups were made using thetudent’s t-test for independent continuous variables or theann-Whitney U-test when the assumption of normality

as not achieved. To analyze categorical data, we performedhe �2 test. Correlation between duration of corticosteroidherapy before biopsy and TAB yield in terms of positivity oregativity was tested by the Spearman test. Statistical signif-

cance was defined as P � 0.05.

ESULTS

tudy Sample and Overall Biopsy Results

he main clinical features and laboratory data of the 78atients included in the study are summarized in Table 1.ll patients were referred to the hospital from the primaryare unit or they self-referred to the emergency room. Therinciple motives for the initiation of corticosteroid treat-ent before TAB were the presence of visual manifesta-

ions at diagnosis (18 cases), delay in performance of theiopsy due to surgical waiting lists (27 cases), and institu-ion of steroid treatment by general practitioners or in themergency unit with subsequent referral to our depart-
ent (33 cases, including 8 patients initially diagnosed

wd

2(tbuss(sqtnt

nwTss

RC

Tt

btiiftta

rrtt

wnaadmoth

ptdrGrcdpto�ssttst

o(1wmPalwtdv

J. Narváez et al. 15

ith isolated polymyalgia rheumatica (PMR) that latereveloped symptoms of GCA).TAB was positive in 57 (73%) patients and negative in

1 (27%). Bilateral biopsy specimens were taken in 79%) patients with a negative unilateral biopsy: in 6 caseshe second biopsy was positive and negative in 1. Repeatiopsy in cases with a negative unilateral biopsy was notsually performed as the patients’ symptoms rapidly re-olved with treatment. Mean biopsy specimen length wasimilar in patients with positive and negative results1.61 � 0.69 versus 1.55 � 0.43 cm; P � 0.28); noignificant differences in the number of slides or the fre-uency of cross-sections/cm of artery were found betweenhe positive and negative biopsies. Furthermore, we didot find significant differences in disease expression be-ween biopsy-proven and biopsy-negative cases (Table 2).

During the same period of time, 43 patients who didot receive corticosteroid before biopsy were diagnosedith GCA; of these, TAB was positive in 31 (72%) cases.he difference in the rate of positivity between cortico-

teroid-treated and untreated patients was not statisticallyignificant (P � 0.9).

elation between Previousorticosteroid Treatment and Biopsy Results

able 3 outlines the relationship between corticosteroid

Table 1 Main Clinical and Laboratory Data of the StudyCohort

Number of patients 78Age (mean y � SD) (range) 74.6 � 7.8 (56–89)Women/men (ratio) 50/28 (1.7)Positive TAB 57 (73%)Clinical features

Headache 76 (97%)Abnormal temporal artery 48 (61%)Jaw claudication 32 (41%)Malaise/anorexia/weight loss 46 (59%)Fever 7 (9%)Polymyalgia rheumatica 38 (49%)Visual manifestations 18 (23%)Diplopia 2 (2%)Transient visual loss 12 (15%)Permanent blindness 4 (5%)Cerebrovascular accidents 3 (4%)Limb claudication 4 (5%)

Laboratory dataESR (mm/h) 94.4 � 26.5CRP (mg/L; Ref. value �5) 51.6 � 30.1Hb (g/dl) 11.4 � 1.2Platelets (�103 cells/mm3) 340 � 135Raised ALT/AST* 8 (10%)Raised alkaline phosphatase* 17 (22%)

Results are presented as mean � standard deviation (SD) ornumber of cases with frequencies.

*Increased ALT/AST and alkaline phosphatase were consideredif values at diagnosis were �1.5 times normal.

reatment before TAB and biopsy results. l

We grouped patients according to treatment durationefore biopsy. In 45 patients the treatment period was upo 2 weeks (1 to 14 days) and biopsy results were positiven 78% (35/45) of the cases. Five of these patients receivedntravenous methylprednisolone pulse therapy (1 g dailyor 3 days) followed by 60 mg/d of prednisone (in all ofhem TAB was positive); in the remaining 40 patients theotal dose of prednisone received was 349 � 196 mg, withn average daily dose of 45.5 � 16.9 mg/d.

In another 20 patients the duration of treatmentanged between 2 and 4 weeks (15 to 28 days) and biopsyesults were positive in 65% (13/20). In this group theotal dose of prednisone received was 881 � 352 mg andhe average daily dose was 40.4 � 13.4 mg/d.

In 13 patients the duration of corticosteroid therapyas 4 weeks or more. Of these, 5 had been newly diag-osed with GCA and treated with high-dose steroid ther-py. Eight patients were originally diagnosed with PMRnd treated with low doses of corticosteroids, but latereveloped an arteritic relapse during the course of treat-ent. Due to the great differences in the employed doses

f corticosteroids, both groups were analyzed separatelyo determine the question of whether treatment withigh-dose steroids reduces the yield of TAB in GCA.Among patients with newly diagnosed GCA, TAB was

ositive in only 40% (2/5) of the cases, after receiving aotal dose of prednisone of 1673 � 1089 mg (averageaily dose, 43.8 � 23.4 mg/d). Taking into account theates of positivity in the 3 groups of patients with newCA treated with high-dose steroid therapy (n � 70), the

aw numbers might suggest a trend that, with increasedorticosteroid exposure, the likelihood of confirming theiagnosis by TAB diminishes. In this sense, when we ex-lored the correlation between duration of corticosteroidherapy before biopsy and TAB yield in terms of positivityr negativity, we found a negative correlation (r �0.169; P � 0.08), although the data failed to reach

tatistical significance probably due to the small sampleize. Similarly, the difference in the rate of positivity be-ween patients treated for less than 2 weeks and thosereated over 4 weeks or more also approached statisticalignificance (P � 0.06) despite the small number of pa-ients included in the last group.

In the 8 patients that developed GCA on a backgroundf a prior history of PMR, biopsy was positive in 88%7/8) of the cases; median duration of treatment was80 � 172 days and the total dose of prednisone receivedas 1702.7 � 1260 mg (average daily dose, 7.1 � 1.4g/d). These 8 patients were diagnosed as having isolatedMR because they did not have clinical evidence of GCAt the time of diagnosis and showed a prompt response toow-dose steroid therapy (10 to 20 mg/d of prednisone),ith normalization of the acute-phase reactants. Despite

his observation, they later experienced an arteritic relapseuring the course of steroid treatment. None sufferedisual loss or other major ischemic complications. Re-
apses usually occurred when prednisone was tapered or as

ame

wt(

maaopwtmdaphadw

D

Pbso

Tcptihu

tGmdctTptpfpsptbInsn

values


consequence of unauthorized discontinuation of treat-ent (1 case). There were no predictive features for this

volution.In all groups the total dose of prednisone received

as greater in patients with negative biopsies, althoughhese differences did not achieve statistical significanceTable 3).

All positive biopsies in patients who had received treat-ent for up to 4 weeks displayed characteristics of active

rteritis (Table 3). Of the 9 patients with a positive biopsyfter more than 4 weeks of corticosteroids, 7 showed featuresf active arteritis and 2 had features of healed arteritis. Bothatients with healed arteritis were cases originally diagnosedith isolated PMR that later experienced a relapse of arthri-

is, one 36 weeks after the initiation of corticosteroid treat-ent and the other after 34 weeks. In patients with newly

iagnosed GCA, the longest treatment period before biopsymong those with active arteritis was 45 days. In the group ofatients that developed GCA on a background of a prioristory of PMR, the longest corticosteroid-to-biopsy intervalmong those with active arteritis was 210 days (mean, 153ays), compared with an average of 246 days in the 2 patientsith healed arteritis.

ISCUSSION

atients with untreated GCA are more susceptible tolindness (which is generally irreversible unless aggres-ively treated within 24 to 48 hours of onset), strokes, and

Table 2 Comparison of Patients with and without a Positiv

Biopsy-pr(n � 5

Age (mean y � SD) 76.4 � 6Women/men (ratio) 36/21 (1Clinical features

Headache 55 (9Abnormal temporal artery 38 (6Jaw claudication 25 (4Malaise/anorexia/weight loss 36 (6Fever 6 (1Polymyalgia rheumatica 28 (4Visual manifestations 13 (2Diplopia 2 (3Transient visual loss 8 (1Permanent blindness 3 (5Cerebrovascular accidents 2 (3Limb claudication 2 (3

Laboratory dataESR (mm/h) 96 � 2CRP (mg/L; Ref. value �5) 59.1 � 2Hb (g/dl) 11.3 � 1Platelets (�103 cells/mm3) 341 � 1Raised ALT/AST* 5 (9Raised alkaline phosphatase* 12 (2

Results are presented as mean � standard deviation (SD) or numb*Increased ALT/AST and alkaline phosphatase was considered if

ther manifestations of cranial ischemic complications. f

his has led to a general consensus among clinicians thatorticosteroid treatment is necessary in subjects with sus-ected GCA. Most authorities recommend starting theherapy as soon as the diagnosis is suspected; as a result,nitiation of corticosteroids on clinical suspicion of GCAas become common practice, generally before the sched-ling of TAB.Controversy exists regarding whether prior steroid

reatment masks the characteristic pathological signs ofCA. Many clinicians believe that corticosteroid treat-ent for suspected GCA before TAB renders the proce-

ure worthless. Allison and Gallagher (21) retrospectivelyompared the percentage of positive biopsies in 132 pa-ients clinically diagnosed and managed as GCA cases.hey found a positive biopsy result in 82% of untreatedatients, in 60% of patients who received corticosteroidherapy for 1 week or less before TAB, and in 10% ofatients who received corticosteroids for over 1 week be-ore biopsy. They suggested that an increase in the pro-ortion of false-negative biopsies occurs within days ofteroid treatment and questioned the value of biopsy inatients who receive high doses of corticosteroid for morehan 1 week before biopsy. This opinion was later de-unked by an article from the Mayo Clinic in 1994 (23).n this study, Achkar and colleagues confirmed that diag-ostic histopathological findings could be readily demon-trated even 2 weeks after the commencement of pred-isone. Although they did not prove that histological

poral Artery Biopsy

Biopsy-negative(n � 21) P

70 � 8.4 0.00114/7 (2) 0.77

21 (100%) 0.4810 (48%) 0.067 (33%) 0.30

10 (48%) 0.121 (5%) 0.66

10 (48%) 0.905 (24%) 0.960 (0%) 0.364 (19%) 0.731 (5%) 0.881 (5%) 0.842 (10%) 0.31

90 � 25 0.3848.6 � 30 0.5311.8 � 1.2 0.10339 � 148 0.95

3 (14%) 0.685 (24%) 0.90

ases with frequencies.at diagnosis were �1.5 times normal.

e Tem

oven7)

.9.7)

6%)7%)4%)3%)0%)9%)3%)%)4%)%)%)%)

66.4.232%)1%)

er of c

eatures are unaffected by corticosteroids, they showed

tssimardIt

rptwrcpom


hat the TAB positivity rate in patients who had receivedteroids at any dose for any period up to 2 weeks wasimilar to that of untreated patients. Since then, somendividual case reports of positive biopsy results up to 6

onths after steroid treatment have appeared in the liter-ture, supporting the idea that vascular lesions are ratheresistant to therapy (28-30). The most recent study ad-ressing this issue seems to confirm this impression (24).n that article, Ray-Chaudhuri and coworkers prospec-

Table 3 Corticosteroid Treatment before Temporal Artery B

1 to 14 daysNumber of patients: 45Positive TAB: 35 (78%)Active arteritis/Healed arteritis: 35/0Total dose of prednisone (excluding 5 patients who receAverage daily dose: 45.5 � 16.9 mg/day

Total dose of prednisone:Average daily dose:

15 to 28 daysNumber of patients: 20Positive TAB: 13 (65%)Active arteritis/Healed arteritis: 13/0Total dose of prednisone: 881 � 352 mgAverage daily dose: 40.4 � 13.4 mg/day


�28 daysPatients with suspicion of GCA at presentation.

Number of patients: 5Positive TAB: 2 (40%)Active arteritis/Healed arteritis: 2/0Total dose of prednisone: 1673 � 1089 mgAverage daily dose: 43.8 � 23.4 mg/day


Patients originally diagnosed with isolated PMRthat later developed relapse of arteritis.

Number of patients: 8Positive TAB: 7 (87.5)Active arteritis/Healed arteritis: 5/2Median duration of treatment: 180 � 172 daysTotal dose of prednisone: 1702.7 � 1260 mgAverage daily dose: 7.1 � 1.4 mg/day

Median duration of treatment:Total dose of prednisone:Average daily dose:

ively examined the effect of up to 6 weeks of corticoste-

oid therapy on TAB diagnostic yield. They included 11atients meeting the 1990 ACR criteria for the classifica-ion of GCA, and patients underwent TAB within 1eek, at 2 to 3 weeks, or after 4 weeks of a standardized

egimen of high-dose corticosteroid treatment. At studyompletion, they found that 9 of 11 (82%) patients hadositive biopsies: 3/4 (75%) were positive within 2 weeksf starting steroid treatment and 6/7 (86%) after 4 orore weeks of treatment.

and Biopsy Results

ntravenous pulses of methylprednisolone): 349 � 196 mg

Biopsy-proven GCAN � 35 (78%)

Biopsy-negative casesN � 10 (22%)

323.5 � 186.1 427 � 215.442.4 � 16.1 55 � 16.4


Biopsy-negative casesN � 7 (35%)

813.5 � 240.4 918.4 � 403.239.9 � 9.98 40.6 � 15.3


Biopsy-negative casesN� 3 (60%)

1238.3 � 407.2 2325 � 1732.436.2 � 18 55.1 � 33.5

Biopsy-proven GCAN � 7 (87.5%)

Biopsy-negative casesN � 1 (12.5%)

165 � 58.7 6421256.7 � 336.7 3852

7.1 � 1.5 6

iopsy

ived i

In agreement with this study, we also observed that

pdptpwcgTslwaf

yGowptapfiao

atrtccttdMmwwm

msaTisl(ntAapc

wa

ospsttHor(dfclaymd1Tin

fmdfsdtAGsdwphdr

R


atients with newly diagnosed GCA treated with high-ose steroid therapy had a good TAB diagnostic yieldossible within 4 weeks of starting treatment, with posi-ive biopsy results in 74% (48/65) of the cases: in 78% ofatients who received corticosteroid therapy for up to 2eeks before TAB and in 65% of patients who received

orticosteroids between 2 and 4 weeks before biopsy. To-ether, these data confirm that the diagnostic yield ofAB in subjects with new GCA does not seem to be

ignificantly altered by high-dose steroid therapy for ateast 4 weeks. When the duration of treatment was longer,e observed that the TAB positivity rate fell to 40%,

lthough, due to the small sample size, this data requireurther confirmation.

It is of interest to emphasize that the decline in TABield was only observed in subjects with newly diagnosedCA. In contrast, in patients that developed GCA

n a background of a prior history of PMR, late TABas generally informative. In our experience, TAB wasositive in 88% (7/8) of these cases, despite long-termreatment with low doses of corticosteroids, which aredequate to improve symptoms and normalize the acute-hase reactants, but do not seem to influence the TABndings. Evolution of arteritis during the course of ther-py is infrequent clinically, reported in 5 to 7% of patientsriginally diagnosed with PMR (31-35).

Although it is still possible to make the diagnosis wellfter commencing corticosteroids, treatment does alterhe histopathological features of GCA (23,26). In thisegard, Achkar and coworkers observed that the propor-ion of positive specimens that yielded atypical histologi-al features of GCA (such as an absence of giant cells oronfinement of inflammation to the adventitia ratherhan in the media) was significantly higher in patientsreated more than 14 days with corticosteroids before un-ergoing biopsy, compared with untreated patients (23).ore recently, Font and colleagues (26) have docu-ented striking histologic differences between patientsith untreated GCA and patients who have been treatedith corticosteroids, as early as 4 to 7 days following treat-ent initiation (26).Another logical consequence of prebiopsy treatment isarked improvement in vascular inflammation. As a re-

ult, some biopsy specimens are characteristic of a healedrteritis (a footprint of active GCA that has been treated).he duration of treatment corresponding to this finding

s not entirely clear and is probably related to the cortico-teroid dosage used. In the main studies that have ana-yzed this issue, this period ranged from 33 to 82 days24,25). In our series, all positive biopsies in patients withewly diagnosed GCA showed features of active arteritis;he longest corticosteroid-to-biopsy interval was 46 days.mong patients that developed GCA on a background ofprior history of PMR, the mean duration of treatment inatients with active arteritis was 153 days (the longest
orticosteroid-to-biopsy interval was 210 days) compared
ith an average of 246 days in the 2 cases with healedrteritis.

Another important issue in the diagnosis of GCA is theptimal TAB sample size. GCA affects vessels focally andegmentally, yielding areas of inflammatory lesions juxta-osed with areas of normal artery. The length of arteryegment available for pathological review and examina-ion of multiple histological sections is therefore impor-ant for confirming the pathological diagnosis of GCA.owever, there is no unanimous consensus regarding the

ptimal length of biopsy. Initially, the majority of authorsecommended a generous biopsy of 4 to 6 cm in length3,36) to confirm a suspected diagnosis of GCA. Nowa-ays, segments of at least 2.5 cm are considered acceptableor pathological review (18,21). However, in a report re-ently published, Taylor-Gjevre and colleagues postu-ated that a threshold length of 1.0 cm postformalin fixedrterial segment was associated with increased diagnosticield of GCA (18). These authors recommended a mini-um TAB length of 1.5 cm to allow for tissue shrinkage

uring fixation, which was estimated to be approximately0%. In support of this, we have observed a relatively highAB positivity rate (near 75%) with a mean biopsy spec-

men length of 1.61 � 0.69 cm (1.55 � 0.43 cm inegative specimens).In summary, TAB is considered the gold standard test

or GCA diagnosis. Current medical practice recom-ends initiating corticosteroid treatment as soon as the

iagnosis is suspected, even before the biopsy is per-ormed. Although it is desirable to perform TAB before oroon after onset of treatment, in practice there is often aelay of several days or weeks in arranging surgery. Failingo do so should not be a reason for not performing TAB.ccording to our results, in patients with newly diagnosedCA, the diagnostic yield of TAB within 4 weeks of

tarting high-dose steroid treatment was 74%; when theuration of treatment extended beyond this time period,e observed that the TAB positivity rate fell to 40%. Inatients that developed GCA on a background of a prioristory of PMR, a late TAB is also generally informativeespite long-term treatment with low doses of corticoste-oids.

EFERENCES

1. Weyand CM, Goronzy MD. Giant-cell arteritis and polymyalgiarheumatica. Ann Intern Med 2003;139:505-15.

2. Smetana GW, Shmerling RH. Does this patient have temporalarteritis? JAMA 2002;287:92-101.

3. Hall S, Hunder GG. Is temporal artery biopsy prudent? MayoClin Proc 1984;59:793-6.

4. Fauchald P, Rygvold O, Oystese B. Temporal arteritis and poly-myalgia rheumatica. Clinical and biopsy findings. Ann InternMed 1972;77:845-52.

5. Allsop CJ, Gallagher PJ. Temporal artery biopsy in giant cellarteritis. A reappraisal. Am J Surg Pathol 1981;5:317-23.

6. Bengtsson BA, Malmvall BE. The epidemiology of giant cell ar-teritis including temporal arteritis and polymyalgia rheumatica:incidences of different clinical presentations and eye complica-
tions. Arthritis Rheum 1981;24:899-904.

1

1

1

11

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3


7. Hall S, Persellin S, Lie JT, O’Brien PC, Kurland LT, Hunder GG.The therapeutic impact of temporal artery biopsy. Lancet 1983;2:1217-20.

8. Hedges TR, Gieger GL, Albert DM. The clinical value of negativetemporal artery biopsy specimens. Arch Ophthalmol 1983;101:1251-4.

9. Nadeau SE. Temporal arteritis: a decision-analytic approach totemporal artery biopsy. Acta Neurol Scand 1988;78:90-100.

0. Fernández-Herlihy L. Temporal arteritis: clinical aids to diagno-sis. J Rheumatol 1988;15:1797-801.

1. Hunder GG, Bloch DA, Michel BA, Stevens MB, Arend WP,Calabrese LH, et al. The American College of Rheumatology1990 criteria for the classification of giant cell arteritis. ArthritisRheum 1990;33:1122-8.

2. Chmelewski WL, McKnight KM, Agudelo CA, Wise CM. Pre-senting features and outcomes in patients undergoing temporalartery biopsy. A review of 98 patients. Arch Intern Med 1992;152:1690-5.

3. Mason JC, Walport MJ. Giant cell arteritis. BMJ 1992;305:68-9.4. Gabriel SE, O’Fallon WM, Achkar AA, Lie JT, Hunder GG. The

use of clinical characteristics to predict the results of temporalartery biopsy among patients with suspected giant cell arteritis.J Rheumatol 1995;22:93-6.

5. Duhaut P, Pinède L, Bornet H, Demolombe-Ragué S, DumontetCH, Ninet J, et al. Biopsy proven and biopsy negative temporalarteritis: differences in clinical spectrum at the onset of the disease.Ann Rheum Dis 1999;58:335-41.

6. Bhatti MT, Tabendeh H. Giant cell arteritis: diagnosis and man-agement. Curr Opin Ophthalmol 2001;12:1217-20.

7. González-Gay MA. The diagnosis and management of patientswith giant cell arteritis. J Rheumatol 2005;32:1186-8.

8. Taylor-Gjevre R, Vo M, Shukla D, Resch L. Temporal arterybiopsy for giant cell arteritis. J Rheumatol 2005;32:1279-82.

9. Klein RG, Campbell RJ, Hunder GG, Carney JA. Skip lesions intemporal arteritis. Mayo Clin Proc 1976;51:504-10.

0. Poller DN, Van Wyk Q, Jeffrey MJ. The importance of skiplesions in temporal arteritis. J Clin Pathol 2000;53:137-9.

1. Allison MC, Gallagher PJ. Temporal artery biopsy and cortico-steroid treatment. Ann Rheum Dis 1984;43:416-7.

2. Liozon F, Lavignac C, Emilie D, Liozon E, Crevon MC, Vidal E,et al. Immunohistochemical study of lesions of temporal arteritisin Horton’s disease. Ann Med Interne (Paris) 1993;144:85-91.

3. Achkar AA, Lie JT, Hunder GG, O=Fallon M, Gabriel SE. How
does previous corticosteroid treatment affect the biopsy findings
in giant cell (temporal) arteritis. Ann Intern Med 1994;120:987-92.

4. Ray-Chaudhuri N, Kiné Dah, Tijani SO, Parums DV, CartlidgeN, Strong NP, et al. Effect of prior steroid treatment on temporalartery biopsy findings in giant cell arteritis. Br J Ophthalmol2002;86:530-2.

5. McDonnell PJ, Moore GW, Miller NR, Hutchins GM, GreenWR. Temporal arteritis: a clinicopathologic study. Ophthalmol-ogy 1986;93:518-30.

6. Font RL, Prabhakaran VC. Histologic parameters helpful to rec-ognize steroid-treated temporal arteritis. An analysis of thirty-fivecases. Br J Ophthalmol 2006; 20 [Epub ahead of print].

7. González-Gay MA, García-Porrua C, Llorca J, González-LouzaoC, Rodriguez-Ledo P. Biopsy-negative giant cell arteritis: clinicalspectrum and predictive factors for positive temporal artery bi-opsy. Semin Arthritis Rheum 2001;30:249-56.

8. To KW, Enzer YR, Tsiaras WG. Temporal artery biopsy after onemonth of corticosteroid therapy [letter]. Am J Ophthalmol 1994;117:265-7.

9. Guevara RA, Newman NJ, Grossniklaus HE. Positive temporalartery biopsy 6 months after prednisone treatment [letter]. ArchOphthalmol 1998;116:1252-3.

0. Kopelman H. Temporal artery biopsy [letter]. BMJ 1988;297:1470.

1. Narváez J, Nolla-Solé JM, Clavaguera MT, Valverde-García J,Roig-Escofet D. Temporal arteritis and polymyalgia rheumaticain north-eastern Spain: clinical spectrum and relationship over a15 year period. Joint Bone Spine 2003;70:33-9.

2. Narváez J, Nolla-Solé JM, Clavaguera MT, Valverde-García J,Roig-Escofet D. Long-term therapy in polymyalgia rheumatica:effect of coexistent temporal arteritis. J Rheumatol 1999;26:1945-52.

3. Myklebust G, Gran JT. A prospective study of 287 patients withpolymyalgia rheumatica and temporal arteritis: clinical and labo-ratory manifestations at onset of disease and at the time of diag-nosis. Br J Rheumatol 1996;35:1161-8.

4. Ayoub WT, Franklin CM, Torreti D. Polymyalgia rheumatica.Duration of therapy and long-term outcome. Am J Med 1985;79:309-15.

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

6. Kent RB 3rd, Thomas L. Temporal artery biopsy. Am Surg 1990;
56:16-21.

Documents

Influence of Previous Corticosteroid Therapy on Temporal Artery Biopsy Yield in Giant Cell Arteritis