Original Article
Temporal artery biopsy size does not matter
Vascular 2014, Vol. 22(6) 406–410 ! The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1708538113516322 vas.sagepub.com
Sarantos Kaptanis1, Joanne K Perera1, Constantine Halkias2, Nadine Caton1, Lida Alarcon1 and Stella Vig1
Abstract This study aimed to clarify whether positive temporal artery biopsies had a greater sample length than negative biopsies in temporal arteritis. It has been suggested that biopsy length should be at least 1 cm to improve diagnostic accuracy. A retrospective review of 149 patients who had 151 temporal artery biopsies was conducted. Twenty biopsies were positive (13.3%), 124 negative (82.1%) and seven samples were insufficient (4.6%). There was no clinically significant difference in the mean biopsy size between positive (0.7 cm) and negative samples (0.65 cm) (t-test: p ¼ .43 NS). Ninetyfour patients fulfilled all three ACR criteria prior to biopsy (62.3%) and four patients (2.6%) changed ACR score from 2 to 3 after biopsy. Treatment should not be delayed in anticipation of the biopsy or withheld in the case of a negative biopsy if the patient’s symptoms improve.
Keywords Giant cell arteritis, temporal artery biopsy, size, ACR criteria
Introduction Giant cell arteritis (GCA) is a large and medium vessel systemic vasculitis1 affecting, among other vessels, the temporal arteries. The American College of Rheumatology (ACR) has published2 a five-point scoring system for temporal arteritis (Table 1), which is still widely used, despite its shortcomings; recent critics include the European League Against Rheumatism (EULAR) consensus group,3 where 38% of the committee were dissatisfied with the definition. Temporal artery biopsy (TAB) can establish the diagnosis3 and is considered an important tool (Level of Evidence: Ia); however, a negative biopsy does not exclude GCA. A prompt TAB may lead to early withdrawal of unnecessary corticosteroid treatment4; it is suggested that corticosteroid treatment for one or two weeks5 does not affect the biopsy result, whereas TAB would be less typical after four weeks or more of treatment. TAB length remains a controversial issue. Size of any biopsy should be the minimum required to obtain a diagnosis in most cases. The TAB specimen contracts following harvesting after exposure to air and histologic processing, resulting in differences between the
preexcision, excision and post-fixation length of the artery.6 Small studies have suggested that this is 20% from excision to fixation (due to contraction of the artery)6 and a further 8.5% following fixation in formalin. According to the latest BSR/BHPR guidelines,7 TAB length should be at least 1 cm. This recommendation is based on a multicentre retrospective study of 1520 biopsies that found no statistically significant difference in size between the groups of positive (1.34 cm) and negative (1.33 cm) biopsies, and in fact suggested 0.5 cm as a sufficient length8; furthermore, the guidelines do not specify whether the 1 cm target is postfixation length or surgical specimen length. Various methods for TAB have been described, by rheumatologists,9 surgeons,10 ENT surgeons,11 dermatologists12 and ophthalmologists.13 Neurosurgeons have described an endoscopic-assisted method of 1 2
Croydon University Hospital, London, UK Imperial College Healthcare NHS Trust, London, UK
Corresponding author: Sarantos Kaptanis, Undergraduate Centre, Croydon University Hospital, 530 London Road, Thornton Heath, CR7 7YE, London, UK. Email: [email protected]
Kaptanis et al.
407
Table 1. ACR criteria for the diagnosis of GCA 1990 (modified from Hunder et al.2). Criterion
Definition
1
Age 50 years
2
Headache
3
Examination abnormal
4 5
ESR 50 mm/hour TAB abnormal
Development of symptoms at age 50 or older New onset, or new type, localized Temporal artery tenderness to palpation or abnormal pulses unrelated to atherosclerosis of the cervical arteries Westergren method
ESR: erythrocyte sedimentation rate; GCA: Giant cell arteritis; TAB: temporal artery biopsy.
harvesting the entire artery for the purpose of anastomosis with the middle cerebral artery.14 Identification of the temporal artery is by palpation or portable Doppler; the course of the artery is marked. Some surgeons shave the hair over the artery if they choose to harvest the parietal branch; then local anaesthetic is infiltrated with or without adrenaline. The area is prepared and draped. An incision, 2–8 cm long, is made through the epidermis and dermis just until subcutaneous fat presents. The superficial temporalis muscle fascia envelops the STA, and it should be carefully divided. An adequate length of artery is dissected and then ligated with ties before dividing the specimen and fixing it in formalin for histopathology; some surgeons use diathermy on the cut ends of the artery or on small branches; others prefer ties. Skin closure and dressings vary according to surgeon’s preference. Bilateral biopsies are rarely performed; most surgeons operate on the side where symptoms are most prominent. This retrospective audit aimed to explore the potential association between TAB specimen length and diagnosis of GCA in a single centre, and highlight the differences in surgical technique leading to the wide variation of collected specimen sizes.
Methods The operation TAB is performed mostly by junior surgeons in our institution, under Consultant supervision. Our method aims to avoid the danger zone where the temporal branch of the facial nerve may be injured, minimize scar, infection rates, wound dehiscence, bleeding, hematoma formation and damage to hair follicles which may lead to bald patches.
After marking the artery, a mixture of lidocaine and bupivacaine local anaesthetic (5–10 cc) is infiltrated; this provides adequate post-operative pain control. We do not use epinephrine to prevent arterial spasm. We aim to biopsy the frontal branch whenever possible, about 2 cm above the zygoma; this avoids the temporal branch of the facial nerve. We use a 2-cm incision to harvest a 2–3 cm specimen. Operating loupes are not routinely used. The artery and any branches are ligated prior to division with 3/0 Polyglactin 910 ties. Skin closure is with Polyglactin 910 deep dermal sutures and Poliglecaprone 25 subcuticular suture. Steri-strips or surgical tape is also used.
Histopathological examination The specimen is embedded in 10% formalin. We examine hematoxylin-eosin sections at three levels. The specimen is examined for the characteristic histological findings in GCA: inflammation (sometimes transmural), the presence of giant cells, intimal hyperplasia, fragmentation and disruption of the internal elastic lamina, arterial lumen occlusion.
Data collection A retrospective audit was registered with the hospital’s audit department, which provided ethics approval for the study. In all, 151 patients who had temporal artery biopsies from April 2006 to February 2012 were identified from pathology and theatre records. Erythrocyte sedimentation rate (ESR) results were retrieved from the patient information system. Data was entered in a Microsoft Excel (2010) spreadsheet. Statistical analysis was performed with Stata version 11.15
Literature review and audit criteria A literature review in Ovid Medline (MESH terms: giant cell arteritis, temporal arteries, biopsy and keywords: length or size) identified 29 reports of positive artery biopsy rate compared with sample length. Further studies were identified by cross-referencing. We sought to clarify whether our positive biopsies had a greater sample length than negative biopsies, in which case we would need to change our practice (biopsy method) to obtain larger samples. We set an arbitrary target of less than 10% insufficient samples for biopsy.
Results In all, 111 females (74.5%) and 38 males (25.5%) underwent a total of 151 temporal artery biopsies.
408
Vascular 22(6)
Table 2. Demographics and results.
Age
Positive
Negative
Insufficient sample
71.4 (SD 9.8)
68.8 (SD 12)
64.7 (SD 12)
Sex Male
4
33
2
Female
16
91
5
ESR
82.9 (SD 35.5))
61.4 (SD 33.4)
79.9 (SD 37.7)
TAB length (mm)
7 (SD 3.2)
6.4 (SD 3)
3.5 (SD 1.3)
ESR: erythrocyte sedimentation rate; TAB: temporal artery biopsy.
Two patients (one male, one female) had bilateral biopsies (all negative) in different occasions (21 months apart and 2 months apart, respectively). Mean age at biopsy was 68.9 years (range 19 to 95, SD 11.8); 144 biopsies were performed on people over 50 years old. An ESR result was available for 148 cases (98%). Mean ESR was 65.2 (SD 34.6); 95 cases had an ESR greater than 50. In total there were 20 positive (13.3%) and 124 negative (82.1%) biopsies; in seven (4.6%) cases the specimen was insufficient for diagnosis. The characteristics of the three groups are in Table 2. Out of the seven biopsies performed in patients less than 50 years old, six were negative and one was an insufficient sample. Out of the 53 cases with an ESR less than 50, four (7.5%) had temporal arteritis, 47 (88.7%) were negative and two (3.8%) were insufficient samples. Length of specimen (post-fixation) was available for 148 cases. Mean length was 6.4 mm (median 6, range 2 to 1.8, SD 3 mm). For positive biopsies mean was 7 mm (median 7, range 2 to 1.5, SD 3.2 mm). For negative biopsies mean was 6.4 mm (median 6, range 2 to 1.8, SD 3 mm) and for insufficient samples length of tissue mean was 3.6 mm (range 2 to 5, SD 1.3 mm). Excluding insufficient samples from further analysis, a t-test showed that the difference between positive and negative specimen length (0.6 mm, 95% confidence interval -0.9 to 2 mm) was not statistically significant (p ¼ .43). Logistic regression identified only ESR as a statistically significant predictor of positive biopsy (p ¼ .02), and not age (p ¼ .46), biopsy size (p ¼ .271) or sex (p ¼ .401). Twenty-seven (18.2%) specimens had a post-fixation length of 1 cm or greater; 76 (48.1%) specimens had a post-fixation length of 0.7 cm or greater. Introduction of the BSR/BHPR guidelines for the management of GCA in 20107 made no difference to harvested specimens. All patients presented with a headache; combining this with ESR, age and biopsy results, 94 patients
fulfilled all three criteria prior to biopsy. Four patients changed ACR classification after biopsy, as their ESR was below 50.
Discussion GCA is the most common systemic vasculitis in the elderly in the western world.16 It can result, without treatment, in irreversible visual loss due to arteritic anterior ischaemic neuropathy,17 arm or leg claudicatio,18 or even aortic dissection.19 Current practice is such that corticosteroids are commenced when a patient’s presentation is suspicious of GCA based on the ACR criteria (Table 1) and they are subsequently referred for a TAB. This procedure should then be undertaken within 2 weeks of initiating corticosteroid therapy to ensure that the inflammatory changes corresponding to a diagnosis of GCA have not resolved, thus leading to a false-negative result.5 Presently, the main implication of a positive temporal artery biopsy result is to clarify a decision on prolonging corticosteroid therapy in cases where the clinical presentation is not entirely diagnostic of GCA, i.e., in patients presenting with less than three of the five criteria, and in whom initial response to corticosteroids is not satisfactory. Conversely, corticosteroids are often continued despite a negative biopsy result, particularly in those with more than three criteria present. In our series, seven patients had a biopsy which resulted in an insufficient sample for diagnosis, only four of whom fulfilled three ACR criteria; none of these patients had a repeat biopsy requested. In this study, we examined length of biopsy and its impact on diagnosis of GCA. One would expect that the longer the segment, the higher the sensitivity, particularly as skip inflammatory lesions can occur in this condition. It has previously been suggested that an increased length of biopsy increases the sensitivity and several studies have recommended a segment length ranging between 2 and 7 cm. This is a strategy not supported by clinical data, as in all studies difference in mean length of positive and negative biopsies is only a few millimetres. The old recommendation for bilateral temporal artery biopsies20 has been abandoned after a number of studies demonstrated minimal benefit.21–23 We find it surprising that discussion continues regarding the optimum length of the unilateral biopsy. Ypsilantis et al.24 recently suggested that specimen length is important, on a retrospective cohort of 966 biopsies where median length of positive specimens was 1.2 cm and median length of negative specimens 1.0 cm; a 0.7 cm minimum post-fixation length was suggested. Mahr et al.8 previously studied 1534 biopsies with similar lengths (mean 1.34 cm for positive and 1.33 cm for negative biopsies, median 1.2 cm), finding
Kaptanis et al. no significant difference. The vast majority of smaller studies report mean and median lengths in this region.5,25–28 Small studies with biopsies of greater size achieve a similar positive rate.29 The 1983 study by Hall et al.,30 achieving median biopsy lengths of 4 cm for positive cases and 5 cm for negative cases, had a 34% positive rate over 134 cases. We doubt that a biopsy length of 17.5 cm would be deemed acceptable today. Arguably, large biopsies avoid the problem of skip lesions, which could occur, according to the Mayo Clinic series,31 in up to 28% of cases. A more recent study by Poller et al.32 has suggested that foci of discontinuous inflammation would occur in around only 10% of cases. This study also stated that skip lesions should be apparent on one set of sections from each segment of temporal artery from a specimen at least 6 mm long. If this is the case, then a post-fixation length of 6 mm should be adequate for histological diagnosis of GCA, negating the need for a more extensive biopsy and thus keeping the incision size to a minimum. It has been suggested that deeper sections,33 multiple levels25 or anti-CD83 immunohistochemistry34 may improve diagnostic accuracy; when quantitatively assessed, these methods have provided at most a 10% increase in sensitivity. It is important to consider the need for temporal artery biopsy given that in most cases, the ACR scoring system is able to provide a clinical diagnosis of GCA. Furthermore, the impact on management rarely differs with a negative TAB result. This has been concluded by Varma and O’Neill35 who studied the clinical application of TAB and discovered that it is of most benefit to patients whose diagnosis is less clear as the result impacts on the decision to continue corticosteroid therapy, which of course has its own adverse effects. The benefit of TAB needs to be carefully considered in those who have 3 of the ACR criteria as although TAB is a relatively simple local anaesthetic procedure, it does have recognised complications such as bleeding, infection, damage to the facial nerve,36 a scar which may be considered cosmetically undesirable and insufficient biopsy sample: these need to be mentioned to the patient while obtaining informed consent. Regarding samples insufficient for diagnosis, our rate of 4.6% is unfortunate. Previously, Mahr et al.37 reported 23/1821 (1.3%) biopsies to have sampling errors; we could not identify other major studies referring to this biopsy outcome. Possibly other authors tend to exclude such specimens from further analysis. In our experience, repeat or contralateral biopsies were not requested for these patients where an ‘‘insufficient sample’’ was reported. If TAB is still to be considered an integral component of GCA diagnosis despite the advent of methods such as ultrasound, CT arteriography, MRI
409 and FDG-PET,18 then perhaps the focus should be put on the minimum length of biopsy that is adequate for histological diagnosis to keep complications to a minimum. The largest meta-analysis comparing ultrasound and temporal artery biopsy so far38 included 988 patients in 17 studies, with a sensitivity of 75% and specificity of 83%. We await the results of the Temporal artery biopsy vs ultrasound (TABUL) in diagnosis of giant cell arteritis study to see whether ultrasound can replace TAB as a first-line investigation. Limitations of our study, because of its retrospective nature, are that there was no way to assess whether patients were already on corticosteroid treatment and for how long prior to the biopsy. Histological findings have not been correlated with follow up. As length of biopsies is not routinely measured or recorded by the operating surgeon in theatre, all measurements refer to post-fixation length measured by histopathology.
Conclusions Temporal artery biopsy is a procedure commonly requested of general and vascular surgeons for the confirmation of the diagnosis of GCAs. Our findings suggest that it can be competently performed by junior surgeons with minimal morbidity. There is no clear benefit in harvesting specimens longer than 0.6 cm post fixation. Corticosteroid treatment should not be delayed in anticipation of the biopsy or withheld in the case of a negative biopsy if the patient’s symptoms improve. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest None declared
References 1. Borchers AT and Gershwin ME. Giant cell arteritis: a review of classification, pathophysiology, geoepidemiology and treatment. Autoimmun Rev 2012; 11: A544–A554. 2. Hunder GG, Bloch DA, Michel BA, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990; 33: 1122–1128. 3. Basu N, Watts R, Bajema I, et al. EULAR points to consider in the development of classification and diagnostic criteria in systemic vasculitis. Ann Rheum Dis 2010; 69: 1744–1750. 4. Taylor-Gjevre R, Vo M, Shukla D, et al. Temporal artery biopsy for giant cell arteritis. J Rheumatol 2005; 32: 1279–1282.
410 5. Achkar AA, Lie JT, Hunder GG, et al. How does previous corticosteroid treatment affect the biopsy findings in giant cell (temporal) arteritis? Ann Intern Med 1994; 120: 987–992. 6. Su GW, Foroozan R and Yen MT. Quantitative analysis of temporal artery contraction after biopsy for evaluation of giant cell arteritis. Can J Ophthalmol 2006; 41: 500–503. 7. Dasgupta B, Borg FA, Hassan N, et al. BSR and BHPR guidelines for the management of giant cell arteritis. Rheumatology (Oxford) 2010; 49: 1594–1597. 8. Mahr A, Saba M, Kambouchner M, et al. Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? Ann Rheum Dis 2006; 65: 826–828. 9. Ton E and Kruize AA. How to perform and analyse biopsies in relation to connective tissue diseases. Best Pract Res Clin Rheumatol 2009; 23: 233–255. 10. Eskridge M and Perrier ND. Temporal artery biopsy. Oper Tech Gen Surg 2002; 4: 239–250. 11. Winkler A and Wudel J. Temporal artery biopsy. eMedicine from WebMD, Colorado, 2010. 12. Albertini JG, Ramsey ML and Marks VJ. Temporal artery biopsy in a dermatologic surgery practice. Dermatol Surg 1999; 25: 501–508. 13. Tomsak RL. Superficial temporal artery biopsy: a simplified technique. J Neuro-Ophthalmol 1991; 11: 202–204. 14. Kubo S, Takimoto H and Yoshimine T. Endoscopically assisted harvesting of the superficial temporal artery: technical note. Neurosurgery 2003; 52: 982–985. 15. StataCorp Stata Statistical Software: Release 11, College Station, TX, StataCorp LP, 2009. 16. Gonzalez-Gay MA, Garcia-Porrua C and MirandaFilloy JA. Giant cell arteritis: diagnosis and therapeutic management. Curr Rheumatol Rep 2006; 8: 299–302. 17. Matson M and Fujimoto L. Bilateral arteritic anterior ischemic optic neuropathy optometry. J Am Optometr Assoc 2011; 82: 622–631. 18. Janssen SP, Comans EH, Voskuyl AE, et al. Giant cell arteritis: Heterogeneity in clinical presentation and imaging results. J Vasc Surg 2008; 48: 1025–1031. 19. Liu G, Shupak R and Chiu BKY. Aortic dissection in giant-cell arteritis. Semin Arthritis Rheum 1995; 25: 160–171. 20. Ponge T, Barrier JH, Grolleau JY, et al. The efficacy of selective unilateral temporal artery biopsy versus bilateral biopsies for diagnosis of giant cell arteritis. J Rheumatol 1988; 15: 997–1000. 21. Boyev LR, Miller NR and Green WR. Efficacy of unilateral versus bilateral temporal artery biopsies for the diagnosis of giant cell arteritis. Am J Ophthalmol 1999; 128: 211–215.
Vascular 22(6) 22. Pless M, Rizzo JF 3rd, Lamkin JC, et al. Concordance of bilateral temporal artery biopsy in giant cell arteritis. J Neuroophthalmol 2000; 20: 216–218. 23. Danesh-Meyer HV, Savino PJ, Eagle RC Jr, et al. Low diagnostic yield with second biopsies in suspected giant cell arteritis. J Neuro-Ophthalmol 2000; 20: 213–215. 24. Ypsilantis E, Courtney ED, Chopra N, et al. Importance of specimen length during temporal artery biopsy. Br J Surg 2011; 98: 1556–1560. 25. Chakrabarty A and Franks AJ. Temporal artery biopsy: is there any value in examining biopsies at multiple levels? J Clin Pathol 2000; 53: 131–136. 26. Arashvand K. The value of temporal artery biopsy specimen length in the diagnosis of giant cell arteritis. J Rheumatol 2006; 33: 2363–2364; author reply 2364. 27. Dalbeth N, Lynch N, McLean L, et al. Audit of the management of suspected giant cell arteritis in a large teaching hospital Intern. Med J 2002; 32: 315–319. 28. Caroe A. Temporal artery biopsy to diagnose temporal arteritis. JAMA 1998; 280: 1992. 29. Cetinkaya A, Kersten RC, Brannan PA, et al. Intraoperative predictability of temporal artery biopsy results. Ophthal Plast Reconstr Surg 2008; 24: 372–376; discussion 377. 30. Hall S, Lie JT, Kurland LT, et al. The therapeutic impact of temporal artery biopsy. Lancet 1983; 322: 1217–1220. 31. Klein RG, Campbell RJ, Hunder GG, et al. Skip lesions in temporal arteritis. Mayo Clin Proc 1976; 51: 504–510. 32. Poller DN, van Wyk Q and Jeffrey MJ. The importance of skip lesions in temporal arteritis. J Clin Pathol 2000; 53: 137–139. 33. Gooi P, Brownstein S and Rawlings N. Temporal arteritis: a dilemma in clinical and pathological diagnosis. Can J Ophthalmol 2008; 43: 119–120. 34. Slobodin G, Lurie M, Bejar J, et al. Biopsy-negative giant cell arteritis: does anti-CD83 immunohistochemistry advance the diagnosis? Eur J Intern Med 2007; 18: 405–408. 35. Varma D and O’Neill D. Quantification of the role of temporal artery biopsy in diagnosing clinically suspected giant cell arteritis. Eye (Lond) 2004; 18: 384–388. 36. Yoon MK, Horton JC and McCulley TJ. Facial nerve injury: a complication of superficial temporal artery biopsy. Am J Ophthalmol 2011; 152: 251–255.e251. 37. Mahr A, Saba M, Kambouchner M, et al. Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? Ann Rheum Dis 2006; 65: 826–828. 38. Ball EL, Walsh SR, Tang TY, et al. Role of ultrasonography in the diagnosis of temporal arteritis. Br J Surg 2010; 97: 1765–1771.
Temporal artery biopsy size does not matter
Vascular 2014, Vol. 22(6) 406–410 ! The Author(s) 2013 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1708538113516322 vas.sagepub.com
Sarantos Kaptanis1, Joanne K Perera1, Constantine Halkias2, Nadine Caton1, Lida Alarcon1 and Stella Vig1
Abstract This study aimed to clarify whether positive temporal artery biopsies had a greater sample length than negative biopsies in temporal arteritis. It has been suggested that biopsy length should be at least 1 cm to improve diagnostic accuracy. A retrospective review of 149 patients who had 151 temporal artery biopsies was conducted. Twenty biopsies were positive (13.3%), 124 negative (82.1%) and seven samples were insufficient (4.6%). There was no clinically significant difference in the mean biopsy size between positive (0.7 cm) and negative samples (0.65 cm) (t-test: p ¼ .43 NS). Ninetyfour patients fulfilled all three ACR criteria prior to biopsy (62.3%) and four patients (2.6%) changed ACR score from 2 to 3 after biopsy. Treatment should not be delayed in anticipation of the biopsy or withheld in the case of a negative biopsy if the patient’s symptoms improve.
Keywords Giant cell arteritis, temporal artery biopsy, size, ACR criteria
Introduction Giant cell arteritis (GCA) is a large and medium vessel systemic vasculitis1 affecting, among other vessels, the temporal arteries. The American College of Rheumatology (ACR) has published2 a five-point scoring system for temporal arteritis (Table 1), which is still widely used, despite its shortcomings; recent critics include the European League Against Rheumatism (EULAR) consensus group,3 where 38% of the committee were dissatisfied with the definition. Temporal artery biopsy (TAB) can establish the diagnosis3 and is considered an important tool (Level of Evidence: Ia); however, a negative biopsy does not exclude GCA. A prompt TAB may lead to early withdrawal of unnecessary corticosteroid treatment4; it is suggested that corticosteroid treatment for one or two weeks5 does not affect the biopsy result, whereas TAB would be less typical after four weeks or more of treatment. TAB length remains a controversial issue. Size of any biopsy should be the minimum required to obtain a diagnosis in most cases. The TAB specimen contracts following harvesting after exposure to air and histologic processing, resulting in differences between the
preexcision, excision and post-fixation length of the artery.6 Small studies have suggested that this is 20% from excision to fixation (due to contraction of the artery)6 and a further 8.5% following fixation in formalin. According to the latest BSR/BHPR guidelines,7 TAB length should be at least 1 cm. This recommendation is based on a multicentre retrospective study of 1520 biopsies that found no statistically significant difference in size between the groups of positive (1.34 cm) and negative (1.33 cm) biopsies, and in fact suggested 0.5 cm as a sufficient length8; furthermore, the guidelines do not specify whether the 1 cm target is postfixation length or surgical specimen length. Various methods for TAB have been described, by rheumatologists,9 surgeons,10 ENT surgeons,11 dermatologists12 and ophthalmologists.13 Neurosurgeons have described an endoscopic-assisted method of 1 2
Croydon University Hospital, London, UK Imperial College Healthcare NHS Trust, London, UK
Corresponding author: Sarantos Kaptanis, Undergraduate Centre, Croydon University Hospital, 530 London Road, Thornton Heath, CR7 7YE, London, UK. Email: [email protected]
Kaptanis et al.
407
Table 1. ACR criteria for the diagnosis of GCA 1990 (modified from Hunder et al.2). Criterion
Definition
1
Age 50 years
2
Headache
3
Examination abnormal
4 5
ESR 50 mm/hour TAB abnormal
Development of symptoms at age 50 or older New onset, or new type, localized Temporal artery tenderness to palpation or abnormal pulses unrelated to atherosclerosis of the cervical arteries Westergren method
ESR: erythrocyte sedimentation rate; GCA: Giant cell arteritis; TAB: temporal artery biopsy.
harvesting the entire artery for the purpose of anastomosis with the middle cerebral artery.14 Identification of the temporal artery is by palpation or portable Doppler; the course of the artery is marked. Some surgeons shave the hair over the artery if they choose to harvest the parietal branch; then local anaesthetic is infiltrated with or without adrenaline. The area is prepared and draped. An incision, 2–8 cm long, is made through the epidermis and dermis just until subcutaneous fat presents. The superficial temporalis muscle fascia envelops the STA, and it should be carefully divided. An adequate length of artery is dissected and then ligated with ties before dividing the specimen and fixing it in formalin for histopathology; some surgeons use diathermy on the cut ends of the artery or on small branches; others prefer ties. Skin closure and dressings vary according to surgeon’s preference. Bilateral biopsies are rarely performed; most surgeons operate on the side where symptoms are most prominent. This retrospective audit aimed to explore the potential association between TAB specimen length and diagnosis of GCA in a single centre, and highlight the differences in surgical technique leading to the wide variation of collected specimen sizes.
Methods The operation TAB is performed mostly by junior surgeons in our institution, under Consultant supervision. Our method aims to avoid the danger zone where the temporal branch of the facial nerve may be injured, minimize scar, infection rates, wound dehiscence, bleeding, hematoma formation and damage to hair follicles which may lead to bald patches.
After marking the artery, a mixture of lidocaine and bupivacaine local anaesthetic (5–10 cc) is infiltrated; this provides adequate post-operative pain control. We do not use epinephrine to prevent arterial spasm. We aim to biopsy the frontal branch whenever possible, about 2 cm above the zygoma; this avoids the temporal branch of the facial nerve. We use a 2-cm incision to harvest a 2–3 cm specimen. Operating loupes are not routinely used. The artery and any branches are ligated prior to division with 3/0 Polyglactin 910 ties. Skin closure is with Polyglactin 910 deep dermal sutures and Poliglecaprone 25 subcuticular suture. Steri-strips or surgical tape is also used.
Histopathological examination The specimen is embedded in 10% formalin. We examine hematoxylin-eosin sections at three levels. The specimen is examined for the characteristic histological findings in GCA: inflammation (sometimes transmural), the presence of giant cells, intimal hyperplasia, fragmentation and disruption of the internal elastic lamina, arterial lumen occlusion.
Data collection A retrospective audit was registered with the hospital’s audit department, which provided ethics approval for the study. In all, 151 patients who had temporal artery biopsies from April 2006 to February 2012 were identified from pathology and theatre records. Erythrocyte sedimentation rate (ESR) results were retrieved from the patient information system. Data was entered in a Microsoft Excel (2010) spreadsheet. Statistical analysis was performed with Stata version 11.15
Literature review and audit criteria A literature review in Ovid Medline (MESH terms: giant cell arteritis, temporal arteries, biopsy and keywords: length or size) identified 29 reports of positive artery biopsy rate compared with sample length. Further studies were identified by cross-referencing. We sought to clarify whether our positive biopsies had a greater sample length than negative biopsies, in which case we would need to change our practice (biopsy method) to obtain larger samples. We set an arbitrary target of less than 10% insufficient samples for biopsy.
Results In all, 111 females (74.5%) and 38 males (25.5%) underwent a total of 151 temporal artery biopsies.
408
Vascular 22(6)
Table 2. Demographics and results.
Age
Positive
Negative
Insufficient sample
71.4 (SD 9.8)
68.8 (SD 12)
64.7 (SD 12)
Sex Male
4
33
2
Female
16
91
5
ESR
82.9 (SD 35.5))
61.4 (SD 33.4)
79.9 (SD 37.7)
TAB length (mm)
7 (SD 3.2)
6.4 (SD 3)
3.5 (SD 1.3)
ESR: erythrocyte sedimentation rate; TAB: temporal artery biopsy.
Two patients (one male, one female) had bilateral biopsies (all negative) in different occasions (21 months apart and 2 months apart, respectively). Mean age at biopsy was 68.9 years (range 19 to 95, SD 11.8); 144 biopsies were performed on people over 50 years old. An ESR result was available for 148 cases (98%). Mean ESR was 65.2 (SD 34.6); 95 cases had an ESR greater than 50. In total there were 20 positive (13.3%) and 124 negative (82.1%) biopsies; in seven (4.6%) cases the specimen was insufficient for diagnosis. The characteristics of the three groups are in Table 2. Out of the seven biopsies performed in patients less than 50 years old, six were negative and one was an insufficient sample. Out of the 53 cases with an ESR less than 50, four (7.5%) had temporal arteritis, 47 (88.7%) were negative and two (3.8%) were insufficient samples. Length of specimen (post-fixation) was available for 148 cases. Mean length was 6.4 mm (median 6, range 2 to 1.8, SD 3 mm). For positive biopsies mean was 7 mm (median 7, range 2 to 1.5, SD 3.2 mm). For negative biopsies mean was 6.4 mm (median 6, range 2 to 1.8, SD 3 mm) and for insufficient samples length of tissue mean was 3.6 mm (range 2 to 5, SD 1.3 mm). Excluding insufficient samples from further analysis, a t-test showed that the difference between positive and negative specimen length (0.6 mm, 95% confidence interval -0.9 to 2 mm) was not statistically significant (p ¼ .43). Logistic regression identified only ESR as a statistically significant predictor of positive biopsy (p ¼ .02), and not age (p ¼ .46), biopsy size (p ¼ .271) or sex (p ¼ .401). Twenty-seven (18.2%) specimens had a post-fixation length of 1 cm or greater; 76 (48.1%) specimens had a post-fixation length of 0.7 cm or greater. Introduction of the BSR/BHPR guidelines for the management of GCA in 20107 made no difference to harvested specimens. All patients presented with a headache; combining this with ESR, age and biopsy results, 94 patients
fulfilled all three criteria prior to biopsy. Four patients changed ACR classification after biopsy, as their ESR was below 50.
Discussion GCA is the most common systemic vasculitis in the elderly in the western world.16 It can result, without treatment, in irreversible visual loss due to arteritic anterior ischaemic neuropathy,17 arm or leg claudicatio,18 or even aortic dissection.19 Current practice is such that corticosteroids are commenced when a patient’s presentation is suspicious of GCA based on the ACR criteria (Table 1) and they are subsequently referred for a TAB. This procedure should then be undertaken within 2 weeks of initiating corticosteroid therapy to ensure that the inflammatory changes corresponding to a diagnosis of GCA have not resolved, thus leading to a false-negative result.5 Presently, the main implication of a positive temporal artery biopsy result is to clarify a decision on prolonging corticosteroid therapy in cases where the clinical presentation is not entirely diagnostic of GCA, i.e., in patients presenting with less than three of the five criteria, and in whom initial response to corticosteroids is not satisfactory. Conversely, corticosteroids are often continued despite a negative biopsy result, particularly in those with more than three criteria present. In our series, seven patients had a biopsy which resulted in an insufficient sample for diagnosis, only four of whom fulfilled three ACR criteria; none of these patients had a repeat biopsy requested. In this study, we examined length of biopsy and its impact on diagnosis of GCA. One would expect that the longer the segment, the higher the sensitivity, particularly as skip inflammatory lesions can occur in this condition. It has previously been suggested that an increased length of biopsy increases the sensitivity and several studies have recommended a segment length ranging between 2 and 7 cm. This is a strategy not supported by clinical data, as in all studies difference in mean length of positive and negative biopsies is only a few millimetres. The old recommendation for bilateral temporal artery biopsies20 has been abandoned after a number of studies demonstrated minimal benefit.21–23 We find it surprising that discussion continues regarding the optimum length of the unilateral biopsy. Ypsilantis et al.24 recently suggested that specimen length is important, on a retrospective cohort of 966 biopsies where median length of positive specimens was 1.2 cm and median length of negative specimens 1.0 cm; a 0.7 cm minimum post-fixation length was suggested. Mahr et al.8 previously studied 1534 biopsies with similar lengths (mean 1.34 cm for positive and 1.33 cm for negative biopsies, median 1.2 cm), finding
Kaptanis et al. no significant difference. The vast majority of smaller studies report mean and median lengths in this region.5,25–28 Small studies with biopsies of greater size achieve a similar positive rate.29 The 1983 study by Hall et al.,30 achieving median biopsy lengths of 4 cm for positive cases and 5 cm for negative cases, had a 34% positive rate over 134 cases. We doubt that a biopsy length of 17.5 cm would be deemed acceptable today. Arguably, large biopsies avoid the problem of skip lesions, which could occur, according to the Mayo Clinic series,31 in up to 28% of cases. A more recent study by Poller et al.32 has suggested that foci of discontinuous inflammation would occur in around only 10% of cases. This study also stated that skip lesions should be apparent on one set of sections from each segment of temporal artery from a specimen at least 6 mm long. If this is the case, then a post-fixation length of 6 mm should be adequate for histological diagnosis of GCA, negating the need for a more extensive biopsy and thus keeping the incision size to a minimum. It has been suggested that deeper sections,33 multiple levels25 or anti-CD83 immunohistochemistry34 may improve diagnostic accuracy; when quantitatively assessed, these methods have provided at most a 10% increase in sensitivity. It is important to consider the need for temporal artery biopsy given that in most cases, the ACR scoring system is able to provide a clinical diagnosis of GCA. Furthermore, the impact on management rarely differs with a negative TAB result. This has been concluded by Varma and O’Neill35 who studied the clinical application of TAB and discovered that it is of most benefit to patients whose diagnosis is less clear as the result impacts on the decision to continue corticosteroid therapy, which of course has its own adverse effects. The benefit of TAB needs to be carefully considered in those who have 3 of the ACR criteria as although TAB is a relatively simple local anaesthetic procedure, it does have recognised complications such as bleeding, infection, damage to the facial nerve,36 a scar which may be considered cosmetically undesirable and insufficient biopsy sample: these need to be mentioned to the patient while obtaining informed consent. Regarding samples insufficient for diagnosis, our rate of 4.6% is unfortunate. Previously, Mahr et al.37 reported 23/1821 (1.3%) biopsies to have sampling errors; we could not identify other major studies referring to this biopsy outcome. Possibly other authors tend to exclude such specimens from further analysis. In our experience, repeat or contralateral biopsies were not requested for these patients where an ‘‘insufficient sample’’ was reported. If TAB is still to be considered an integral component of GCA diagnosis despite the advent of methods such as ultrasound, CT arteriography, MRI
409 and FDG-PET,18 then perhaps the focus should be put on the minimum length of biopsy that is adequate for histological diagnosis to keep complications to a minimum. The largest meta-analysis comparing ultrasound and temporal artery biopsy so far38 included 988 patients in 17 studies, with a sensitivity of 75% and specificity of 83%. We await the results of the Temporal artery biopsy vs ultrasound (TABUL) in diagnosis of giant cell arteritis study to see whether ultrasound can replace TAB as a first-line investigation. Limitations of our study, because of its retrospective nature, are that there was no way to assess whether patients were already on corticosteroid treatment and for how long prior to the biopsy. Histological findings have not been correlated with follow up. As length of biopsies is not routinely measured or recorded by the operating surgeon in theatre, all measurements refer to post-fixation length measured by histopathology.
Conclusions Temporal artery biopsy is a procedure commonly requested of general and vascular surgeons for the confirmation of the diagnosis of GCAs. Our findings suggest that it can be competently performed by junior surgeons with minimal morbidity. There is no clear benefit in harvesting specimens longer than 0.6 cm post fixation. Corticosteroid treatment should not be delayed in anticipation of the biopsy or withheld in the case of a negative biopsy if the patient’s symptoms improve. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest None declared
References 1. Borchers AT and Gershwin ME. Giant cell arteritis: a review of classification, pathophysiology, geoepidemiology and treatment. Autoimmun Rev 2012; 11: A544–A554. 2. Hunder GG, Bloch DA, Michel BA, et al. The American College of Rheumatology 1990 criteria for the classification of giant cell arteritis. Arthritis Rheum 1990; 33: 1122–1128. 3. Basu N, Watts R, Bajema I, et al. EULAR points to consider in the development of classification and diagnostic criteria in systemic vasculitis. Ann Rheum Dis 2010; 69: 1744–1750. 4. Taylor-Gjevre R, Vo M, Shukla D, et al. Temporal artery biopsy for giant cell arteritis. J Rheumatol 2005; 32: 1279–1282.
410 5. Achkar AA, Lie JT, Hunder GG, et al. How does previous corticosteroid treatment affect the biopsy findings in giant cell (temporal) arteritis? Ann Intern Med 1994; 120: 987–992. 6. Su GW, Foroozan R and Yen MT. Quantitative analysis of temporal artery contraction after biopsy for evaluation of giant cell arteritis. Can J Ophthalmol 2006; 41: 500–503. 7. Dasgupta B, Borg FA, Hassan N, et al. BSR and BHPR guidelines for the management of giant cell arteritis. Rheumatology (Oxford) 2010; 49: 1594–1597. 8. Mahr A, Saba M, Kambouchner M, et al. Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? Ann Rheum Dis 2006; 65: 826–828. 9. Ton E and Kruize AA. How to perform and analyse biopsies in relation to connective tissue diseases. Best Pract Res Clin Rheumatol 2009; 23: 233–255. 10. Eskridge M and Perrier ND. Temporal artery biopsy. Oper Tech Gen Surg 2002; 4: 239–250. 11. Winkler A and Wudel J. Temporal artery biopsy. eMedicine from WebMD, Colorado, 2010. 12. Albertini JG, Ramsey ML and Marks VJ. Temporal artery biopsy in a dermatologic surgery practice. Dermatol Surg 1999; 25: 501–508. 13. Tomsak RL. Superficial temporal artery biopsy: a simplified technique. J Neuro-Ophthalmol 1991; 11: 202–204. 14. Kubo S, Takimoto H and Yoshimine T. Endoscopically assisted harvesting of the superficial temporal artery: technical note. Neurosurgery 2003; 52: 982–985. 15. StataCorp Stata Statistical Software: Release 11, College Station, TX, StataCorp LP, 2009. 16. Gonzalez-Gay MA, Garcia-Porrua C and MirandaFilloy JA. Giant cell arteritis: diagnosis and therapeutic management. Curr Rheumatol Rep 2006; 8: 299–302. 17. Matson M and Fujimoto L. Bilateral arteritic anterior ischemic optic neuropathy optometry. J Am Optometr Assoc 2011; 82: 622–631. 18. Janssen SP, Comans EH, Voskuyl AE, et al. Giant cell arteritis: Heterogeneity in clinical presentation and imaging results. J Vasc Surg 2008; 48: 1025–1031. 19. Liu G, Shupak R and Chiu BKY. Aortic dissection in giant-cell arteritis. Semin Arthritis Rheum 1995; 25: 160–171. 20. Ponge T, Barrier JH, Grolleau JY, et al. The efficacy of selective unilateral temporal artery biopsy versus bilateral biopsies for diagnosis of giant cell arteritis. J Rheumatol 1988; 15: 997–1000. 21. Boyev LR, Miller NR and Green WR. Efficacy of unilateral versus bilateral temporal artery biopsies for the diagnosis of giant cell arteritis. Am J Ophthalmol 1999; 128: 211–215.
Vascular 22(6) 22. Pless M, Rizzo JF 3rd, Lamkin JC, et al. Concordance of bilateral temporal artery biopsy in giant cell arteritis. J Neuroophthalmol 2000; 20: 216–218. 23. Danesh-Meyer HV, Savino PJ, Eagle RC Jr, et al. Low diagnostic yield with second biopsies in suspected giant cell arteritis. J Neuro-Ophthalmol 2000; 20: 213–215. 24. Ypsilantis E, Courtney ED, Chopra N, et al. Importance of specimen length during temporal artery biopsy. Br J Surg 2011; 98: 1556–1560. 25. Chakrabarty A and Franks AJ. Temporal artery biopsy: is there any value in examining biopsies at multiple levels? J Clin Pathol 2000; 53: 131–136. 26. Arashvand K. The value of temporal artery biopsy specimen length in the diagnosis of giant cell arteritis. J Rheumatol 2006; 33: 2363–2364; author reply 2364. 27. Dalbeth N, Lynch N, McLean L, et al. Audit of the management of suspected giant cell arteritis in a large teaching hospital Intern. Med J 2002; 32: 315–319. 28. Caroe A. Temporal artery biopsy to diagnose temporal arteritis. JAMA 1998; 280: 1992. 29. Cetinkaya A, Kersten RC, Brannan PA, et al. Intraoperative predictability of temporal artery biopsy results. Ophthal Plast Reconstr Surg 2008; 24: 372–376; discussion 377. 30. Hall S, Lie JT, Kurland LT, et al. The therapeutic impact of temporal artery biopsy. Lancet 1983; 322: 1217–1220. 31. Klein RG, Campbell RJ, Hunder GG, et al. Skip lesions in temporal arteritis. Mayo Clin Proc 1976; 51: 504–510. 32. Poller DN, van Wyk Q and Jeffrey MJ. The importance of skip lesions in temporal arteritis. J Clin Pathol 2000; 53: 137–139. 33. Gooi P, Brownstein S and Rawlings N. Temporal arteritis: a dilemma in clinical and pathological diagnosis. Can J Ophthalmol 2008; 43: 119–120. 34. Slobodin G, Lurie M, Bejar J, et al. Biopsy-negative giant cell arteritis: does anti-CD83 immunohistochemistry advance the diagnosis? Eur J Intern Med 2007; 18: 405–408. 35. Varma D and O’Neill D. Quantification of the role of temporal artery biopsy in diagnosing clinically suspected giant cell arteritis. Eye (Lond) 2004; 18: 384–388. 36. Yoon MK, Horton JC and McCulley TJ. Facial nerve injury: a complication of superficial temporal artery biopsy. Am J Ophthalmol 2011; 152: 251–255.e251. 37. Mahr A, Saba M, Kambouchner M, et al. Temporal artery biopsy for diagnosing giant cell arteritis: the longer, the better? Ann Rheum Dis 2006; 65: 826–828. 38. Ball EL, Walsh SR, Tang TY, et al. Role of ultrasonography in the diagnosis of temporal arteritis. Br J Surg 2010; 97: 1765–1771.
