Takayasu's Disease: Diagnostic and Therapeutic Value of Subclavian Artery Biopsy Anne Long,* MD, Patrick Bruneval,* MD, Claude Laurian, MD, ~ Jean-Michel Cormier, ~ MD, Jean-Pierre Camilleri, MD,* and Jean-NoEl Fiessinger,* MD, Paris, France

Seventeen surgical biopsy specimens of the axillary or subclavian artery were obtained from patients who had inflammatory, nonatherosclerotic arterial lesions and were studied retrospectively. Initial roentgenoiogic clinical, and histologic findings were evaluated and compared with final diagnosis. Histological examination suggested Takayasu's disease in 11 patients whereas the roentgenologic and clinical signs led us to suspect Takayasu's disease in only eight instances. Takayasu's disease was confirmed in these eight cases later in the course of the disease. In six patients subclavian artery biopsy specimens showed lesions consistent with Horton's disease in contrast to roentgenologic and clinical signs which suggested Takayasu's disease. In five patients, the final diagnosis was Horton's disease and in one, the diagnosis remained unknown. Of 13 patients who had revascularization procedures, bypasses remained patent in patients followed from nine months to 16 years (mean: 68 months). Histopathologic criteria are more reliable than roentgenologic or clinical signs in the diagnosis of Takayasu's disease. The subclavian artery is frequently involved in Takayasu's disease and is a desirable site for retrieval of diagnostic specimen when a revascularization procedure is entertained. (Ann Vasc Surg 1990;4:151 - 155). KEY WORDS: Takayasu's disease; subclavian artery biopsy; Horton's disease; arteritis.

Takayasu's disease is characterized by inflammatory lesions of the aorta or the larger vessels, and tends to occur in young people [1-8]. In most published series, diagnosis has been based on roentgenographic or clinical evidence [3-5,9,10]. The histopathological lesions which characterize TakaFrom the Service de Pathologie Vasculaire and the ¢ Laboratoire d'Anatomopathotologie, HOpital Broussais, and the Service de Chirurgie Vasculaire, "~HOpital Saint Joseph, Paris, France. Presented at the Annual Meeting of the Society de Chirurgie Vascutaire de Langue Franqaise, Strasbourg, France, June 23-24 1989. Reprint requests: Anne Long, MD, Service de Pathotogie Vasculaire, H6pital Broussais, 96 rue Didot, 75674 Paris Cedex 14, France.

yasu's disease [2,6,7,11] are rarely included in diagnostic criteria because they have not been considered as specific and may be confused with Horton's disease [9]. The goal of this study was to evaluate the limitations of the roentgenologic and clinical aspects of Takayasu's disease, based on a retrospective study of 17 subclavian artery specimens obtained because Takayasu's disease was suspected. The importance of histopathological examination to final diagnosis is emphasized. PATIENTS AND METHODS Seventeen consecutive biopsy specimens of the subclavian artery showing non-atherosclerotic in151

ANNALS OF VASCULAR SURGERY

S U B C L A V I A N A R TER Y B I O P S Y

152

TABLE I.mClinical criteria of Takayasu's disease (according to Bletry) Clinical and Biologic Criteria Age less than 40 years Absence of vascular risk factors (hyperlipidemia, diabetes, tobacco consumption, estrogen/progesterone replacement drugs) Biologic inflammatory syndrome (ESR > 20 mm at first hour or fibrinemia > 4 g/I) Arterial Criteria Stenosis or aneurysm of the descending thoracic or suprarenal aorta Ectasia of the descending thoracic or suprarenal aorta Stenosis or occlusion of the postvertebral subclavian artery Stenosis or occlusion of the common carotid artery Multiple site arterial lesions Arterial wall thickening (sonography, CT scan, surgery) Criteria of exclusion: Horton's disease Characteristic cephalalgia or biopsy specimens positive for Horton's disease, after age 55 years. Diagnosis is retained when at least three criteria (two arterial) are present.

flammatory arteritis were obtained between 1971 and 1987 in the Pathology Laboratory of Broussais Hospital. Clinical reports and roentgenologic data were obtained for these patients, who had been hospitalized in the Vascular Disease Unit of Broussais Hospital. These data were retrospectively analyzed and compared to the histologic diagnosis. There were 17 patients, 15 women and two men, mean age 43 years (25-65 years). Radioclinical diagnosis

All clinical, biologic, and arteriographic findings were analyzed in the search for arguments in favor of the diagnosis of Takayasu's disease, according to previously proposed diagnostic criteria (Table I) [12]. The aspect and extension of associated arterial lesions, the presence or absence of thickened arterial walls, the presence or absence of an inflammatory syndrome (erythrocyte sedimentation rate > 20 mm in one hour and/or fibrinogenemia > 4 g/l) and, in three patients, negative results from histological examination of a temporal artery specimen enabled us to define two groups of patients: group A had a strong presumption of Takayasu's disease because the patients had suggestive criteria and group B had a weak presumption of Takayasu's disease because the arguments were not sufficient to establish its diagnosis. Histopathologic diagnosis

Biopsy specimens were obtained through a supraclavicular approach to the subclavian artery. In

each case a cylindrical arterial segment was retrieved, enabling the pathologist to study all three tunica of the artery. A revascularization procedure was indicated during the same operation in 13 patients. Seven patients had severe chronic ischemia in the upper limb. Six arteries were patent. The procedure performed was a carotidobrachial or subclavian-toaxillary bypass. The conduit used was a prosthetic graft in two cases and vein in 11 cases. Four patients with functionally well-tolerated chronic ischemia associated with an occluded subclavian artery did not undergo a revascularization procedure. The 17 specimens were studied under optical microscopy by two different pathologists using orcein, hematoxylin-eosin-safranin, and Masson's trichrome stains. Histopathologic diagnosis of Takayasu's disease was made when at least three of the following criteria were found [13]: adventitial and external medial sclerosis, destruction of the external elastic lamella, thickening of the intima, or medioadventitial inflammatory infiltrate. In several specimens, the findings revealed interruption of elastic fibers near the internal tunica, intimal-medial inflammation, numerous giant cells, or medial infarctoid necrosis [13,14]. The coexistence of at least three of these lesions was suggestive of the diagnosis of Horton's disease. Follow-up

During the follow-up period, the histopathological diagnoses were compared with the initial clinical findings and the final diagnosis which included the appearance of new arterial lesions in specific sites. Follow-up ranged from nine months to 16 years with a mean of 43 months. During this same period, the patency of the bypass was verified by palpation of pulses, sonography or arteriography.

RESULTS There were 14 patients in group A and three in group B. Histopathological findings suggested Takayasu's disease in I1 cases (Fig. 1) and Horton's disease in six cases (Fig. 2). Final diagnosis was Takayasu's disease in 11 patients and Horton's disease in five. In one instance, diagnosis remained unknown. According to Bletry's score [12], 14 patients (Group A) were suspected of having Takayasu's disease based on clinical findings. Takayasu's disease was confirmed by histology, then, during follow-up, by the appearance of new arterial lesions in specific sites in only eight of these patients (mean age 35 years, range 25-51 years). Paradoxically, six patients had subclavian artery

VOLUME 4 N o 2 - 1990

SUBCLAVIAN ARTERY BIOPSY

a

153

b

Fig. 1. Subclavian involvement in Takayasu's disease. (a) Adventitial and external medial sclerosis (stars). Lumen (L). Masson's trichrome stain (X 20), (b) Medioadventitial area: external elastic lamellae are destroyed by sclerosis and inflammatory reaction. Adventitia (A), media (M). Orcein x 190.

specimens which were histopathologically suggestive of Horton's disease. Five patients aged more than 55 years (range 57--65 years) had a history of atypical headache or scapular arthralgia, in two instances associated with poor general health. Erythrocyte sedimentation rate was invariably accelerated. In three patients, the diagnosis of Horton's disease was entertained in spite of the absence of clinical signs. The biopsy specimen of the temporal artery showed ateriosclerotic lesions only. For these five patients, a late diagnosis of Takayasu's disease could not be excluded according to the course of the disease based on radioclinical signs and Bletry's exclusion criteria. The remaining pa-

a

tient, a 30 year-old woman, presented with a stenosis of the common carotid subclavian artery and the descending thoracic aorta. Her erythrocyte sedimentation rate was high. Her disease was controlled by steroid therapy. This was the only case in which the outcome did not enable us to determine whether the patient had histologically atypical Takayasu's disease or Horton's disease at a young age. Group B included three patients (mean age 38 years, range 30-51 years). The roentgenologic and clinical signs consisted of a postvertebral subclavian lesion associated with declining general health in one case and an inflammatory syndrome in another. Takayasu's disease was revealed by subcla-

b

Fig. 2. Subclavian Involvement In Horton's disease. (a) Destruction of the medial elastic tissue Is limited to Internal elastic lamella (large arrow). The external medial area and adventitla are normal (small arrow heads). Lumen (L). Masson's trichrome stain (X 27). (b) Inflammatory Infiltrate of the internal elastic lamella with presence of giant cells and phenomenon of elastophagla (arrow). (Hematoxylin-eosln-safranin X 240).

154

ANNALS OF VASCULARSURGERY

SUBCLAVIAN ARTERY BIOPSY

vian histology and confirmed by the discovery of a tuberculous lymph node lateral to the subclavian artery in two cases and by extension of inflammatory arteritis in two cases, one patient manifesting both symptoms. There was no pre- or postoperative mortality or morbidity. Three vein grafts occluded early after the biopsy of a patent (1) or occluded (2) artery, but they did not require reoperation. One patient was lost to follow-up. Nine bypasses remained patent with a mean follow-up of 68 months (nine months to 16 years).

DISCUSSION Histopathologic examination of a biopsy specimen of the subclavian artery from 17 patients confirmed the roentgenologic and clinical diagnosis of Takayasu's disease in eight cases, led to its discovery in three cases, and eliminated this diagnosis in five cases. In one further case, neither the histopathologic nor the clinical findings were contributory and follow-up was inconclusive. The specificity of lesions in Takayasu's disease is a controversial topic. Predominant medial and adventitial involvement, both characteristic findings in Takayasu's disease, was first described by Nasu [6] and Sen [7]. In a recent review, Camilleri [13] discussed the presence of massive adventitial sclerosis which can penetrate the external media after destruction of the external elastic lamellae. The inflammatory cellular infiltrate found in the initial phase or during evolutionary phases of the disease also predominates at the medioadventitial junction. The histologic individuality of Takayasu's disease and Horton's disease had been commonly accepted [2,5,6] until Hall [9] contested it and declared that it was impossible to distinguish histologically between the two. Moreover, the specific distribution of arterial lesions and the coexistence of stenoses and aneurysms which are highly suggestive of Takayasu's disease have led Lande and associates [4,15,16] to propose total aortography as a diagnostic method. Bletry [12] emphasized the young age of patients, the absence of vascular risk factors, and the presence of inflammatory syndrome and proposed that Horton's disease characteristics might serve as exclusion criteria for diagnosis. Are roentgenographic and clinical criteria sufficient to suggest the diagnosis of Takayasu's disease and to distinguish it from Horton's disease? If not, can histological examination differentiate the two entities and if so, is there readily accessible anatomic material with which to do so? According to our study, use of roentgenographic and clinical scores led to three misinterpreted early cases of Takayasu's disease in patients with isolated subclavian lesions. As well, atypical findings of Horton's disease associated with isolated subcla-

vian lesions and negative temporal artery specimen histological findings led to the misdiagnosis of Takayasu's disease in three cases. On the other hand, the histologic diagnosis was in agreement with the final diagnosis, including follow-up findings in l l patients with Takayasu's disease and five of six patients with Horton's disease. This suggests that histologic criteria should be added to the diagnostic score of Takayasu's disease, as already proposed by Fiessinger [2]. In a recent study, (17) we have shown that it is possible to diagnose Horton's disease in a subclavian artery specimen when clinical signs of temporal arteritis are lacking, confirming our preliminary reports [2,11]. Intimal-medial lesions, numerous giant cells and infarctoid necroses are the hallmarks of Horton's disease [7,11,13], and they contrast with the peripheral sclerosis and medioadventitial predominance of lesions in Takayasu's disease. It is not only conceivable but necessary to certify the diagnosis of Takayasu's disease by pathology. This is especially true when the disease is in its earlier stages and arterial lesions are still isolated, or when the subject is elderly and one must eliminate atypical Horton's disease as a diagnosis. The surgical approach to the subclavian artery is easy. Long-term patency of revascularization procedures has never been compromised by the arterial retrieval in our experience. We therefore suggest retrieving a segment of subclavian artery for histopathology, as a valuable aid to correct diagnosis and proper therapy for Takayasu's disease.

REFERENCES 1. Committee report. Clinical and pathological studies of aortitis syndrome. Jap Heart J 1968;9:76-87. 2. FIESSINGER JN, TAWF|K-TAHER S, CAPRON L, et al. Maladie de Takayasu: crit~res diagnostiques. Nouv Press Med 1982;11:583-586. 3. LACOMBE P, FRIJA G, DUBOURG O, et al. Place de l'angiographie num6ris6e intra-veineuse dans la maladie de Takayasu: ~i propos de 44 cas. Arch Mal Coeur 1986; 3:273-280. 4. LANDE A, BERKMEN YM. Aortitis: pathologic, clinical and arteriographic review. Radiol Clin North Am 1976; 14:219-240. 5. LUPI-HERRERA E, SANCHEZ-TORRES G, MARCUSHAMER J, et al. Takayasu's arteritis: clinical study of 107 cases. A m Heart J 1977;93:94-103. 6. NASU T. Pathology of pulseless disease: a systemic study and critical review of twenty-one autopsy cases reported in Japan. Angiology 1963;14:225-242. 7. SEN K, KINARE JG, KELNAR MD, et al. Non-specific aorto-arteritis: a monograph based on a study of 101 cases. Bombay, New Delhi 1973. Tata McGraw Hill Publishing Company. 8. SHELHAMER JH, VOLKMAN DJ, PARRILLO JE et al. Takayasu's arteritis and its therapy. Ann Intern Med 1985; 103:121-126. 9. HALL S, BARR W, LIE JT, et al. Takayasu's arteritis: a

VOLUME 4

No 2 - 1990

10.

11.

12.

13.

SUB CLA VIA N A R TER Y BIOPS Y

study of 32 North American patients. Medicine 1985;64: 89-99. TEOH PC, TAN LKA, CHIA BL, et al. Non specific aorto-arteritis in Singapore with special reference to hypertension+ Am Heart J 1978;95:683-690. FIESSINGER JN, CAMILLIERI JP, CHOUSTERMAN M, et al. Maladie de Horton et maladie de Takayasu: crit~res anatomopathologiques. 2 observations. Nouv Press Med 1978;7:639-642. BLETRY O, K1EFFER E. THOMAS D. Le point rapide aujourd'hui sur l'art6rite de Takayasu. Nouv Press Med 1985;14:1311-1312. CAMtLLERI JP. BRUNEVAL P. Art~riopathies inflamma+ toires. In: CAMILLERI JP. BERRY CC, FIESSINGER JN,

I l l

14.

15. 16. 17.

155

et al (eds). Les Maladies de la Paroi Art~rielle. Paris, 1987, pp 426-461+ LIOZON F, CATANZANO G. L'art4rite temporale de Horton: etude anatomopathologique en microscopie optique. A p r o p o s de 123 biopsies temporales. Rev Med Int 1982 ;3:295-301. LANDE A, GROSS A. Total aortography in the diagnosis of Takayasu's arteritis. Am J Roentgenol 1972;116:165-178. LANDE A, ROSS1 P+ The value of total aortography in the diagnosis of Takayasu's arteritis. Radiology 1975;114:287297~ LONG A, BRUNEVAL P, LAURIAN C. Maladie de Horton a localisatio extra temporale: diagnostic parla biopsie sous clari~re+ 4 observation Arch Mal Coeur 1989: 82:1727-1731.