Carotid Steal: Report of Ten Cases

Transcription

1 Carotid Steal: Report of Ten Cases Süha Süreyya Özbek, MD, Ahmet Memiş, MD, Refik Killi, MD, Mir Ali Pourbagher, MD, Gülgün Demirpolat, MD, İsmail Oran, MD, Ayşin Pourbagher, MD The blood flow may be diverted from the external to the internal carotid artery via the carotid bulb in the absence of flow in the common carotid artery. We aimed to investigate the prevalence, hemodynamics, and clinical features of this condition. Reviewing the records of color duplex ultrasonographic examinations, we found carotid steal in 12 carotid arteries of 10 patients. In three patients flow in the ipsilateral external carotid artery was bidirectional. The steal had been demonstrated angiographically in six patients. The cause of the phenomenon was atherosclerosis, Takayasu arteritis, or trauma. Awareness of this pathway of collateral circulation may contribute to diagnostic work-up and expand management alternatives. KEY WORDS: Doppler ultrasonography; Angiography; Carotid arteries; Atherosclerosis; Takayasu arteritis. It is well known that brain is provided with alternative pathways for the maintenance of arterial perfusion after cerebrovascular occlusion. These collateral anastomoses may prevent or minimize serious neurovascular disease. The circle of Willis ABBREVIATIONS ECA, External carotid artery; ICA, Internal carotid artery; CCA, Common carotid artery; CDUS, Color duplex ultrasonography Received April 2, 1998, from the Department of Diagnostic Radiology, Ege University School of Medicine, İzmir, Turkey. Revised manuscript accepted for publication June 27, Address correspondence and reprint requests to S. Süreyya Özbek, MD, Ege University School of Medicine, Department of Diagnostic Radiology, Bornova, TR İzmir, Turkey. constitutes the main and best known pathway of collateral circulation. Unfortunately, developmental or acquired malfunction of this pathway may be present. In this case, or as a supplementary mechanism to the circle of Willis, other channels of collateral flow may contribute to the brain perfusion. One of these alternative collateral flow mechanisms is the diversion of ECA flow into the ipsilateral ICA in cases of occlusion in the CCA or innominate artery. This pattern of collateral flow, namely the carotid steal or ECA-to-ICA steal phenomenon, has been known for a long time. 1,2 It was not reported frequently and was hypothesized to occur only when the cross circulation via the anterior part of the circle of Willis is inadequate. 3 5 Until the 1980s the carotid steal phenomenon was reported more commonly in patients who had undergone therapeutic ligation of the CCA. 1,2,5 10 However, since then relevant series have mostly consisted of spontaneously occurring cases. 4, by the American Institute of Ultrasound in Medicine J Ultrasound Med 17: , /98/$3.50

2 624 CAROTID STEAL J Ultrasound Med 17: , 1998 The first purpose of this study was to assess the prevalence of the ECA-to-ICA steal in patients who had undergone CDUS in our center. A second purpose was to evaluate the hemodynamic features and accompanying clinical features in this subset of patients. MATERIALS AND METHODS The CDUS records of 3463 patients who had undergone examination of the extracranial segments of the carotid and vertebral arteries between August 1991 and March 1998 were reviewed. Diversion of flow from the ECA to the ICA was noted in 10 patients. The ages of the four female and six male patients ranged from 14 to 71 years (mean, 47 years) (Table 1). The CDUS examinations were performed using one of two Toshiba SSA-270A units (Toshiba Corporation, Tokyo, Japan). A 5.0 or 7.5 MHz lineararray transducer was used routinely with both spectral and color displays. Angiograms obtained concomitantly with the CDUS examinations were available for six of the patients. Five patients had angiographic examinations after the CDUS examinations (mean duration, 25 days). Another patient had undergone angiography 3 months prior to CDUS, which was performed without knowledge of angiographic results (patient 1). The angiographic examinations of the carotid and vertebral arteries were performed using a transfemoral approach and Seldinger technique. Cut-film angiography was obtained in two patients. Three patients underwent cine-angiography, whereas intraarterial digital subtraction angiography was available for one patient. During the diagnostic work-up of these patients, the final clinical diagnosis of the cause of the steal was based on clinical history and radiologic and clinical findings. RESULTS In this series 10 patients with carotid steal phenomenon represented a prevalence of 0.3% among 3463 patients referred for CDUS examination of the carotid and vertebral arteries. The clinical history of each patient is presented in Table 1. Symptoms and clinical findings indicative of cerebrovascular disease were the most common finding in the clinical history (six patients). The patient who had been referred after receiving a knife wound was the only patient with carotid steal in this series whose CCA was ligated therapeutically. Four of the 10 patients were outpatients. None of the patients was unconscious during the radiologic examinations. CDUS examinations revealed six right-sided, two left-sided, and two bilateral cases of carotid steal in 10 patients (Fig. 1). The CDUS features of the patients are summarized in Table 2. The ipsilateral CCA was occluded totally or subtotally in all except one case Table 1: Patient and Clinical Data Patient Age (yr)/ Clinical Duration* Final Cinical Number Sex Status History Angiography (days) Diagnosis 1 40/F Outpatient Right hemiparesis, + 90 Takayasu arteritis no pulse on both arms 2 14/M Inpatient Knife wound lacerating Therapeutic right CCA CCA ligation 3 53/M Outpatient Neurologic stroke Atherosclerosis 4 58/M Outpatient TIA, no pulse on right arm + 34 Atherosclerosis 5 51/M Inpatient Gunshot wound in the right Gunshot wound cervical region 6 38/F Inpatient Left hemiplegia, no pulse + 12 Takayasu arteritis on left arm 7 57/F Outpatient Weak pulse on left arm Takayasu arteritis 8 49/M Inpatient Neurologic stroke + 36 Atherosclerosis 9 39/M Inpatient Left hemiparesis + 20 Atherosclerosis 10 71/F Inpatient Coronary heart disease + 25 Atherosclerosis *Duration between CDUS and angiography. Final diagnosis of the pathologic cause of carotid steal. CDUS was performed after angiography, but without knowledge of its findings. TIA, Transient ischemic attack.

3 J Ultrasound Med 17: , 1998 ÖZBEK ET AL 625 (patient 4). In this patient the innominate artery was occluded, which caused partial reversal of flow in the right CCA in the systolic phase. CDUS showed antegrade flow in the distal and retrograde flow in the proximal segment of ECA in three patients (patients 7 to 9). The flow supplied by a caudally directed branch of the proximal ECA diverted into two opposite directions at the level of its origin. This artery with reversal of flow was presumed to be the superior thyroid artery (Fig. 2). In one patient (patient 4), CDUS demonstrated partial reversal of flow during systole in the left vertebral artery. Partially retrograde flow progressed to complete reversal of flow after a provocative maneuver performed as described earlier. 17 Angiography confirmed this finding and delineated subclavian arterial stenosis in the prevertebral segment of this artery. CDUS also showed partial reversal of blood flow during systole in the right CCA in the same patient. In two other patients (patients 7 and 10), partially retrograde flow in systolic phase was demonstrated in the ICA during the cardiac cycle. Angiographic examination showed the ECA-to- ICA steal in all of six patients who underwent this procedure. The angiographic findings are presented in Table 3. In four patients the collateral flow to the ECA could be evaluated in detail. In all of the patients ipsilateral occipitovertebral anastomoses were delineated angiographically. In one case (patient 6), angiography also showed thyrocervical anastomosis. Bidirectional ECA flow that was demonstrated sonographically could not be shown with angiography. In five of the 10 patients the carotid steal was caused essentially by atherosclerosis. Vasculitis, more precisely Takayasu arteritis, was considered to be the causative factor of the ECA-to-ICA steal in three female patients. Figure 1 Patient 6. CDUS and angiographic findings in a patient with bilateral carotid steal. A, Reversed flow in the right ECA. B, Antegrade and damped flow in the right ICA, supplied by the flow in the ECA. The CCA is completely occluded except for the bulbar segment. C, Left subclavian arteriogram taken in left lateral projection delineates patency of the left ICA and ECA in the presence of a CCA occluded except in its most distal segment. Occipitovertebral anastomosis is evident. A B DISCUSSION C In our study, we diagnosed 10 patients with the carotid steal, which represented a prevalence of 0.3%. This rate is in accordance with that of the study by Morvay and coworkers, 12 who found carotid steal in 0.1% of CDUS examinations performed in 3759 patients. In contradiction to findings of series published up until the 1980s, the ECA-to-ICA steal had developed spontaneously in the large majority of patients in our series. Only one in 10 patients had undergone therapeutic ligation of CCA. This trend also is apparent in more recently reported cases in the literature. It may be partly attributed to the omission of CCA ligation as a neurosurgical procedure in

4 626 CAROTID STEAL J Ultrasound Med 17: , 1998 Table 2: CDUS Findings Patient Right Left Number CCA ICA ECA VA CCA ICA ECA VA 1 Occluded A (CS) R (CS) A Occluded?? A 2 Subtotally A (CS) R (CS) A A A A A occluded 3 Occluded A (CS) R (CS) A Partially Occluded A A thrombosed 4 PR A (CS) R (CS)? A Occluded A PR 5 Subtotally A (CS) R (CS) A A A A A occluded 6 Occluded A (CS) R (CS) A Subtotally A (CS) R (CS) A occluded 7 A A A A Occluded PR (CS) A+R (CS) A 8 Occluded A (CS) A+R (CS) A A A Occluded A 9 A A A A Occluded A (CS) A+R (CS) A 10 Occluded PR (CS) R (CS) A Occluded A (CS) R (CS) A VA, Vertebral artery; A, antegrade flow; R, retrograde flow; CS, carotid steal;?, unsuccessful demonstration; PR, partially retrograde flow. the recent past. Another reason for this may be the widespread availability of Doppler ultrasonography in recent decades, which has made the diagnosis of asymptomatic hemodynamic disorders possible. The present study consists of one of the largest groups of patients demonstrated with CDUS. In this series, the patient with a patent right CCA and carotid steal due to occlusion of the innominate artery is a unique case. To our best of knowledge, only few patients with similar pathologic findings have been reported in the literature. 16 In cases of CCA occlusion, another pattern of carotid steal, the diversion of blood flow from the ICA to the ECA, also has been reported in relatively few cases. 5,11,18 However, we have not encountered that type of steal in our CDUS or angiography practice. Although Sutton and Davies 2 reported atherosclerosis as a rare cause in the pre-ultrasonography era, the carotid steal occurred basically as a result of atheromatous arterial occlusion in five of the 10 patients in the present series. In one of them the ECAto-ICA steal was diagnosed incidentally during the investigation for coronary heart disease. Two of the remaining four subjects were outpatients with minor symptoms. These aspects of the clinical data are indicative of the etiopathologic role of atherosclerosis in a larger number of patients than was thought previously. A relatively rare entity, Takayasu arteritis, was concluded to cause the steal phenomenon in three female patients. Although occlusion of the brachiocephalic arteries is a common phenomenon in Takayasu arteritis, our review of literature did not reveal another case in which it resulted in carotid steal. In one of two cases of trauma, the ECA-to-ICA steal occurred spontaneously, as a result of thrombosis in the CCA. In the other patient, the steal resulted from therapeutic ligation of the proximal CCA. A review of the data in this series shows that the ECA-to-ICA steal effectively contributes to the collateral circulation to the brain. In six of the 10 patients, symptoms and clinical findings associated with cerebrovascular insufficiency were predominant. In the remaining cases, only the symptoms and signs resulting from deficient blood supply to upper extremity or local findings in cases of trauma were present. Even in one of the two patients with bilateral carotid steal, predominant clinical findings and history were indicative of cardiovascular disease. Rossi and colleagues 7 described occipitovertebral, occipital deep cervical, and carotid-basilar anastomoses as the major collateral channels involved in cases of carotid steal. In anatomic studies it was shown that communications between two ECAs occur mainly through their maxillary branches via widespread anastomoses in the nose and palate, and also by the anastomoses between occipital arteries, superficial temporal arteries, and the thyroid arterial plexus. 3 Of these alternative channels, occipitovertebral anastomosis is the one reported most commonly in studies based on angiographic findings. 2,7,8,19 Rossi and coauthors 7 reported the presence of occipito-

5 J Ultrasound Med 17: , 1998 ÖZBEK ET AL 627 vertebral anastomosis in every case in their series. Likewise, we demonstrated this anastomosis angiographically in all of four patients in whom collateral circulation could be evaluated in detail. Angiographic examinations delineated thyrocervical anastomoses in two patients. In one of these patients, and in two others, CDUS demonstrated another finding: bidirectional flow in the ECA on the side of the carotid steal. According to the Doppler sonographic findings, ECA flow was antegrade in the segment distal to the origin of a proximal branch, presumably the superior thyroid artery, whereas the flow was retrograde proximal to that point. This hemodynamic information suggested that the superior thyroid artery was one of the main suppliers to the ECA and ICA, and the main one to the extracranial segment of ICA in these cases. We do not have objective data to explain the bidirectional flow phenomenon in the ECA. However, we can speculate that the resistances exerted on the incoming sidebranch flow at its origin by the vascular territories of the distal ECA and ICA (via the proximal ECA and carotid bulb) were in equilibrium in these cases. The equilibrium might have been lowered resistance of the distal ECA system (owing to reasons like ECA-to- ICA collateralization) with or without increased resistance of the ICA system (due to intracranial or extracranial steno-occlusive pathology) on incoming flow. Angiographic examination, which was performed in two of these three patients, did not show this pattern of collateral flow. This failure may be partly attributed to the relative insensitivity of angiography to slow flow and hemodynamic changes distal to an occluded artery. 4,14 Another reason for the inability to show the pattern of collateral flow may be the fact that angiography causes physiologic alterations, which might have prevailed or changed the hemodynamic status in these patients. 19 Unlike angiography, CDUS provides the ability to investigate hemodynamics in completely physiologic circumstances. Other noninvasive modalities, such as magnetic resonance angiography, also show promise in noninvasive evaluation of the cerebrovascular system. Angiography supplies more information on anatomic details, whereas CDUS demonstrates hemodynamics more precisely. The two modalities seem to be complementary to one another rather than being rivals. Demonstrated with either one of them, carotid steal appears as a unique and clinically important phenomenon. Awareness of this pathway of collateral circulation in a given patient may contribute greatly to diagnostic work-up and may avoid possible complications during operations in the cervical and orofacial regions. Figure 2 Patient 8. CDUS images demonstrating bidirectional flow in the ECA in a patient with carotid steal. A, Antegrade flow in the distal segment of the right ECA. B, The flow is reversed in the proximal segment of the same artery. The image also shows complete occlusion of the CCA except at the carotid bulb. C, Forward flow is maintained in the right ICA. A B C

6 628 CAROTID STEAL J Ultrasound Med 17: , 1998 Table 3: Angiographic Findings Demonstrated Patient Angiography Collaterals Additional No. Type Right Left to ECA Findings CCA ICA ECA VA CCA ICA ECA VA 1 Cut film Occluded A (CS) R (CS) N Occluded Occluded Occluded N? Bilateral SCA occlusions 4 Cut film PR A (CS) R (CS)? N Occluded N PR? Innominate artery occluded at its origin; left SCA stenosis (prevertebral); left VA stenosis (origin) 6 DSA Occluded A (CS) R (CS) N Occluded A (CS) R (CS) N Bilateral occipito- Left SCA stenosis (postvertevertebral, right bral); both hemispheres mainly thyrocervical perfused by both VA via posterior communicating arteries 8 Cine- Occluded A (CS) R (CS) N N N Occluded N Right occipito- Right ACA territory perfused angiography vertebral by left carotid system, but that of right MCA is not 9 Cine- N N N N Occluded A (CS) R (CS) N Left occipitoangiography vertebral and thyrocervical 10 Cine- Occluded A (CS) R (CS) N Occluded A (CS) R (CS) N Left occipito- Right VA is dominant angiography vertebral and supplies right MCA and both ACA; left MCA perfused mainly by left carotid steal VA, Vertebral artery; SCA, subclavian artery; MCA, middle cerebral artery; ACA, anterior cerebral artery; DSA, digital subtraction angiography; A, antegrade flow; R, retrograde flow; CS, carotid steal;?, unsuccessful demonstration; PR, partially retrograde flow; N, normal or nonsignificant lesions.

7 J Ultrasound Med 17: , 1998 ÖZBEK ET AL 629 Furthermore, demonstration of a patent ICA in the presence of ipsilateral CCA occlusion may make surgical endarterectomy or bypass procedure in selected patients beneficial. REFERENCES 1. Dorrance GM: Ligation of the great vessels of the neck. Ann Surg 99:721, Sutton D, Davies ER: Arch aortography and cerebrovascular insufficiency. Clin Radiol 17:330, Hawkins TD: The collateral anastomoses in cerebrovascular occlusion. Clin Radiol 17:203, Dashefsky SM, Cooperberg PL, Harrison PB, et al: Total occlusion of the common carotid artery with patent internal carotid artery: Identification with color flow Doppler imaging. J Ultrasound Med 10:417, Tindall GT, Odom GL, Dillon ML, et al: Direction of blood flow in the internal and external carotid arteries following occlusion of the ipsilateral common carotid artery: Observation in 19 patients. J Neurosurg 20:985, Sweet WH, Stanley JS, Bakay L: A clinical method for recording internal carotid pressure: Significance of changes during carotid occlusion. Surg Gynecol Obstet 90:327, Rossi P, Rosenbaum AE, Zingesser LH: The fate of the carotid artery after occlusion for treatment of aneurysm. Radiology 95:567, Wolpert SM: The thyroidea ima artery: An unusual collateral vessel. Radiology 92:335, Gossman HH, Gryspeerdt GL, Tomlinson BE: Development of collateral arteries at the site of carotid ligation. Br J Surg 52:634, Somach FM, Shenkin HA: Angiographic end-results of carotid ligation in the treatment of carotid aneurysm. J Neurosurg 24:966, Grant EG, Wong W, Tessler F, et al: Cerebrovascular ultrasound imaging. Radiol Clin North Am 26:1111, Morvay Z, Milassin P, Barzó P: Assessment of steal syndromes with color and pulsed Doppler imaging. Eur Radiol 5:359, Sidhu PS, Morarji Y: Case report: A variant of the subclavian steal syndrome: Demonstration by duplex Doppler imaging. Clin Radiol 50:420, Keller HM, Valavanis A, Imhof H-G, et al: Patency of external and internal carotid artery in the presence of an occluded common carotid artery: Noninvasive evaluation with combined cerebrovascular Doppler examination and sequential computer tomography. Stroke 15:149, Blackshear WM Jr, Phillips DJ, Bodily KC, et al: Ultrasonic demonstration of external and internal carotid patency with common carotid occlusion: A preliminary report. Stroke 11:249, Barnett HJM, Gladstone RM: Diversion and reversal of cerebral blood flow: External carotid artery steal. Neurology 20:1, Schneider PA, Rossman ME, Bernstein EF, et al: Noninvasive evaluation of vertebrobasilar insufficiency. J Ultrasound Med 10:373, Hardesty WH, Roberts B, Toole JF, et al: Studies on carotid artery. Surgery 49:251, Hessel SJ, Rosenbaum AE: True and false external carotid steals. Clin Radiol 25:303, 1974