Evaluation of Cross-Circulation Through Circle of Willis Using An Ultrasonic Doppler Technique

Transcription

1 478 Evaluation Cross-Circulation Through Circle Willis Using An Ultrasonic Doppler Technique Part I. Comparison Between Blood Flow Velocity by Ultrasonic Doppler Flowmetry and Angiogram SHOTARO YONEDA, M.D., TADAATSU NUKADA, M.D., KAZUFUMI KIMURA, M.D., KENICHI TANAKA, M.D., KEIICHI ASHIDA, M.D., TSUTOMU ASAI, M.D., HIDEKI ETANI, M.D., MASATOSHI IMAIZUMI, M.D., AND HIROSHI ABE, M.D. SUMMARY In 24 patients with cerebrovascular disease and 6 without organic brain lesions, the increased Telocity blood flow in both the internal carotid and rertebral arteries daring a contralateral carotid compression was compared with the angiographlc appearance the drde Willis. The flow vdodty was measured using ultrasonic Doppler flowmetry. It was not possible to investigate fully the relationship between the increase velocity blood flow in the internal carotid artery and the anatomical variations the circle Willis, specifically the anterior cerebral and communicating arteries. The velocity flow in the patients with an aplastic proximal portion the anterior cerebral artery showed no increase. There were no differences in the increase velocity flow in the vertebral artery in patients with hypoplastk, normal and fetal posterior communicating arteries. It is considered that although anatomical variations the drde vessels influence the cross-drculation via the drde Willis, peripheral vascular factors distal to the drde also play an Important role in the quantity cross-circulation through the drde. Stroke,Voll2,No4,1981 VARIOUS METHODS 1-8 are used for assessing cerebral cross-circulation through the circle Willis, including the ultrasonic Doppler techniques. e " 1 ' A new method for evaluating both the efficiency and adequacy cross-circulation through the circle Willis has been presented," and this revealed that the blood flow increase in the internal carotid and vertebral arteries, as measured by means ultrasonic Doppler flowmetry during a contralateral carotid compression, was related to the functional capacity the cross-circulation via the circle Willis. Also, it was suggested that this functional capacity might be altered by atherosclerotic lesions in the cervical and cerebral arteries and be related to an increase cerebral vascular resistance. Anatomical 18 " 20 variations in the circle Willis are observed with high frequency. The relationship between the anatomical variations in the circle and the efficiency cross-circulation has been reviewed in previous studies. 1 ' 19> " This study was performed to evaluate the significance angiographically demonstrated variations the circle Willis and cross-circulation using ultrasonic Doppler flowmetry. From the First Department Medicine, Osaka University Medical School, Fukushima, 1 chome, Fulcushima-ku, Osaka 553, (Drs. Yoneda, Kimura, Ashida, Asai, Etani, Imaizumi and Abe), and the National Osaka South Hospital, 677-2, KJdo-cho, Kawachi- Nagano city, Osaka 586, Japan (Drs. Nukada, Tanaka). Reprints: Dr. Shotaro Yoneda, First Dept. Medicine, Osaka Univ. Medical School, Fukushima, 1 chome, Fukushima-ku, Osaka 553, Japan. Materials and Methods This study included 24 patients with chronic ischemic cerebrovascular disease (CBVD) due to carotid system disease and 6 without organic brain lesions (normal brain). The 30 patients consisted 28 men and 2 women with an age range from 15 to 70 years (average age 50.0 ± 12.3 SD). Among the 24 patients with cerebrovascular disease there were 17 with completed stroke and 7 with transient ischemic attack. Carotid angiography was unilateral in 14 patients and bilateral in 3. A unilateral carotid and vertebral angiogram was done on 2 patients. Bilateral carotid and vertebral angiograms were done in the other 8 patients. No angiography was carried out in 2 patients with "normal brain." Patients who had moderate or severe stenotic and occlusive lesions in both the carotid and vertebral arteries, and also on the anterior, middle and posterior cerebral arteries were all excluded. Angiographic findings in 24 patients with cerebrovascular disease are presented in table 1. Angiography Positive cross-circulation was believed present when in the antero-posterior projection the angiogram branches the contralateral carotid artery filled; it was believed to be absent when contralateral branches were not shown. The proximal segment the anterior cerebral

2 ULTRASONIC DOPPLER FOR CROSS CIRCULATION. I. ANGIOGRAM/Yoneda et al. 479 TABLE 1 Angiographic Findings Patients with Cerebrovoscular Disease Unilateral CAG (n = 12) Wall irregularity or mild stenosis IC A 2 Wall irregularity ACA &MCA 3 Wall irregularity or mild stenosis ICA, ACA & MCA 2 Wall irregularity ICA & MCA 1 Branch occlusion MCA 1 No evidence vascular disease 3 Bilateral CAG (n = 3) Wall irregularity or mild stenosis bilateral ICA 2 No evidence vascular disease 1 Unilateral CAG & VAG (n = 1) Wall irregularity ICA and kinking VA 1 Bilateral CAG & VAG (n = 8) Wall irregularity unilateral ICA 2 Wall irregularity bilateral ICA 1 Wall irregularity peripheral branch unilateral MCA and VA 1 Wall irregularity unilateral ICA, MCA & ACA and stenosis branch PC A 1 No evidence vascular disease 3 ICA; Internal Carotid Artery MCA: Middle Cerebral Artery PCA: Posterior Cerebral Artery ACA: Anterior Cerebral Artery VA: Vertebral Artery artery was judged to be hypoplastic when this vessel was less than 1 mm in diameter.* 1 The caliber the posterior communicating artery was determined on a lateral projection the angiogram by the ratio the diameter that artery to the diameter the posterior cerebral artery. It was said to be hypoplastic when the caliber the posterior communicating artery was less than 25 percent that the posterior cerebral artery or when the posterior communicating artery was revealed as a string-like vessel by carotid angiogram. The posterior communicating artery was labeled fetal when its caliber was either equal to or approached that either the anterior or the middle cerebral artery.* 1 Doppler Flowmetry Apparatus and Basis Examination Technique The blood flow velocity was measured by a directional Doppler flowmeter (Hitachi EUD-3B, 5MHz, Hitachi Medical Corporation, Tokyo). All examinations were performed with the patient supine. The technique for internal carotid detection was as follows: The probe was directed to the internal carotid artery and was positioned so as to obtain a maximum audible audio signal from a speaker. To be sure that the signal came from the internal carotid artery the contralateral carotid artery was compressed. This resulted in an increase in the flow velocity signal. The technique for vertebral flow detection is as follows: The patient's head was turned away from the side the examination. The probe was placed beneath the posterior-inferior margin the mastoid process and the ultrasonic beam directed medially and slightly superiorly. To determine that the signal came from the vertebral artery the ipsilateral carotid artery was compressed. This resulted in no change or consequential increase in the flow velocity signal. The signal from the external carotid artery was abolished by this maneuver. The blood flow velocity (mean velocity) curve obtained from the directional Doppler flowmeter was integrated every 3 seconds. This value was employed so as to simplify the calculation blood flow velocity increase during a carotid compression. The blood flow velocity, the electrocardiogram, and the ear plethysmogram were all recorded simultaneously on a polygraph (Hitachi ECP-1, Hitachi Medical Corporation, Tokyo) (fig. 1). Examination Procedure Cross-circulation via the circle Willis was evaluated from the flow velocity changes in the internal carotid and vertebral arteries by an ultrasonic Doppler flowmeter. The changes in flow velocity in both internal carotid and vertebral arteries were measured before and during digital compression the common carotid artery (fig. 2). The changes in blood flow velocity were expressed as percent change from the bloodflowvelocity in the steady state. The flow change in the internal carotid artery was measured during a compression the contralateral common carotid artery which redistributes blood flow in the internal carotid artery to the contralateral cerebral hemisphere through the anterior communicating artery. Flow change in the vertebral artery was measured during a compression the ipsilateral common carotid artery which redistributes blood flow from the vertebral artery to the ipsilateral cerebral hemisphere the carotid system through the posterior communicating artery. Complete compression the common carotid artery was confirmed by a plateau wave in the ipsilateral ear plethysmogram. Patients who had a significant change in blood pressure and heart rate during the carotid compression were excluded. Results 1. Percent Increase in Blood Flow Velocity The percent increase in the velocity bloodflowin the internal carotid artery was measured on 11 arteries FIGURE 1. Schematic layout Doppler ultrasonic system employed in this study.

3 480 STROKE VOL 12, No 4, JULY-AUGUST 1981 compression compression E.C.G. rt-ear Plethysmogram Doppler Signal It-Internal Carotid Artery Integrated Doppler Signals \l\l\ \ F(500 Hz)dt FIGURE 2. A polygraphic record monitoring electrocardiogram (ECG), ear plethysmogram, Doppler signal from the internal carotid artery, and its integrated value. This figure shows that a digital compression the right carotid artery produced an increase in both Doppler signal and its integrated value and a reduction wave amplitude the right ear plethysmogram. 10 sec. from 6 patients with normal brain function and on 45 arteries from 23 patients with CBVD (fig. 3). The percent increase the internal carotid blood flow velocity was 36.7 ± 0.8 (SD) % in 6 patients with normal brain function and 22.1 ± 12.7 (SD) % in 23 with CBVD (table 2). The difference is significant (p < 0.005). The mean ages in the normal brain function group and in the CBVD group were respectively 44.0 ±16.2 (SD) and 51.3 ± 11.2 (SD) years. There was no significant difference in the mean ages between these 2 groups. The percent increase the vertebral blood flow velocity was measured on 9 arteries from 5 patients with normal brain function and on 42 arteries in 23 patients with CBVD (fig. 3). The percent increase the velocity blood flow in the vertebral artery in patients with normal brain function and CBVD was respectively 41.6 ± 16.0 (SD) and 18.9 ± 12.0 (SD) %, the difference being highly significant (p < 0.001). The mean ages in the normal brain function group and in the CBVD group were respectively 45.6 ± 17.6 (SD) and 52.4 ± 10.3 (SD) years. There was no significant difference in the mean (table 2). COMPRESSION FIGURE 3. The blood flow velocity increase the contralateral carotid and ipsilateral vertebral arteries during a digital compression the carotid artery. The black shaded arteries represents the vessels on which measurement the flow velocity was performed. The dotted artery is the collateral channel. The large dotted arrow represents the velocity increase during a carotid compression. The small dotted arrow shows the flow direction through the circle Willis. 2. Flow Velocity Increase in Internal Carotid Artery There was spontaneous angiographic cross-circulation through the anterior communicating artery observed in 27 out 28 patients who had angiograms. Bilateral anterior cerebral arteries were found in 20 patients. The relation between the percent increase in velocity flow in the internal carotid artery and the findings from angiograms was examined in those 20 patients. Thirty-four internal carotid arteries in 17 patients had a bilaterally normal proximal segment in their anterior cerebral artery (normal ACA) and 4 internal carotid arteries in 2 patients had a unilateral hypoplastic proximal segment in the anterior cerebral artery (hypoplastic ACA). The percent increase the velocity blood flow in the internal carotid was 24.0 ± 10.8 (SD) and 20.9 ± 15.9 (SD) % (figs. 4, 5). The patients with a normal ACA had higher mean values than those with a hypoplastic ACA. Because the paucity patients with hypoplastic ACA it was not possible to draw firm conclusions concerning the significance. Flow studies from bilateral internal carotid arteries in one patient with a unilateral aplastic proximal segment the anterior cerebral artery (aplastic ACA), showed no increase the velocity blood flow in the internal carotid artery. 3. Flow Velocity Increase in Vertebral Artery Thirty-five posterior communicating arteries 23 patients were indirectly evaluated by study the flow velocity increase in the ipsilateral vertebral artery during ipsilateral carotid compression. Fifteen arteries in

4 ULTRASONIC DOPPLER FOR CROSS CIRCULATION. I. ANGIOGRAM/Yoneda et al. 481 TABLE 2 Percent Increase Cerebrouascular Disease in Blood Flow Velocity in Internal Carotid and Vertebral Arteries in Normal Brain and in patients Normal Brain 6 CBVD 23 arteries Internal carotid artery Age (years) 44.0 ± ±11.2 Percent increase in flow velocity 36.7 ± 13.8* 21.4 ±12.9 patients 5 23 arteries 9 42 Vertebral artery Age (years) 45.6 ± ± 10.3 Percent increase in flow velocity 41.6 ± 16.0** 18.9 ± 12.0 *: Significant difference between normal brain and CVD at p < **: Significant difference between normal brain and CVD at p < CBVD Cerebrovascular disease. these 11 patients with hypoplastic posterior communicating arteries (hypoplastic PCA) and 14 arteries in those 12 patients with normal posterior communicating arteries (normal PCA) during an ipsilatcral carotid compression had a vertebral blood flow velocity increase 20.9 ± 16.0 (SD) and 24.0 ± 10.8 (SD) %, respectively (figs. 6, 7). There was no significant difference between the hypoplastic and the normal PCA groups. A fetal posterior communicating artery (fetal PCA) was observed in 6 arteries among 4 patients. The mean percent increase their vertebral flow velocity was 21.7%, within a distribution 8 to 61%. There was no difference in percent in- crease when compared to the other 2 groups. In patients with a fei&l PCA and carotid compression one had high increase and 4 low increases. Discussion A variety methods electroencephalographic monitoring, 1 "* measurement carotid artery back "3 5 o COMPRESSION m o 40- (0 O To I 30- </> 20 n o> u g 10 2 Q. I I 1 FIGURE 4. The blood flow velocity increase in the internal carotid artery during digital compression the contralateral common carotid artery. The black shaded artery represents the vessel on which measurement the flow velocity was performed. The dotted artery is the collateral channel. The large dotted arrow represents the velocity increase during a carotid compression. The small dotted arrow shows the flow direction through the circle Willis. NORMAL ACA (n=34) HYPOPLASTIC ACA (n=4) FIGURE 5. Percent increase the velocity flow in the internal carotid artery the patients with a normal anterior cerebral artery (n = 34) and with a hypoplastic anterior cerebral artery (n = 4). The values in patients these 2 groups was 24.0 ± 10.8 (SD) and 20.3 ± 13.9 (SD) %, respectively.

5 482 STROKE VOL 12, No 4, JULY-AUGUST /e J 50- I COMPRESSION ioh FIGURE 6. The blood flow velocity increase in the vertebral artery during digital compression the ipsilateral common carotid artery. The black shaded artery represents the vessel on which measurement the flow velocity was performed. The dotted artery is the collateral channel. The large dotted arrow represents the velocity increase during a carotid compression. The small dotted arrow shows the flow direction through the circle Willis. p pressure *- Ti ' ** jg jugular venous blood gases measurel cerebral bloodflow*"*- d * ' * are ment,' 1 and regional being used to assess the potential cerebral collateral circulation. In our previous report, 14 a non-invasive method for evaluating both the efficiency and the adequacy collateral circulation via the circle Willis by the ultrasonic Doppler technique was presented. In the present study, it was found that the percent increase in flow velocity in both the carotid and vertebral arteries in patients without an organic brain lesion during digital carotid compression was significantly higher than that those patients with CBVD. Spontaneous angiographic cross-circulation through the anterior communicating artery was observed in 27 out 28 patients. This ratio is higher than that in previous reports.* 1 " It is known that several technical factors influence the distribution radiographic contrast medium injected into the carotid artery. These are related to both the pressure and the volume injection, the arterial pressure, and the point the arterial pulse cycle at the time the injection the contrast medium. u> * The demonstration cross-circulation through the anterior communicating artery on angiography is not a particu- o J hypo plastic PCA (n=16) normal PCA (n = 13) fetal PCA (n=5) FIGURE 7. Percent increase velocity flowin the vertebral ariery in the patients with hypoplastic (n = 15), normal (n = 14) and fetal (n = 6) posterior communicating arteries. The values in the patients the three groups were 20.9 ± 16.0 (SD), 24.0 ± 10.8 (SD) and 21.7 ± 19.9 (SD) %, respectively. larly reliable guide to the tolerance unilateral carotid ligation.' 1 " " It is not possible to visualize the anterior communicating artery on the antero-posterior angiogram projection and absence the anterior communicating artery is seen only rarely." All but one the patients studied who had cerebral angiograms had cross-filling the anterior cerebral arteries. Based on this finding it was concluded that there was no aplasia the anterior communicating artery. As to the one exceptional patient, the percent increase the internal carotid blood flow velocity was 39 and 23% and was not necessarily a low increase. It has been reported in a previous study 1 ' that a hypoplastic ACA was accompanied with poor crossfilling through the circle Willis. The percent increase in the velocity blood flow in the carotid artery those 2 patients with a hypoplastic ACA was 6 to 35%, and in 2 arteries it was within normal limits. The number patients with the hypoplastic ACA studied is small and it is difficult to assess the significance the changes. No flow increase in the internal carotid artery following carotid compression in the contralateral side was observed in patients with an aplastic proximal segment the anterior communicating artery. This suggests that in patients with no anatomical connection between the internal carotid arteries via the anterior communicating artery, con-

6 ULTRASONIC DOPPLER FOR CROSS CIRCULATION. I. ANGIOGRAM/Yoneda et al. 483 tralateral carotid compression is not accompanied by increased collateral circulation. The flow velocity the patients with a normal ACA had a wide range increase from 0 to 61% with carotid compression. This result may suggest that the peripheral vascular factors distal to the circle Willis play a role in determining the difference in increased velocity. The percent increase in flow velocity in the vertebral artery in patients with hypoplastic, normal, and fetal PCA were 20.9 ± 15.9 (SD), 24.0 ± 10.8 (SD) and 21.7 ± 19.9 (SD) % respectively. The differences between these groups were not significant and there were no differences in the angiographic variations the PCA. Levy et al. 17 reported that vasodilation the circle Willis occurred immediately after a bilateral carotid occlusion. It is suspected that the PCA may possibly dilate during carotid compression, making it difficult to clarify the differences in the percent increase the blood flow and the angiographic variations the artery. Beatty and Richardson 1 reported that the presence a bilateral fetal PCA was an indicator poor tolerance for carotid ligation and that the presence a unilateral fetal PCA was not associated with ischemia. Jawad et al. 8 stated that most patients with this anomaly satisfactorily tolerated carotid ligation without showing a significant increase in the incidence cerebral ischemia following the ligation. In our study, one patient with a unilateral fetal PCA had a high flow velocity increase and 2 others with a bilateral fetal PCA had a low increase. There was no difference in the percent increase in flow velocity between patients with a fetal PCA or the other patients. A pressure drop probably less than 30 mm Hg occurs between the aorta and the circle Willis and its large distributing branches. There is a pressure drop 3 to 4 times this magnitude in the vascular bed distal to the circle Willis 18 and only a small part the total vascular resistance comes from the circle Willis itself. 19 Jawad et al.' demonstrated that most resistance in cerebral vessels, which determines collateral supply in the brain, lies distal to the circle Willis. This study indicates that although anatomical variations the circle Willis influence crosscirculation through the circle, a peripheral vascular factor distal to the circle Willis also plays an important role. References 1. Beatty RA, Richardson AE: Predicting intolerance to common carotid artery ligation by carotid angiography. J Neurosurg 28: 9-13, Sundt TM Jr, Sharbrough FW, Anderson RE, Michenfelder JD: Cerebral blood flow measurements and electroencephalograms during carotid endarterectomy. J Neurosurg 41: , Trojaborg W, Boysen G: Relation between EEG, regional cerebral blood flow and internal carotid artery pressure during carotid endarterectomy. Electroencephal Clin Neurophysiol 34: 61-69, Sharbrough FW, Messick JM, Sundt TM Jr: Correlation continuous electroencephalograms with cerebral blood flow measurements during carotid endarterectomy. Stroke 4: , Waltz AG, Sundt TM Jr, Michenfelder JD: Cerebral blood flow, jugular venous PO a and lactate concentration, and arterial-venous oxygen content during carotid endarterectomy. Europ Neurol 6: , 1971/72 6. Galbraith JG: Monitoring brain oxygenation during carotid ligation. J Neurosurg 27: , Boysen G, Engell HC, Ladegaard-Pedersen HJ, Henrik»en H: Effect induced hypertension on internal carotid artery stump pressure and rcbf during carotid surgery. Europ Neurol 6: , 1972/72 8. Jawad K, Miller JD, Wyper DJ, Rowan JD: Measurement CBF and carotid artery pressure compared with cerebral angiography in assessing collateral blood supply after carotid ligation. J Neurosurg 46: , Keller H, Meier W, Yonekawa Y, Kumpe D: Noninvasive angiography for the diagnosis carotid artery disease using Doppler ultrasound (carotid artery Doppler). Stroke 7: , Yoneda S, Nukada T, Tada K, Imaizumi M, Takano T, Abe H: Subclavian steal in Takayasu's arteritis. A hemodynamic study by means ultrasonic Doppler flowmetry. Stroke 8: , Hodek-Demarin V, Mailer HR: Reversed ophthalmic artery flow in internal carotid artery occlusion. A re-appraisal based on ultrasonic Doppler investigations. Stroke 10: , Blackshear WM Jr, Phillips DJ, Chikos PM, Harley JD, Thiele BL, Strandness DE Jr: Carotid artery velocity patterns in normal and stenotic vessels. Stroke 11: 67-71, Spencer MP, Reid JM: Quantitation carotid stenosis with continuous-wave (C-W) Doppler ultrasound. Stroke 10: , Tada K, Nukada T, Yoneda S, Kuriyama Y, Abe H: Assessment the capacity cerebral collateral circulation using ultrasonic Doppler technique. J Neurol Neurosurg Psychiatry 38: , Alpers BJ, Berry RG, Paddison RM: Anatomical studies the circle Willis in normal man. Arch Neurol Psychiatry 81: 409^118, Fisher CM: The circle Willis: Anatomical variations. Vase Dis 2: , Riggs HE, Rupp C: Variation in form circle Willis. The relation the variations to collateral circulation: Anatomic analysis. Arch Neurol 8: 8-14, Saltzman GF: Circulation through the anterior communicating artery studied by carotid angiography. Acta Radiol 52: , Mount LA, Taveras JM: Arteriographic demonstration the collateral circulation the cerebral hemisphere. Arch Neurol Psychiatry 78: , Saltzman GF: Angiographic demonstration the posterior communicating and posterior cerebral arteries. I. Normal angiography. Acta Radiol 52: 1-20, Kirgis HD, Llewellyn RC, Peebles EM: Functional trifurcation the internal carotid artery and its potential clinical significance. J Neurosurg 17: , Jen net t B, Miller JD, Harper AM: Clinical studies on 100 patients with intracranial aneurysm. In Jennett B, Miller JD, Harper AM (eds) Effect Carotid Artery Surgery on Cerebral Blood Flow. Amsterdam, London, New York, Excerpta Medica, pp , Moore WS, Yee JM, Hall AD: Collateral cerebral blood pressure. An index tolerance to temporal carotid occlusion. Arch Surg 106: , Wilkinson HA, Wright RL, Sweet WH: Correlation reduction in pressure and angiographic cross-filling with tolerance carotid occlusion. J Neurosurg 22: , Small JM, Holmes JM, Connolly RC: The prognosis and role surgery in spontaneous intracranial haemorrhage. Br Med J 2: , Mitterwallner von F: Variationsstatitishe Untersuchungen an den basalen HirngefSben. Acta Anat 24: 51-88, Levy LL, Wallace JD, Stolwijk JAJ, Poindexter ER: Cerebral

7 484 STROKE VOL 12, No 4, JULY-AUGUST 1981 blood flow regulation: Vascular resistance adjustments in the circle Willis. Stroke 7: , Kanzow E, Dieckhf D: On the location the vascular resistance in the cerebral circulation. In Brock M, Fieschi C, Ingvar DH, Lassen NA, Schflrmann K (eds) Cerebral Blood Flow. Berlin, Heidelberg, New York, Springer-Verlag, pp 96-97, Clark ME, Himwich WA, Martin JD: Simulation studies factors influencing the cerebral circulation. Acta Neurol Scand 43: , 1967 Evaluation Cross-Circulation Through Circle Willis Using An Ultrasonic Doppler Technique Part II. Comparison Between Blood Flow Velocity by Ultrasonic Doppler Flowmetry and Cerebrovascular Resistance SHOTARO YONEDA, M.D., TADAATSU NUKADA, M.D., KAZUFUMI KIMURA, M.D., KENICHI TANAKA, M.D., KEIICHI ASHIDA, M.D., TSUTOMU ASAI, M.D., HIDEKI ETANI, M.D., MASATOSHI IMAIZUMI, M.D., AND HIROSHI ABE, M.D. SUMMARY The correlation between the increase in velocity blood flow in both the internal carotid and vertebral arteries during a carotid compression and the cerebrovascnlar resistance (CVR) was investigated in 11 patients with chronic ischemic cerebrovascular disease and 4 without organic brain lesions. The velocity blood flow was measured by an ultrasonic Doppler flowmeter. CVR was calculated from cerebral blood flow and arterial blood pressure. There was no correlation between the increased velocity blood flow in the internal carotid and vertebral arteries and CVR. The increased velocity blood flow in patients with low CVR was, however, significantly higher than that patients with high CVR. The investigation cross-circulation by ultrasonic Doppler flowmetry is a useful non-invasive method for the detection changes in cerebral vascular resistance. Stroke, Vol 12, No 4, 1981 ELECTROENCEPHALOGRAPHIC monitoring," measurement carotid artery back pressure, 1-4 ~* measurement jugular venous blood gases, 7 ' and regional cerebral blood flow 18 are used for assessing the potential cerebral collateral circulation. In a previous report, 9 cross-circulation via the circle Willis was examined by ultrasonic Doppler flowmetry and was compared with angiographic variations the circle Willis. It was suggested that a peripheral vascular factor distal to the circle might play an important role in regulating increases in carotid and vertebral blood flow. This study was performed to evaluate the relationship between the cross-circulation through the circle Willis and cerebrovascular resistance (CVR). Crosscirculation via the circle Willis was evaluated by an ultrasonic Doppler flowmeter measuring changes From the First Department Medicine, Osaka University Medical School, Fukushima, 1 chome, Fukushima-ku, Osaka 553 (Drs. Yoneda, Kimura, Ashida, Asai, Etani, Imaizumi, and Abe), and National Osaka South Hospital, 677-2, Kido-cho, Kawachi- Nagano city, Osaka 586, Japan (Drs. Nukada, Tanaka). Reprints: Dr. Yoneda, First Dept. Medicine, Osaka University Medical School, Fukushima, 1 chome, Fukushima-ku, Osaka 553, Japan. velocity blood flow in the internal carotids and vertebrals during carotid compression. Materials and Methods Eleven patients with chronic ischemic cerebrovascular disease (CBVD) in the brain supplied by the carotid system without an aplastic anterior cerebral artery on angiogram, and 4 without organic brain lesions (normal brain) were examined. The ages patients varied from 15 to 70 years with an average age 48.9 years. Among the 11 patients with cerebrovascular disease there were 10 with completed stroke and one with transient ischemic attacks. Unilateral carotid angiography was performed in 7 patients and bilateral angiography in 3. Unilateral carotid and vertebral angiograms were made in one patient, and bilateral carotid and vertebral angiograms were made in the other 2 patients. Angiography was not carried out on the 2 patients without cerebrovascular disease. Angiographic findings the 11 patients with CBVD are summarized in the table. Patients were excluded if they had moderate or severe stenotic and occlusive lesions either the carotid and vertebral arteries, or the anterior, middle, and posterior cerebral arteries.