Br HeartJ 1993;70:193-194

CASE REPORTS

Magnetic resonance angiography in subclaviansteal syndrome

Paul D Flynn, David J Delany, Huon H Gray

AbstractA case is reported of a patient with thesubclavian steal syndrome in whom thereversed blood flow of the vertebralartery was shown by phase encoded mag-netic resonance angiography.

Department ofCardiology, WessexCardiothoracicCentre, SouthamptonGeneral Hospital,SouthamptonP D FlynnD J DelanyH H GrayCorrespondence to:Dr D J Delany, Departmentof Cardiology, WessexCardiothoracic Centre,Southampton GeneralHospital SouthamptonS09 4XY.

(Br HeartJ_ 1993;70: 193-194)

Reversed blood flow in the vertebral arteryassociated with neurological dysfunction wasdescribed first in 1961 by Reivich et al,'although the concept of retrograde flow in thecerebral circulation had been presaged byWillis in his "cerebri anatome" of 1664.2Reivich et al described two patients withstenosis of the left subclavian artery proximalto the origin of the left vertebral artery, and

Figure 1 Spin echo coronalMRI in the plane of the aorta and subclavian, and vertebralartery junctions. The absence offlow in the proximal subclavian artery (arrow) wasconfirmed on transverse images.

showed angiographically that blood in thevertebral artery was flowing from the verte-brobasilar junction to the subclavian artery.Furthermore in experiments on anaesthetiseddogs it was shown that occlusion of the sub-clavian artery resulted in a reversal of flow inthe ipsilateral vertebral artery, and increasedflow in the contralateral vertebral and bothcarotid arteries. Antegrade basilar and totalcerebral blood flow were, however, reduced.It was postulated that, with subclavian steno-sis, activity that increased the reversed verte-bral flow, such as exercise of the ipsilaterallimb, could result in transient vertebrobasilaror cerebral hypoperfusion. This phenomenonwas labelled "the subclavian steal syndrome"by Fisher.3

Subsequently, subclavian steal wasreported in many other patients and detectedeither by arteriography4 or more recently byDoppler ultrasound of the extracranial arter-ies.56 We report here a case of subclaviansteal syndrome in which the steal was shownnon-invasively by phase contrast magneticresonance angiography (MRA).

Case reportA 72 year old woman was admitted a fewhours after a collapse at home. She had beenwell, and had been standing at her kitchensink washing elderberries for 20 minuteswhen she gradually developed some mildbackache. A few minutes later she slumpedbackwards and was caught and lowered to thefloor by her husband. She did not lose con-sciousness but seemed disoriented. There hadbeen no chest pain, no palpitation, nopremonitory aura, or any other epileptiformfeatures. Her general practitioner was called,and on arrival he noted weak pulses in her leftarm and arranged her admission to hospitalwith a diagnosis of a possible thoracic aorticdissection. She had fainted 11 years previ-ously, but otherwise she had been fit andwell. She had smoked cigarettes for manyyears.

Examination showed a regular pulse at 80beats per minute, and confirmed the relativeweakness of the left arm pulses with a differ-ential blood pressure of 148/80 mm Hg in theright arm and 112/80 mm Hg in the left. Aleft-sided carotid bruit was noted. It was atthis point that the patient remarked that, as aformer nurse, she had on occasion felt herown pulses and had noted their weakness inher left arm for some years. There were noother cardiovascular signs and neurologicalexamination was normal.

193

on March 6, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.70.2.193 on 1 August 1993. D

ownloaded from

Flynn, Delany, Gray

qbp__.Figlure 2 Phase conztrast 'IRA in the samiie planie as fig 1. Flozt enzcodin?g is blackfor cephalic and white for caudal. The vertebral artery positions are showtn by linemiarkers.

Normal investigations included a full bloodcount, biochemical screen (including glu-cose), measurement of a series of cardiacenzymes, and electrocardiograms. Her chest xray film was normal and showed no evidenceof mediastinal widening. On computed tomo-graph scanning of the thoracic aorta there wasatheromatous disease of the descending tho-racic and abdominal aorta but no evidence ofdissection. It was noted that the left subcla-vian artery did not opacify with contrast. Spinecho MRI images confirmed occlusion of theleft subclavian (fig 1), and MRA with flowencoding gradients was performed showing

antegrade flow in the right vertebral artery,but retrograde flow in the left vertebral artery,constituting a classical subclavian steal (fig 2).Thus a combination ofMRI and MRA was

able both to exclude other diagnostic consid-erations and accurately and definitively makethe true diagnosis without the need for anyinvasive procedure.

DiscussionThe combination of history, physical findings,and retrograde flow in the left vertebral arteryon MRI was considered diagnostic of the sub-clavian steal syndrome. In the 1960s treat-ment of this condition was usually surgicalbut Fields and Lemak in 1972 showed no sig-nificant difference in mortality between thosepatients who had undergone surgery andthose who had been managed conservatively.4The proportion of patients achieving a goodoutcome was also similar, and the rate ofstrokes was marginally higher in the surgicalgroup. In 1988 Hennerici et al showed thatmost patients with isolated unilateral sub-clavian steal were symptom free, and sug-gested that in most of those who were not,the symptoms were most likely to result fromcoexistent cerebrovascular disease. In supportof this they followed up 54 patients with uni-lateral subclavian steal alone with Dopplerexaminations over seven years. Elevenpatients died, only one of them from a stroke.Eight further patients had transient ischaemicattacks, all of whom were shown on repeatDoppler examination to have developedcarotid artery stenosis.The consensus now seems to be that iso-

lated subclavian steal is a relatively benignphenomenon, but that it may be associatedwith other cerebrovascular disease for whichit could be considered a marker. Our patientwas advised to stop smoking if possible, andto avoid any prolonged exertion of her leftarm, especially while standing. Otherintervention will only be considered if shedevelops further evidence of disablingvertebrobasilar insufficiency.

1 Reivich M, Holling HE, Roberts B, Toole JF. Reversal ofblood flow through the vertebral artery and its effect oncerebral circulation. NEnglJMed 1961;265:878-85.

2 Willis T. Practice of physick. London: S Pordage, 1684;Part 6: 59.

3 Fisher CM. A new vascular syndrome-"the subclaviansteal". NEnglJMed 1961;265:912-3.

4 Fields WS, Lemak NA. Joint study of extracranial arterialocclusion. VII. Subclavian steal-a review of 168 cases.JAAMA 1972;222:1139-43.

5 Bomstein NM, Norris JW, Subclavian steal: a harmlesshaemodynamic phenomenon? Lancet 1986;ii:303-5.

6 Hennerici M, Klemm C, Rautenberg W. The subclaviansteal phenomenon: a common vascular disorder withrare neurologic deficits. Neurology 1988;38:669-73.

194

on March 6, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.70.2.193 on 1 August 1993. D

ownloaded from