Sign Up for a Free Account
  • Updated 06.22.2023
  • Released 01.27.2015
  • Expires For CME 06.22.2026

Subclavian steal

Introduction

Overview

With subclavian artery stenosis proximal to the takeoff of the vertebral artery, there is a compensatory shunting of blood from the ipsilateral vertebral artery to supply the arm, thus, “stealing” blood from the vertebrobasilar system. In most patients with subclavian steal, the steal is asymptomatic (subclavian steal phenomenon), but a minority may have ischemic symptoms referable to the posterior circulation or the arm. In a minority of cases, neurologic symptoms may be precipitated by ipsilateral arm exercise, head-turning, or standing up. The key signs of a subclavian steal are markedly diminished, delayed, or absent radial pulses on the affected side; supraclavicular systolic bruits; and asymmetric brachial blood pressures, with the lower systolic pressure on the affected side, and generally with a minimum difference of 20 mmHg. Asymptomatic cases should be managed conservatively with vascular risk factor modification and antiplatelet therapy. Endovascular stenting and extrathoracic surgical bypass or transposition (eg, carotid-subclavian) are safe and lasting therapeutic options for selected cases of subclavian steal syndrome, particularly among those with disabling symptoms.

Key points

• With hemodynamically significant subclavian artery stenosis proximal to the takeoff of the vertebral artery, there is a compensatory shunting of blood from the ipsilateral vertebral artery to supply the arm, thus, “stealing” blood from the vertebrobasilar system.

• In most patients with subclavian steal, the steal is asymptomatic (ie, subclavian steal phenomenon), but a minority may have ischemic symptoms referable to the posterior circulation or the arm (ie, subclavian steal syndrome). In some cases, neurologic symptoms may be precipitated by ipsilateral arm exercise, head-turning, or standing up.

• Key signs of a subclavian steal are (1) markedly diminished, delayed, or absent radial pulses on the affected side; (2) supraclavicular systolic bruits; and (3) asymmetric brachial blood pressures, with the lower systolic pressure on the affected side, and generally with a minimum difference of 20 mm Hg.

• Asymptomatic cases (subclavian steal phenomenon) should be managed conservatively with vascular risk factor modification and antiplatelet therapy.

• Endovascular stenting and extrathoracic surgical bypass or transposition (eg, carotid-subclavian) are safe and lasting therapeutic options for selected symptomatic cases (subclavian steal syndrome), particularly for those with disabling symptoms.

Historical note and terminology

Recognition of the phenomenon of subclavian steal dates back to the early 19th century and the insightful observations and deductions of two surgeons (73; 150; 151; 52; 53; 54; 62; 42; 37).

A remarkably astute discussion of vascular collaterals and reversed flow in the vertebral artery with occlusion of the subclavian artery was given by Irish surgeon and anatomist Robert Harrison (1796-1858), Professor of Anatomy and Physiology in the School of Surgery at Trinity College in Dublin:

The Student having now concluded the dissection of the arteries of the Neck and superior extremity, may reconsider the various inosculations that exist between these vessels in the different regions of the Neck, Axilla, Arm, fore Arm, and Hand; and he may contemplate the chain of vascular communication extending from the Shoulder to the Fingers, so that if the main artery of the superior extremity be obliterated in any part of its course, he may comprehend those several links by which collateral circulation can be established; for it is well known that in a few hours after the operation of tying the principal artery, the pulse at the Wrist may be distinctly felt.

This communication is maintained partly by distinct vessels, which are rendered obvious by dissection; such exist around the Scapula and Elbow, and in the Hand; during life, however, there are numerous inosculations ["To unite parts such as blood vessels, nerve fibers, or ducts by small openings"—in modern parlance "collaterals"] between small arteries from distant sources in the integuments and cellular membrane through the whole of the superior extremity, even on the periosteum and within the bones; these inosculations the Dissector seldom has an opportunity of observing, but they constitute a complete vascular tissue, extending from the Shoulder to the Fingers. Indeed a careful dissection of the arteries of a limb, in which the main trunk has been for some time obliterated, clearly proves, that the anastomosing arteries are derived not from any one particular series of vessels, but that they are supplied by every contiguous ramification. It cannot, however, he uninteresting to the Student to reflect on those particular vessels which constitute the more obvious and direct media of communication, in case obstruction to the flow of blood exists in any part of the artery of the superior extremity. Suppose this obstruction to have occurred in the Subclavian Artery, in the first stage of its course, and before it has given off any branch, the Arm will be then indebted for its principal supply of blood to the following inosculations: — The Vertebral Artery, from its anastomosis with the opposite Vertebral, and with the internal Carotid Arteries, will receive a considerable share of blood, which it will transmit into the Subclavian beyond the obstruction; the inferior Thyroid Artery, from its free communication with the superior Thyroid, will contribute to the same effect (73).

In 1864, at the suggestion of New York surgeon David L Rogers (1799-1877), American surgeon Andrew Woods Smyth (1833-1916) of New Orleans, then a house surgeon at the Charity Hospital, first successfully ligated the innominate artery for subclavian artery aneurysm (151). To prevent collateral and recurrent blood flow to the aneurysmal sac, Smyth also ligated the right common carotid artery. Then, two weeks later, the ligature came away from the carotid, and the patient sustained a major hemorrhage and syncope. Over the ensuing several weeks, recurrent hemorrhages occurred that proved difficult to control. In a later procedure, 57 days after the initial procedure, Smyth ligated the right vertebral artery. Smyth's patient remained alive and functional for 10 years.

It was the occurrence of syncope and the consequent arrest of hemorrhage that first directed my attention to the vertebral artery as being the one from which the bleeding took place. This artery is capable of draining the blood directly from the brain, therefore the one most likely to produce these effects, and are petition of hemorrhage had been a striking feature in almost all the cases operated upon. Believing the hemorrhage to come from the distal side of the ligature, and from the subclavian artery, the carotid having been tied, I determined on July 8th to ligature the right vertebral artery, this being the principal branch carrying a retrograde current into the subclavian (Smith 1869).

Smyth's case was publicized by American physician and physiologist Bennett Dowler (1797-1879), who suggested after the success of the operation became apparent, that the delay between ligating the vertebral after prior ligation of the innominate and carotid arteries contributed to the procedure’s success.

The success of your operation was clearly owing to your happy resolution in relation to tying the vertebral artery. But it appears to me in reflecting on your case, that there is coupled with this, another element to be accredited to your success; and that is your having tied it at the time you did, rather than at the time of the first operation (151).

In 1960, a century after Smyth's successful surgery, Italian surgeon L Contorni described angiographic findings of retrograde flow in the vertebral artery and stenosis of the proximal subclavian artery in a patient with an absent radial pulse but without neurologic symptoms (41; 42). In 1960, neurologist James F Toole (1925-2022) suggested that proximal subclavian artery stenosis or occlusion might cause vertebrobasilar insufficiency, and the following year, Reivich, Holling, Roberts, and Toole reported two patients with evidence of cerebral ischemia and reversal of blood flow through the vertebral artery; they provided experimental results in dogs that confirmed the clinical observations (133; 162; 119; 10; 165; 76). An anonymous editorial in the New England Journal of Medicine designated the syndrome as “the subclavian steal,” and this label served to popularize the condition (11); Canadian neurologist C Miller Fisher (1913-2012) coined the term “the syndrome of the subclavian steal” (63; 119). Toole and colleagues were instrumental in the subsequent elaboration of understanding of various steal syndromes as well as the evaluation and management of affected patients (163; 164; 40; 119; 166; 10; 165). Because steals are often asymptomatic, Vollmar suggested in 1971 that the term “steal syndrome” be reserved for situations in which shunting occurs with symptoms of deficient blood flow, and “steal phenomenon” (or “steal effect”) should be used for asymptomatic steals (173). Vollmer’s suggestion has subsequently been adopted.

Various synonyms for subclavian steal have been used in the past, but none of these older terms gained acceptance, and none are current terminology. These include brachio-basilar insufficiency syndrome, brachio-vertebral syndrome, subclavian suction syndrome, and vertebral grand larceny.

Since the initial description of subclavian steal, various other steal phenomena have been described. Cerebral steal can occur in the setting of focal cerebral ischemia if the patient’s blood pressure drops, if the patient is given a vasodilator, or if the patient is hypoventilated (PaCO2 increases and pH decreases), causing arterioles in nonischemic brain to dilate and, thereby, shunting blood flow from the ischemic areas that require oxygen (56; 165; 06). This has been called the “reversed Robin Hood syndrome” on the basis that it serves to “rob the poor to feed the rich” (06). Intracranial steal phenomena also occur with angiomas and arteriovenous fistulas (105; 139). Similarly, coronary steal (cardiac steal) occurs when a coronary vasodilator is used in the setting of narrowing of the coronary arteries, causing blood to be shunted away from the coronary vessels supplying the ischemic areas and, thereby, causing further ischemia; this phenomenon is used diagnostically in drug-based cardiac stress tests. Coronary-subclavian steal syndrome is a rare but well-recognized complication of coronary artery bypass graft surgery when a left internal mammary artery (LIMA) graft is used and proximal left subclavian artery stenosis is either present or develops subsequently; the result is retrograde flow from the LIMA graft to the distal subclavian artery to perfuse the left arm, causing accentuation of, rather than relief from, cardiac ischemia (43; 127; 86; 170; 44). Vascular access steal (or dialysis-associated steal) refers to vascular shunting resulting from a surgically created arteriovenous fistula or synthetic vascular graft-AV fistula (57; 137; 26; 121; 45; 09; 39; 116; 131; 149; 174); vascular access steal syndrome is associated with pain distal to the fistula as well as pallor, depressed pulses distal to the fistula, necrosis, a decreased wrist-brachial index (below 0.6), and rarely with neurologic manifestations of vertebrobasilar insufficiency (137; 84). Axillo-femoral bypass stead due to subclavian artery stenosis has also been reported (146).

This is an article preview.
Start a Free Account
to access the full version.

  • Nearly 3,000 illustrations, including video clips of neurologic disorders.

  • Every article is reviewed by our esteemed Editorial Board for accuracy and currency.

  • Full spectrum of neurology in 1,200 comprehensive articles.

  • Listen to MedLink on the go with Audio versions of each article.

Questions or Comment?

MedLink®, LLC

3525 Del Mar Heights Rd, Ste 304
San Diego, CA 92130-2122

Toll Free (U.S. + Canada): 800-452-2400

US Number: +1-619-640-4660

Support: service@medlink.com

Editor: editor@medlink.com

ISSN: 2831-9125