16 EC-IC Bypass Evidence

Paul R. Gigante, Christopher P. Kellner, E. Sander Connolly, Jr.

The first description of an operation intended to provide alternative cerebral blood flow (CBF) was given by German and Taffel in 1939, in which a temporalis muscle pedicle was placed over the cerebral cortex in monkeys (Figure 16–1).¹ They later subjected both treated and untreated controls to carotid and vertebral occlusion, and reported the monkeys with temporalis pedicles overlying the cortex survived, whereas untreated controls did not. In 1942, Kredel first performed the procedure in three human patients who had suffered ischemic stroke, reporting postoperative clinical improvement in all three patients.² However, the technique was largely abandoned over the next decade when postoperative angiograms failed to provide evidence of collateral flow through the graft.

Figure 16–1 Outline of Advances in EC-IC Bypass Techniques.

C. Miller Fisher shed new light on the concept of re-establishing cerebral perfusion in 1951 when he postulated that extracranial to intracranial (EC-IC) arterial bypass might be performed as a treatment for occlusive vascular disease.³ That same year, Pool and Potts created a superficial temporal artery (STA) to anterior cerebral artery (ACA) shunt during treatment of a giant ACA aneurysm. Although the shunt was successful intraoperatively, postoperative angiogram confirmed the shunt had become occluded.⁴

The first EC-IC arterial bypass was not described until 1966 when Yasargil and Donaghy successfully performed an STA to middle cerebral artery (MCA) anastomosis in a dog. Shortly thereafter, Yasargil successfully performed the same procedure in a human, providing the neurosurgical community with what was deemed a breakthrough in the treatment of intracranial atherosclerosis. Following Yasargil’s description of the EC-IC arterial bypass, neurosurgeons around the world utilized the procedure with the intent of re-establishing CBF in a variety of cerebrovascular diseases. Surgeons subsequently described bypasses from the occipital and middle meningeal arteries to the MCA, from the occipital artery to the posterior inferior cerebellar artery,⁵ from the occipital artery to the anterior inferior cerebellar artery (AICA), and from the STA to the superior cerebellar artery.⁵ ^,6

Over time surgical technique improved, and the most frequently employed bypass procedure became the STA to MCA bypass for the indication of symptomatic internal carotid artery (ICA) or MCA stenosis. However, as was the case with most other surgically treated diseases at the time, there were no extant data to demonstrate that performing the procedure provided patients with any significant short- or long-term benefit. The only early report came in the form of a small retrospective series published by Sundt et al. in 1977, the utility of which was limited by multiple biases. The group reported on a retrospective series of 56 patients who underwent STA-MCA bypass for a wide variety of indications, including TIAs, “orthostatic ischemia,” progressive stroke, and aneurysmal bypass. With regards to graft patency, those who had grafts prior to 1973 had low patency (25%) whereas those operated after 1974 had high patency (95%). The other outcome described was major ischemic stroke or death from the surgery, of which they reported none. The report served more to describe the acceptable safety of the procedure, rather than to examine its efficacy.⁷

As evidenced by this first series, the greatest impediments to a meaningful conglomerate of data were the lack of consistency in patient selection and evaluation, the lack of an appropriate medically treated control population, and the lack of consistent outcomes follow-up. As evidence-based medicine emerged and the Food and Drug Administration’s requirements for proving pharmacologic safety and efficacy became increasingly stringent, neurologists and neurosurgeons too recognized the imperative of defining the “efficacy” of risk-containing surgical procedures. In a 1977 Stroke editorial introducing the initiation of the first ever EC-IC bypass study, Fletcher H. McDowell wrote,

Too often surgical treatment for a particular condition gains enthusiasm and wide use before any clear evidence of effectiveness. This problem has been particularly evident in the field of atherosclerotic vascular disease involving the treatment of its cardiac and cerebral complications. The recent suggestions for extracranial/intracranial arterial anastomosis in the prevention and treatment of stroke have without question a logical basis. This is especially true when atherosclerosis in cerebral vessels lies distal to the surgically accessible portion of the internal carotid artery. Anastomosis of the superficial temporal artery to the middle cerebral artery or one of its branches to improve vertebral circulation in such situations is beginning to be carried out frequently in a number of centers across the United States. In an effort to determine the effectiveness of this therapy 20 major medical centers in the United States and three centers outside the United States have joined together in a collaborative study of this therapy.⁸

Pending the results of the EC-IC Bypass Study, neurosurgeons published a handful of case reports, series, and commentaries on the appropriate indications, safety, and utility of EC-IC bypass.⁹^–¹⁵ In 1983, Chater reported on 400 patients who underwent EC-IC bypass for TIAs and hemodynamic intracranial lesions, with a permanent neurological morbidity rate of approximately 2% and operative mortality of 2.5%, and an ipsilateral postoperative stroke incidence of 0.9% per year after a 55-month follow-up.¹⁶ Chater et al. then reported on 105 patients with intracranial ICA stenosis (60% to 98%) and ischemia 1 to 3 months prior to EC-IC bypass. Of these, surgical mortality was 1%, permanent surgical morbidity was 2%, and after 54 months of follow-up, 22 patients had died, of which three were related to stroke; thus the overall late death rate was 4% per year. Although Chater’s reports had begun to examine a more uniform patient population, both of his retrospective series contained no medically treated control group against which the success of surgery could be measured.¹⁷