Introduction

The first description of a carotid endarterectomy (CEA) for stroke was published in the Lancet in 1954.⁵ Since that time, more clinical trials have been completed for CEA than for any other surgical procedure, first addressing the question of surgery versus best medical practice (clearly answered in favor of surgery in most cases) and, most recently, addressing the question of surgery versus endovascular carotid intervention. (These latest trials currently also favor open surgery versus endovascular intervention.)

We can review current indications for surgical intervention for carotid stenosis as follows. The first large cooperative trials randomized surgical reconstruction versus best medical therapy. The North American Symptomatic Carotid Endarterectomy Trial (NASCET) established that symptomatic patients with internal carotid artery stenoses of 50% or more benefit from surgical resection of the offending lesion.^1,3 The Asymptomatic Carotid Atherosclerosis Study (ACAS) and the European Asymptomatic Carotid Surgery Trial (ACST) trial^6,7 provided compelling evidence that asymptomatic patients with 60% or more stenosis of the cervical internal carotid artery have a more favorable outcome with endarterectomy, provided that perioperative morbidity and mortality remain low (<3%),² and that the patient otherwise has a 5-year life expectancy (the time needed to achieve a surgical benefit).

Surgical reconstruction (CEA) has now been randomized against carotid stenting (CAS) in several trials. To date, none has shown an advantage of CAS in either safety or efficacy. The latest trials, EVA-3S and SPACE, continue to demonstrate that open surgery is the safest and best treatment in most cases.^28,29 The SAPPHIRE trial randomized “high-risk” asymptomatic and symptomatic carotid stenosis patients to CAS or CEA.³² Both techniques were effective, with statistically equal risk, in the treatment of symptomatic patients. For asymptomatic patients, there was a trend (p = NS) to lower risk with CAS. Questions have been raised, however, whether high-risk asymptomatic patients should be treated at all, since American Heart Association (AHA) guidelines have specified that surgical treatment is inappropriate for these patients.⁴ The only true answer to this question would be a study of CAS versus medical therapy in this group. No such study exists at present, although one large but non-randomized three-arm trial, the Japan Carotid Atherosclerosis Study (JCAS), will be forthcoming from Japan.

The National Institute of Neurological Disorders and Stroke (NINDS) and National Institute of Health (NIH)–funded Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) completed enrollment of 2522 patients (53% symptomatic, 47% asymptomatic) on July 18, 2008 using the same criteria as NASCET and ACAS.^10,30 Final results of this trial have yet to be published. CREST and other similar trials in Europe (The International Carotid Stenting Study–Carotid and Vertebral Artery Transluminal Angioplasty Study [ICSS–CAVATAS 2]) and in Japan will hopefully provide level 1 evidence for therapeutic decision making. Curiously, the lead-in CREST data show that the risk of CAS, 30-day rates of stroke and death = 11.1%, is unacceptably high in elderly patients, aged >75, a group that was thought to represent excellent candidates for the procedure because of medical comorbidities.^8,10

Indications for surgery

Patients with neurological symptoms localizing to the side of their carotid stenosis are considered symptomatic. Neurological symptoms include ipsilateral monocular blindness, contralateral motor weakness with greater involvement of the arm and face compared to the leg, contralateral sensory deficits, and aphasias (if the dominant hemisphere is involved). Symptoms may take the form of transient ischemic attacks (TIAs), reversible ischemic neurological deficits (RINDs), or cerebrovascular accidents (CVAs), the duration of the deficit being the primary difference. Non-specific neurological complaints, such as dizziness, syncope, or ill-defined visual disturbances, do not meet the criteria for symptomatic carotid stenosis.

Most patients with symptomatic carotid artery stenoses of ≥50%, as measured by the NASCET method, are offered surgery at our institution. We offer endovascular counseling to these patients particularly if they have major medical comorbidities but we educate them that there is no level 1 evidence that stenting is equivalent to surgery in any but the highest-risk cases. The patients are counseled pre-operatively about surgical risks and benefits. There is good evidence that nondiabetic men with hemispheric symptoms have the greatest benefit from surgery; diabetes certainly increases the surgical risk, and amaurosis fugax appears to be a marker for lower stroke risk carotid lesions.²⁶

Asymptomatic patients with linear stenoses of 60% or more benefit from surgery, as determined by the ACAS and ACST trials. It is important to appreciate, however, that the patients should have an expected 5-year survival following surgery for the benefit to be realized; in addition, the benefit is not as great in women, and the combined surgical morbidity and mortality must remain below 3%. Asymptomatic patients with metastatic cancer, profound medical comorbidities, or other life expectancy-reducing conditions should generally not be offered treatment.

Preoperative evaluation and preparation

Carotid stenosis is identified in many different ways. Some patients come for surgical evaluation following workup for a TIA or stroke, where carotid duplex scanning, CTA, or MRA reveals a significant carotid lesion. Other patients present with clinically silent carotid bruits. Regardless of the referral type, we prefer catheter angiography (arch, both carotids, cervical, and cranial) for all patients being considered for surgery. This allows both accurate measurement of the stenosis using NASCET methodology and appropriate preoperative planning. The NASCET equation, where N is the linear diameter at the region of greatest narrowing and D is the greatest diameter of the normal artery distal to the carotid bulb, follows:

Preoperative imaging should also include CT or MRI of the brain to exclude the possibility of a mass lesion mimicking cerebral ischemia. Patients should also undergo appropriate preoperative medical risk stratification and workup especially for cardiac risks factors, and their medical management of cardiac disease, hypertension, hyperlipidemia, and diabetes should be optimized.

Surgical technique

Positioning

The patient is placed in the supine position with the head extended and turned away from the side of the operation (Figure 19–1). Folded sheets or pillowcases are placed beneath the shoulder blades to aid in extension of the neck; the preoperative arteriogram is reviewed to visualize the relationship between the internal and external carotid arteries (ICA and ECA) to choose the appropriate degree of head turning. Normally, the carotid vessels are superimposed in the anteroposterior plane, and rotating the head away from the operative side improves intraoperative visualization of the ICA by forcing it into a more lateral position. When the ICA is medial to the ECA (perhaps 10% of cases), no amount of head rotation will bring the ICA lateral enough, and the surgeon should recognize this and be prepared to dissect posterior and inferior to the ECA to find the “hidden” ICA and mobilize it laterally into the surgical position.

Figure 19–1 Initial positioning. The patient is in the supine position with the head extended and turned away from the operation side. The degree of rotation will be dependent on the relationship of the ICA and ECA. Generally, rotation brings the ICA into a more lateral position. Note the marking for the linear incision along the medial border of the sternocleidomastoid muscle.

Operative technique

The cerebral angiogram is used to locate the height of carotid bifurcation, and the incision is planned accordingly. We prefer a linear incision along the medial border of the sternocleidomastoid muscle (see Figure 19–1), which may extend as high as the retroaural region and as low as the suprasternal notch, as needed. Following sterile prep (never a vigorous scrub) and drape, the skin and subcutaneous tissues are dissected sharply down through the platysma, unavoidably transecting the transverse cervical nerve. (For this reason, patients should be instructed preoperatively to expect numbness anterior to the incision, which usually resolves after 6 months.) Meticulous hemostasis is obtained using monopolar and bipolar cautery superficially, and bipolar alone once the dissection is deep to the sternomastoid muscle. The medial edge of the sternocleidomastoid muscle is located and kept in view by placing a superficial blunt-bladed self-retaining retractor and dissecting through the overlying fat (Figure 19–2). The medial retractor blade (of this and all retractors) remains superficial to prevent injury to the recurrent laryngeal nerve in the tracheo-esophageal groove, while the lateral blade may safely be placed more deeply, on or under the sternocleidomastoid muscle.

Figure 19–2 Initial dissection. After sharp and electrocautery dissection, the anterior edge of the sternocleidomastoid muscle is identified. Care is taken that the medial blunt retractor remains superficial, avoiding delicate structures below, particularly the recurrent laryngeal nerve in the tracheo-esophageal groove.

Attention is next focused on the middle of the incision where the dissection proceeds down the sternocleidomastoid muscle until the internal jugular vein is reached. When dissection is carried under the muscle, the surgeon must be careful not to injure the laterally-placed spinal accessory nerve by inadvertent transection or over-retraction. The jugular vein is a key landmark in the dissection, and it typically lies lateral, parallel, and slightly anterior to the ICA and common carotid artery (CCA). The medial jugular border is fully exposed, and the vein can be retracted as needed using blunt blades over a cottonoid pattie; it is crucial that the retractor blades are blunt to prevent vascular injury. In this portion of the dissection, the rather large common facial vein, and occasionally several smaller veins, are doubly ligated and divided. The common facial vein is a key landmark as it generally crosses the field at the level of the carotid bifurcation and carotid bulb (Figure 19–3).

Figure 19–3 Dissection. Identification of the common facial vein is an important step in the dissection. The common facial vein often marks the level of the carotid bifurcation. We always ligate and divide the common facial vein to gain adequate carotid exposure.

The CCA is typically encountered first, and at first visualization we instruct the anesthesiologist to administer 5000 units of intravenous heparin, which we do not reverse with protamine sulfate at the end of the procedure. The carotid is lifted and separated from surrounding structures upon opening of the carotid sheath with the aid of four sutures (4-0 silk) tied to more superficial tissue (Figure 19–4