Asymptomatic carotid stenosis

Introduction
Carotid stenosis refers to narrowing of the internal carotid artery and is associated with ischemic stroke via two mechanisms: rupture of atherosclerotic plaque causing in-situ thrombosis or artery to artery embolism, or ischemia from hypoperfusion usually in the setting of a concurrent episode of hypotension. The risk of stroke from rupture of the plaque is considerably higher, and significant research has been down over the years to evaluate the best methodologies to mitigate this risk.

Carotid stenosis is considered to be symptomatic when it presents with an associated ischemic stroke or transient ischemic attack, including retinal artery ischemia which, when transient, is often referred to as amaurosis fugax.

Symptomatic carotid stenosis is associated with a significant recurrent stroke risk. However, in asymptomatic carotid stenosis, meaning those patients in whom carotid stenosis was either screened for or detected incidentally, the risk of stroke remains less certain.

Epidemiology
In a large population study of 23,706 patients, asymptomatic carotid stenosis of 50-69% on carotid ultrasound was found in 2% of the population, and of >70% stenosis was found in 0.5% of the population.

Risk factors
A risk prediction scorehas been developed by de Weerd et al entitled the Stenosis Score, which includes the following components:
 * Advanced age
 * Male sex
 * Hypertension
 * Diabetes mellitus
 * Current smoking
 * Hyperlipidemia
 * History of vascular disease

Carotid endarterectomy
Asymptomatic carotid stenosis is probably best-managed with medical treatment only. Carotid endarterectomy may be reasonable in men under the age of 75 with asymptomatic carotid stenosis of ≥60%, provided that the surgeon has a complication rate of less than 3%, in which case the NNT is ~33. Surgery on asymptomatic carotid stenosis is probably not useful in women, and should be used with caution in those over age 75. Moreover, all data is compared with antiquated medical management. We may find that carotid intervention is of no benefit in these patients, as medical management has improvement significantly since the 1990s-2000s.

Trial data
====Veterans Affairs Cooperative Study (VACS, 1993) ====
 * Methods and Results:
 * This trial randomized 444 men with asymptomatic carotid stenosis of ≥50% to CEA vs medical management (aspirin 650 mg BID). The incidence of ipsilateral neurological events was 8.0% in the surgical group and 20.5% in the medical group at an average of 47.9 months, giving a RR of 0.38 (95% CI 0.22-0.67).
 * Problems:
 * In the combined analysis of perioperative death and any stroke, 11.3% of surgical patients met this outcome while 12.4% of medical patients met this outcome, which did not reach statistical significance. Thus perioperative morbidity was a potential problem.  This the basis for the subsequent Cochrane Review's analysis quoting an ARR of ~1% for this trial.

====Asymptomatic Carotid Atherosclerosis Study (ACAS, 1995) ====
 * Methods and Results
 * This trial randomized 1,662 patients with asymptomatic carotid stenosis of ≥60% to CEA plus aspirin 325 mg daily vs. medical management (aspirin 325 mg daily alone).
 * Primary outcome of ipsilateral stroke or any perioperative stroke or death occurred in 5.1% of surgical patients and 11% of medical patients (5-year RR 0.53, 95% CI 0.22 to 0.72) for a NNT of 17. This value was estimated from data obtained with a median follow-up of 2.7 years.
 * Problems:
 * In the combined analysis of perioperative death and any stroke at observed follow-up of average 2.7 years, this outcome occurred in 7.2% of the surgical arm and 10.3% of the medical arm. Thus perioperative morbidity continued to be a potential problem.  This is the basis for the subsequent Cochrane Review's analysis quoting an ARR of ~3% for this trial.
 * Subgroup analysis suggested no benefit in women.
 * 40% of surgeon applicants to participate in the trial were rejected, raising questions about generalizability. This is compounded by the low perioperative morbidity numbers of 2.3% compared with national average around that time of 4.5%.
 * Few patients had follow-up at 5 years, so the data was extrapolated. Had analysis been done at 4 years instead of 5, the trial would have been negative.

====Asymptomatic Carotid Surgery Trial 1 (ACST-1, 2004 with follow-up in 2010) ====
 * Methods and Results
 * This trial randomized 3,120 patients with asymptomatic carotid stenosis of ≥60% to CEA vs possible delayed CEA (medical group). The 5-year stroke rate was 3.8% in the surgical group vs 11% for the medical group. However, this value was also estimated from a mean follow-up time of 3.4 years and importantly excluded perioperative events which were significant.
 * At 10-year follow-up, the risk of stroke or perioperative death in the immediate CEA group was 13.4% compared with 17.9% in the medical treatment group. This corresponds to an ARR of only 0.46% per year.
 * Problems:
 * In the combined analysis of perioperative death and any stroke at observed follow-up of average 3.4 years, this outcome occurred in 5.2% of the surgical arm and 8.4% of the medical arm. Thus perioperative morbidity continued to be a potential problem.  This is the basis for the subsequent Cochrane Review's analysis quoting an ARR of ~3% for this trial.
 * Subgroup analysis suggested no significant benefit for women. However, at 10 years there appeared to be a small benefit for women under age 75, with any stroke occurring in 10.2% compared with 16.0% in the medical treatment arm, so an ARR of 0.58% per year, which may not be significant.
 * No benefit for patients over age 75 years
 * No association seen between stroke risk and severity of carotid stenosis

Cochrane Review (2005)
Taking the available data from the VA study, ACAS, and ACST-1, benefit was confirmed for surgery in preventing perioperatigve stroke or death or any subsequent stroke, with a RR 0.69 (95% CI 0.57 to 0.83). The ARR was 3% overall, corresponding to a NNT of ~33. Additionally, the overall net excess of operation-related stroke or death was 2.9%, which is the basis for the recommendation that surgical complications must be less than 3%, otherwise wll benefit would be negated. Additionally, when stratified by gender all of the benefit seemed to be in mean and not in women, which is hypothesized to do with more complicated surgery in women due to smaller arteries. Younger patients benefited while older patients did not but there were limited numbers of older patients in the trials so definitive conclusions can't be made (defined as under age 68 in ACAS and under age 75 in ACST.

Major trial criticisms

 * Poor medical management in the trials: Medical management was limited to aspirin for the majority of patients in each trial with very few patients receiving statins. In fact, only 10% of patients in the most recent trial, ACST-1, were on lipid-lowering medications at the beginning of the trial, with this number reaching ~80% by 2007. The rates of annual ipsilateral stroke with medical intervention alone has fallen since the 1980s, from 3.3% in 1985 to 0.6% in 2007 with a meta-analysis of data through 2009 suggesting an annual stroke risk of 0.5% pear year. Carotid plaque appears to be stabilized by statin therapy. . A decision analysis has estimated that there would be no benefit of surgery if risk of fatal and disabling stroke with medical treatment is less than 1.09% per year. Medical treatment may now be sufficient to meet this goal.
 * Surgical risk is a potential problem. From 1995-1996, 30-day stroke or death after CEA for asymptomatic patients occurred in  >3% of cases in 6 of the 10 US states studied, with average risk of 30-day stroke or death in 10 US states being 3.7%. From 1998-1999 there was little improvement, with >3% 30-day stroke or death after CEA in 7 of the 10 US states studied, with average risk of 5.4%. A decision analysis has estimated that there would be no benefit of surgery if the perioperative rate of death or disabling stroke is more than 2.1%.
 * As CEA appears to have limited benefit and most strokes are not due to previously asymptomatic carotid disease, it has been suggested that even if every patient with asymptomatic carotid stenosis of 60-99% underwent CEA, 96% of strokes would still occur.

Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE, 2004)
This trial randomized patients with symptomatic carotid stenosis of ≥50% and asymptomatic carotid stenosis of ≥80% who were considered high risk for CEA to either CEA or carotid stenting (CAS). They randomized 334 patients and found CAS to be noninferior to CEA with a trend towards benefit in their primary end point (combination of death, stroke, or MI at 30 days plus ipsilateral stroke or death from neurological causes up to 1 year), 12.2% in CAS group and 20.1% in CEA group (p=0.05). This was driven by a decrease in the incidence of MI and major ipsilateral stroke in the CEA group. There were statistically fewer cranial nerve palsies in the CAS group (occurred in 5% of CEA group and no CAS patients), and numerically fewer deaths in the CAS group (7.4% vs 13.5%, p=0.08). Importantly this trial required the use of emboli protection devices.

Carotid Revascularization Endarterectomy vs. Stenting trial (CREST, 2010 with follow-up in 2016)

 * Methods and Results
 * This trial initially set out to randomize only symptomatic patients, but they subsequently included patients with asymptomatic carotid stenosis of ≥60% on angiography, ≥70% on ultrasonography, or ≥80% on CTA/MRA if 50-69% on ultrasonography. 2,502 patients were randomized. The incidence of stroke was higher in the CAS group (4.1%) vs the CEA group (2.3 %) for a HR of 1.79 (95% CI 1.14 to 2.82).  In contrast, the incidence of myocardial infarction (MI) was lower in the CAS group (1.1%) vs. the CEA group (2.3%) for a HR of 0.50 (95%C CI 0.26 to 0.94).  The primary endpoint of any periprocedural stroke, MI, death, or postprocedural ipsilateral stroke was not different between the two groups.  At 10 year follow-up there remained no difference in the primary endpoint, but there continued to be a higher risk of stroke or periprocedural death at 10 years (11.0% vs 7.9%, HR 1.37 with 95% CI 1.01 to 1.86).  Long-term stroke incidence was not significantly different, with all of the data suggesting that periprocedural stroke incidence with CAS was the problem, and that otherwise it was likely as durable a treatment as CEA.
 * Problems

Stenosis severity
Neither ACAS or ACST-1 showed any evidence that stenosis severity predicted late stroke. . However, in 2005 the nonrrandomized Asymptomatic Carotid Stenosis and Risk of Stroke (ACSRS) study showed an increase in risk of TIA/stroke with degree of stenosis in 1,115 patients at a mean follow-up of 37.1 depending on severity of stenosis, as follows:

Importantly, however, reliance on stenosis severity has not been sufficient to identify patients at high risk of stroke that will benefit from intervention.

Stenosis progression
An increase in stenosis by at least one prespecified category (categories include 0-29%, 30-49%, 50-69%, 70-89%, 90-99%, and 100%) over 6-9 months has been associated with an increase in the 3-year risk of stroke from 2.5% to 5.0% (OR 2.00, 95% CI 1.02-4.11). However, this has not been well-enough studied to determine its utility in stratifying patients for intervention.

Contralateral TIA or stroke
A history of contralateral TIA or stroke is associated with a 3.4% annual risk of stroke compared with 1.2% in those without such a history.

Silent ipsilateral infarct on imaging
In ACSRS, of the patients who had a CT scan at the time of recruitment, those with evidence of silent embolic infarction had an excess risk of ipsilateral stroke (3.6% vs 1.0%, p=0.002).

Ultrasound parameters
INSERT TABLE FROM Naylor, A. R. Nat. Rev. Cardiol. 9, 116–124 (2012); published online 11 October 2011

Microemboli detection
Transcranial doppler (TCD) detection of microemboli has been shown to be highly predictive of the risk of ipsilateral stroke in otherwise asymptomatic patients, with HR 7.46 (95% CI 2.24-24.89). Microemboli may be useful in patients started on best medical therapy to evaluate several months later whether they are responding, and if not, they could be considered for CEA or CAS.

The Asymptomatic Carotid Emboli Study (ACES) observed 467 patients with asymptomatic carotid stenosis of at least 70% from 26 different centers worldwide. Patients had two 1 hour transcranial doppler (TCD) recordings from the ipsilateral MCA at baseline and then at 6, 12, and 18 months, with follow-up at 2 years. Embolic signals were detected on TCD in 77 patients, and the HR for ipsilateral or stroke or TIA up to 2 years for those with embolic signals compared with those who did not was 2.54 (95% CI 1.20-5.36) and for ipsilateral stroke was even higher at 5.57 (95% CI 1.61-19.32). The authors also performed a meta-analysis including prior smaller studies and showed that in the presence of embolic signals, the HR for ipsilateral stroke was 7.46 (95% CI 2.24-24.89) and for ipsilateral stroke or TIA was 7.57 (2.32-24.69). With embolic signals present, stroke occurred in 8.7% of patients, and without they occurred in only 1.4%. Microemboli also improve over time with medical therapy.

MRI plaque imaging
In patient with 50-79% asymptomatic carotid stenosis, MRI evidence of thinned or ruptured fibrous caps (HR 17.0), intraplaque hemorrhage (HR 5.2), larger lipid cores, or larger maximum wall thickness are associated with higher rates of ipsilateral stroke. The effect of intraplaque hemorrhage was separately confirmed.