I. Diagnosis

  1. Carotid aa. bruits aren't specific or sensitive for "surgical" stenoses (70-99%); in most severe stenoses bruits are uncommon
  2. Supraclavicular bruit is 96% specific (but not very sensitive) for "surgical" stenoses (both of above from Ann. Int. Med 120:633; n = 1268)
  3. Carotid artery duplex scanning
    1. Uses plain ultrasound plus Doppler to assess peak systolic velocity, on the basis of which is calculated degree of stenosis
    2. Accuracy of Carotid Duplex was 80-89% for mild (< 50%) and severe (> 70%) carotid stenoses but only 46% for moderate (50-69%) stenoses, in one study using angiography as the gold standard (Radiol. 214:247, 2000--JW)
  4. Magnetic Resonance Angiography-Probably more accurate than duplex ultrasound
    1. In a meta-analysis of 63 studies comparing either MRA or duplex u/s to DSA as the gold standard for dx of carotid artery stenosis (Stroke 34:1324, 2003):
      1. For diagnosis of 70-99% stenosis, MRA had sensitivity 95%/spec 90% and u/s had sensitivity 86%/spec 87% (MRA sig. better discriminatory power)
      2. For diagnosis of total occlusion, MRA had sensitivity 98%/spec 100%; u/s had sensitivity 96%/spec 100% (no sig. diff. in discriminatory power)
  5. In a meta-analysis of 41 studies comparing non-invasive carotid imaging techniques with intra-arterial angiography in pts with clinical evidence of cerebrovascular disease (Lancet 367:1503, 2006--JW)

II. Risks of carotid endarterectomy

  1. 30d risk of death = 1.1%, disabling CVA = 1.8%; nondisabling CVA = 3.7%. Independent predictors of perioperative CVA or death were as follows ("NASCET" trial, Stroke 30:1751, 1999--JW)
    1. Hemispheric (as opposed to retinal) TIA
    2. Left-sided procedure
    3. Contralateral carotid a. occlusion
    4. Ipsilateral ischemic lesion on CT
    5. Ipsilateral ulcerated placque

III. Carotid endarterectomy (CEA) to treat symptomatic carotid stenosis (positive h/o TIA or CVA)

  1. Carotid endarterectomy associated with sig. reduction in risk of ipsilateral CVA c/w medical management alone in pts with h/o TIA or nondisabling CVA and >70% extracranial carotid artery stenosis (Lancet 337:1235, 1991; "NASCET" trial--NEJM 325:445, 1991)
    1. Pts > 75yo more likely to benefit from endarterectomy with less baseline stenosis--In a follow-up trial of a subset of 2,885 NASCET pts > 75yo with symptomatic internal-carotid a. stenosis (Lancet 357:1154, 2001--JW)
      1. Those w/baseline stenosis 70% or more had RR for ipsilateral ischemic CVA of 0.71 among pts > 75yo, 0.85 in pts 65-74, and 0.9 in pts < 65yo
      2. Those w/baseline stenosis 50-69%, only pts > 75yo had sig. lower risk for ipsilateral ischemic CVA (RR 0.83)
      3. Perioperative risk of (CVA or death) was 5% for pts > 65yo
  1. European Carotid Surgery Trial--Multicenter trial of 3,000 pts with carotid stenosis and at least one carotid territory TIA randomized to endarterectomy vs. observation; mean f/u 6y. No sig. diff. in rate of major stroke or death in overall comparison. Among pts with at least 80% stenosis, surgical group had sig. lower 3y risk of major stroke or death (14.9 vs. 26.5%). Among women, sig. benefit from surgery only seen with 90% or more stenosis (Lancet 351:1379, 1998--JW)
  1. 858 pts with h/o TIA or nondisabling CVA and 50-69% carotid a. stenosis randomized to endarterectomy vs. medical management, sig. less risk of ipsilateral CVA over 5y of f/u (15.7% vs. 22%; greater benefit in men); no sig. diff. in 5y risk of ipsilateral CVA in 1,368 pts with < 50% stenosis (NEJM 339:1415, 1998--JW--a f/u of NASCET?)
  2. In data pooled from the European Carotid Surgery Trial and NASCET, both of which randomized pts with symptomatic carotid stenosis to surgery vs. medical tx, perioperative risk of (death or CVA) was sig. greater in pts who were: female, diabetic, had occlusion of the contralateral carotid, and pts with ulcerated or irregular placques (Lancet 363:915, 2004-AFP)

IV. Carotid endarterectomy to treat asymptomatic carotid stenosis (no h/o TIA or CVA)

  1. "Asymptomatic Carotid Atherosclerosis Study" ("ACAS"; JAMA 273:1421, 1995)
  1. Multicenter trial randomized 1662 pts 40-79yo with >60% stenosis & no sx to ASA or carotid endarterectomy; "selected for low surgical risk"
  2. Median followup 2.7y
  3. 2.3% perioperative incidence of CVA or death in CEA group
  4. Estimated 5y risk for ipsilateral CVA or any perioperative CVA was 5.1% in CEA group vs. 11% in control group
  1. 372 asymptomatic pts with carotid bruits and at least 50% stenosis in at least one carotid artery were randomized to 325mg ASA QD or placebo, followed for mean 2.4y; no difference in cerebral ischemic events, progression to higher grade of stenosis, or death. (Ann. Int. Med. 123:649, 1995)
  2. Meta-analysis of 5 trials (total 2440 pts) with asymptomatic carotid stenosis 50% or more randomized to surgery or no; over 3y f/u, risk of (ipsilateral stroke or perioperative stroke or death) was 4.4% for surgical pts vs. 6.4% for those not randomized to surgery (sig.). 30d post-op risk of stroke or death was 2.4% in the surgery groups (BMJ 317:1477, 1998--JW)
  3. 3,120 pts with 60% or greater carotid stenosis and no related sx randomized to carotid endarterectomy vs. nonsurgical tx; over 5y, risk of any CVA was sig. lower in CEA group (6.4% vs. 11.8%); ditto for fatal CVA (3.5% vs. 6.1%) ("Asymptomatic Carotid Surgery Trial" (ASCT) Lancet 363:1491, 2004--JW)
    1. In a follow-up report on the ACST trial, over up to 10y f/u, the pts randomized to endarterectomy had sig. lower overall incidence of non-perioperative CVA (10.8% vs. 16.9%); effect was sig. in subgroup of pts on lipid-lowering drugs but not in pts > 75yo at enrollment (Lancet 376:1074, 2010-JW)

See Lancet 353:2105, 1999 for a retroactively validated scoring system to predict benefit from carotid endarterectomy in pts with > 70% stenosis

V. Carotid stenting

  1. Only 2% restenosis rate at 19mos in one uncontrolled study of 170 pts (JACC 35:1721, 2000--JW)
  2. May cause intra-operative embolization of placque debris to the brain
  3. In a randomized trial in 334 pts with carotid stenosis (> 50% with symptoms or > 80% w/o sx) randomized to CEA vs. stenting (using a filter basket to "catch" placque debris during stent placement), incidence of (death, CVA, or MI in first 30d or ipsilateral CVA or neurologic death in first 1y) was lower in stenting pts; the difference was of borderline statistical significance (12% vs. 20%, p = 0.05) (NEJM 351:1493, 2004--JW)
  4. In a study in 520 pts with 60-99% symptomatic carotid stenosis with h/o CVA or TIA within prior 120d randomized to CEA vs. carotid stenting, 6mo incidence of (CVA or death) as sig. lower in CEA group (6.1% vs. 11.7%) ("EVA-3S" trial; NEJM 355:1660, 2006--JW)
  5. In a study in 1,200 pts with > 69% carotid-artery stenosis and moderate ischemic CVA or TIA in the prior 180d randomized to CEA vs. carotid stenting, 30d incidence of (ipsilateral CVA or death) was not sig. diff. in the two groups ("SPACE" trial; Lancet 368:1239, 2006--JW)
  6. In a study in 2,502 adults (47% asymptomatic) with carotid stenosis randomized to endarterectomy vs. stenting, over median 2.5y f/u, there was no sig. diff. in incidence of (CVA, MI, or death) though stenting group had sig.  higher incidence of CVA (4.1% vs. 2.3%) and endarterectomy group had sig.  higher incidence of MI (2.% vs. 1.1%) (NEJM 363:11, 2010-JW)
  7. In a meta-analysis of 11 randomized controlled trials comparing carotid endarterectomy vs. carotid artery stenting involving a totla of 4.796 pts, the 30d weighted incidence of (death or CVA) was sig. lower with endarterectomy (5.4% vs. 7.3%); there was no sig. diff. in 30d mortality incidence.  Carotid endarterectomy was associated with higher incidence of periprocedural MI (2.6% vs. 0.9%) and facial nerve injury (7.5% vs. 0.45%).  There was no sig. diff. in "intermediate-term" (1-4y) incidence of (death or CVA) (BMJ 340:c467, 2010-AFP)
  8. In a meta-analysis of three randomized trials comparing carotid endovascular stenting vs. endarterectomy in 3,454 pts with moderate-to-severe carotid stenosis, in intent-to-treat analysis, 120d risk of (CVA or death) was sig. higher in stenting group (8.9% vs. 5.8%); subgroup analysis showed the diff. to be sig. in pts > 70yo but not in pts < 70yo. (Lancet 376:1062, 2010-JW)