September 9th, 2011
Asymptomatic Carotid Stenosis: Medical Management or Revascularization?
Allan Brett, MD
Every so often, one of my patients with no history of stroke or transient ischemic attack asks my opinion on what to do about a carotid ultrasound — not ordered by me — showing high-grade stenosis. I am a general internist, and I don’t obtain these studies in patients without a history of cerebrovascular symptoms. Guidelines don’t endorse carotid screening in such patients, yet they end up getting screened in various ways: Some cardiologists and vascular surgeons routinely get carotid studies in their patients with coronary disease; some clinicians order ultrasound for patients with asymptomatic carotid bruits; some clinicians include routine carotid ultrasound inappropriately in their “syncope workup”; and companies offer direct-to-consumer ultrasound screening. Because many of these patients are told “you need surgery” or “you need stenting” as if there’s no choice in the matter, a brief look at asymptomatic carotid stenosis is warranted.
Two large randomized trials compared carotid endarterectomy (CEA) and medical therapy in patients with asymptomatic carotid stenosis — a North American study published in 1995 (JAMA 1995; 273:1421) and a European trial published in 2004 (Lancet 2004; 363:1491). In both trials, the 5-year risk for stroke (including perioperative stroke or death) was significantly lower with CEA than with medical therapy, but the difference was only about 5 percentage points (5%–6% vs. 11%–12%), and no benefit was seen in women (Curr Opin Neurol 2007; 20:58). Given the 2% to 3% rate of perioperative stroke or death, it took several years for the benefit of CEA to clearly surpass that of medical therapy.
Because medical therapy has improved since these trials were conducted, researchers have examined whether stroke rates in patients with asymptomatic carotid stenosis have declined during the past decade. In fact, rates have fallen to around 1% annually in medically treated patients (Stroke 2010; 41:e11 and Stroke 2009; 40:e573). Thus, we must ask whether CEA has any role in patients with asymptomatic carotid stenosis. Recently, researchers have proposed several imaging findings that might identify high-risk subgroups — plaque echolucency, plaque ulceration, and embolic signals on transcranial Doppler ultrasound of the ipsilateral middle cerebral artery.
In one study of 435 patients with asymptomatic carotid stenosis (>70% stenosis by ultrasound), only 10 patients (2%) had strokes during an average follow-up of 2 years (Neurology 2011; 77:751). However, 4 of these strokes occurred among the 27 patients with both echolucent plaque and embolic signals (15% stroke rate). In contrast, only 1.5% of patients without these 2 findings had strokes.
In another study of 253 patients with asymptomatic carotid stenosis (>60% stenosis by ultrasound), only 6 patients (2.5%) had strokes during an average follow-up of 3 years (Neurology 2011; 77:744). Three of these strokes occurred in the 42 patients with at least 2 carotid ulcerations (7%); in contrast, the stroke rate was only 1.4% in the 211 patients with one or no ulcers. In addition, the stroke rate was 13% in patients with embolic signals (2 of 15 patients) but only 1.7% in those without embolic signals (4 of 238 patients).
The most striking aspect of these 2 studies is their confirmation of a very low overall incidence of stroke — about 1% per year. Thus, many asymptomatic patients who now undergo CEA (or carotid stenting, which is not safer than CEA) are likely risking harm without commensurate benefit. The use of embolic signals and plaque characteristics to identify candidates for CEA is promising but requires larger numbers and assurance that the techniques are reliable in community settings. Editorialists argue for “intensified medical management rather than revascularization procedures in patients with asymptomatic carotid stenosis,” until strategies to identify high-risk patients have been thoroughly investigated (Neurology 2011; 77:710). I find their position, which happens to reinforce my previous thinking on this topic, to be compelling. What’s your point of view?
13 Responses to “Asymptomatic Carotid Stenosis: Medical Management or Revascularization?”
NEJM Journal Watch — Recent Cardiology Articles- Persistent Symptoms After Valve Replacement for Severe Aortic Stenosis
- Is Diltiazem Safe in Patients with AF Taking Apixaban or Rivaroxaban?
- Does Lowering High Childhood Lipid Levels Prevent Adverse Cardiovascular Events in Adulthood?
- Self-Expanding Supraannular Valve vs. Balloon-Expandable Valve in TAVR for Small Aortic Annulus
- A New TAVR Prosthesis for Aortic Regurgitation Shows Promise
I agree nearly completely. There is also some data with FDG scanning to predict plaque vulnerability and MRI plaque characteristics as well. Not ready for prime time. The revascularization studies generally do not account for the fact that a procedural stroke occurs earlier than a stroke on medical therapy.
Regarding carotid ultrasounds in general, I do have some patients with identified stenosis >50% or raised plaque (we dont do CIMT) who are adherent with aggressive medical therapy solely due to these findings.
There are some cases where it is hard to avoid an urge to revascularize…very tight lesions and/or bilateral stenoses.
We do a follow-up of large groups of patients especially in large companies as part of occupational health assessment. We do do both duplex and doppler measurements. Assessment of those showed progression in almost all cases when no intervention was done. Our group came to the conclusion that an operation would be the best procedure as those who had refused operation had 5 times more likely that the average normal population on a stroke. In our hands we feel that cimt plus doppler assessment are a great tool to predict the likelyhood of both a stroke or a heart attack. Other studies have confirmed our findings like Hodis and coworkers at USC.
Dr.Brett pointed out a very nice subject. A common problem seen in clinical practice, summed up quite well. In my opinion, much needs to consider before the interventional procedure in patients with asymptomatic carotid stenosis.
Competing interests pertaining specifically to this post, comment, or both:
None
Such patients mandate close surveillance. They are more likely to sustain a myocardial infarction than they are a carotid-related stroke. In addition, revascularization will only prevent one particular cause of future stroke from one particular arterial segment, and do nothing to prevent all the other heterogeneous etiologies of future strokes (embolic, lacunar, contralateral circulation, posterior circulation, etc). I would follow such patients very closely to ensure plaque regression over time.
Competing interests pertaining specifically to this post, comment, or both:
None.
Appendum: The above statistic comes from John Norris’s classic study published in Arch Neurol in 2002 (pretty the pre-statin era for these patients). Here is the abstract:
rch Neurol. 2002 Jul;59(7):1162-6.
Long-term risk of stroke and other vascular events in patients with asymptomatic carotid artery stenosis.
Nadareishvili ZG, Rothwell PM, Beletsky V, Pagniello A, Norris JW.
Source
Stroke Research Unit, Sunnybrook and Women’s College Health Sciences Centre, University of Toronto, Ontario, Canada. nadareishviliz@ninds.nih.gov
Abstract
CONTEXT:
The annual risk of ischemic stroke in patients with asymptomatic carotid artery stenosis is about 2% during the short-term (2-3 years), but the long-term risks of stroke and other vascular events are unknown, although they may affect surgical decision making.
OBJECTIVE:
To evaluate the long-term risk of stroke and other vascular events in patients with asymptomatic carotid artery stenosis.
DESIGN:
Cohort study with a median follow-up of 10 years (range, 5-18 years).
SETTING:
The teaching hospital of the University of Toronto, Toronto, Ontario.
PATIENTS:
From the initial cohort of 500 patients, 106 patients with asymptomatic carotid artery stenosis were selected because they had completed at least 5 years of follow-up.
MAIN OUTCOME MEASURES:
Ipsilateral stroke, myocardial infarction, and nonstroke vascular death.
RESULTS:
The 10- and 15-year actuarial risks of ipsilateral stroke were 5.7% (95% confidence interval [CI], 0%-12%) and 8.7% (95% CI, 1%-17%), respectively, in patients with 0% to 49% internal carotid artery stenosis, and 9.3% (95% CI, 1%-18%) and 16.6% (95% CI, 1%-32%) in patients with 50% to 99% internal carotid artery stenosis. The 10- and 15-year risks of myocardial infarction and nonstroke vascular death were 10.1% (95% CI, 4%-16%) and 24.0% (95% CI, 14%-34%). Age (P =.02), diabetes mellitus (P =.02), and internal carotid artery stenosis of 50% or more (P =.04) were predictive of increased risks of myocardial infarction and nonstroke vascular death. Internal carotid artery stenosis of 50% or more did predict the risk of ipsilateral stroke (P =.003) when all 181 asymptomatic carotid arteries were included.
CONCLUSIONS:
The annual stroke risk in patients with asymptomatic carotid artery stenosis was low and remained stable during long-term follow-up. Any benefit from carotid surgery is therefore unlikely to increase significantly with long-term follow-up. The high long-term risks of myocardial infarction and nonstroke vascular death suggest that prevention strategies should concentrate on coronary risk more than stroke risk.
Plaque characteristics, and embolic signals are very important for identification a vulnerable plaques and high risk subgroup ptns with Asymptomatic CA stenosis,and i think Asymptomatic bilateral CAs stenosis need close observation with very strict medical Antiplatelet and Statins therapy…?
I have trouble reconciling these two statements regarding carotid atherosclerosis.
“I don’t obtain these studies in patients without a history of cerebrovascular symptoms. Guidelines don’t endorse carotid screening in such patients”
“Because medical therapy has improved since these trials were conducted, researchers have examined whether stroke rates in patients with asymptomatic carotid stenosis have declined during the past decade. In fact, rates have fallen to around 1% annually in medically treated patients (Stroke 2010; 41:e11 and Stroke 2009; 40:e573).”
You are saying that we can significantly reduce event rates by treating asymptomatic atherosclerosis medically, yet you do not believe that we should screen for this disease that kills 10% of Americans. I have great difficulty reconciling these two statements.
The National Preventive Task Force determined that screening for asymptomatic carotid stenosis was not worth doing because of the very marginal value of surgical intervention in these subjects. They did not in any way factor the potential benefit of medical management.
If carotid ultrasound can identify subjects with disease who could benefit from improved medical management, why are you opposed to doing this?
Competing interests pertaining specifically to this post, comment, or both:
I have a practice that uses EBT calcium imaging and carotid ultrasound to screen for heart attack and stroke risk. I treat those subjects medically to a goal of stable plaque. I see very very few heart attacks or strokes and have seen no MI or stroke death for over 7 years.
This is another classic example of doing a unnecessary test in asymptomatic patients, and then landing in a clinical dilemma !Once you get to see a significant stenosis, the chances of an unnecessary intervention are very high with marginal, if any, benefits.It really pays, if one follows the simple principle of treating all patients holistically and aggressively addressing their risk factors.
Competing interests pertaining specifically to this post, comment, or both:
None.
Dr. Blanchet has trouble reconciling two of my comments: First, I questioned the widespread performance of carotid endarterectomy in patients with asymptomatic carotid stenosis, and second, I questioned screening for asymptomatic carotid stenosis. By inference, Dr. Blanchet apparently believes that routine carotid screening is warranted, regardless of one’s position on the role of surgery.
If one accepts the premise that surgery is no longer warranted for asymptomatic stenosis, I can think of only two possible reasons to screen. Carotid screening would be reasonable (1) if the incremental information gained by screening would change our management of the patient’s risk factors, with demonstrable improvement in hard clinical outcomes, or (2) if asymptomatic patients who knew they had stenosis would have increased motivation to adhere to medical management, with demonstrable improvement in hard clinical outcomes. But I’m not aware of data from controlled trials to support either of these reasons to screen. Thus, widespread carotid screening cannot be endorsed at the present time. Obviously, these views are open to revision if controlled clinical trials eventually show that screening (perhaps augmented by knowledge of plaque characteristics or embolic signals) results in improved clinical outcomes.
Competing interests pertaining specifically to this post, comment, or both:
None
Well said, Dr. Brett
Competing interests pertaining specifically to this post, comment, or both:
None
Dr Brett,
I accept the premise that surgery is not recommended in asymptomatic patients.
Responding to your two possible justifications of carotid ultrasound screening in asymptomatic patients:
Does atherosclerosis imaging improve stratification of vascular risk?
The MESA heart study demonstrated that carotid ultrasound was dramatically stronger at predicting stroke than all risk factors combined. Coronary calcium >300 was also found to be 10 times more predictive of MI than all conventional risk factors combined. So yes, the DATA is there, imaging plaque is strikingly more predictive of events than risk factors.
MESA found that of non-diabetic women under the age of 70, 90% were considered low risk. However 32% of these women had coronary calcium and a 6.5 HR for CHD compared to a CAC score of 0. Put another way, traditional risk factors found 1/3rd of the women at risk while coronary calcium found over 97% of these women at risk.
Does the identification of plaque with atherosclerosis imaging improve patient compliance with medical intervention?
Taylor AJ. American Heart Association 2006 Scientific Sessions; November 12-15, 2006; Chicago, IL demonstrated a dramatic improvement in compliance with ASA and statin based upon detection of asymptomatic sub-clinical plaque with coronary calcium imaging.
Kalia et al Atherosclerosis 2006; 185:394-399 demonstrated that of patients in the top quartile of risk by CAC imaging, 91% of subjects were compliant with statin use after 3 years. This is dramatically better than statin compliance reported in any other study and dramatically better than statin compliance in subjects in this study with less coronary plaque by EBT imaging.
Now I realize that the studies that best show improved motivation of subjects were involving coronary calcium imaging, there is no reason to believe that demonstrating plaque by carotid imaging should not have similar benefit in motivating changes.
Competing interests pertaining specifically to this post, comment, or both:
I prefer to be at home with my family at night rather than admitting patients to the hospital with heart attacks and strokes. I find atherosclerosis imaging to be necessary to accurately identify individuals at risk and to motivate compliance with medication and lifestyle modification needed to prevent heart attacks and strokes.
The commercialized ” carotid screens” are not for CIMT (which has its own issues: quality and consistency most notably). ARIC showed that properties of the vascular wall (not luminal narrowing) reclassified Framingham risk for intermediate patients. But this was CIMT and plaque, whose definition varies and also requires the expertise and equipment used for CIMT.
There is a” recommendation” that a carotid stenosis >50% should be a” CAD risk equivalent”…like diabetes (although that is controversial, particularly young new onset diabetics). Although reasonable for a known >50 % stenosis, screening for this seems rediculous.
As for showing that screening and risk reclassification affects outcomes…no data for any imaging.
Shaw et al have estimated that you would need to enroll 100,000 patients into a trial of imaging vs guideline-based care to show a mortality reduction of 10%. This is based on the assumption that you have a therapy that reduces mortality by 10% (say statins), and your imaging technology can increase the frequency of prescribing of that intervention in your imaging group by 10%. This would be equivalent to a 60/40 split in statin prescribing.
Of course there are other therapies such aspirin, niacin, omega-3 fatty acids but they tend not to have the mortality track record of statins. I do not think anyone is going to fund a 5-year 100,000 patient study of imaging vs no imaging any time soon. The ability to identify and then treat subclinical vascular disease with multi-modality anti-atherosclerotic therapy is quite clear to anyone who has ever looked at a 3-dimensional ultrasound image of the carotid arteries and uncovered a heckuva lot of plaque in someone without conventional Framingham risk factors. This is where you identify the patient who fits BETWEEN ‘primary’ and ‘secondary’ prevention (which I agree is a wholely false dichotomization). This is where the rubber meets the road.