Response to Dr. Spence

Asymptomatic Carotid Atherosclerosis Study (ACAS)

Of the 313 of 834 medically treated patients who had angiograms, only 4.2% had > 90% stenosis; only 28.1% had > 80% stenosis. From the surgical angiographic data, as many as 8% had < 60% stenosis. The numbers were likely the same for the medically treated patients who did not have angiograms. By any measure, the vast majority had mild or moderate stenosis (< 80%). In addition, 45 (5.4%) randomized to the medical arm actually had carotid surgery but were kept as part of the medical group in the intention-to-treat analysis. The reason for the operations was not explained, but one can safely assume that surgery was advised because of the severity of the stenosis. Were this the case, it is very possible that none of the > 90% stenoses were left for long-term follow-up and as few as 22.6% (189 of 834) of patients in the medical arm who were followed had > 80% stenosis. No matter how one interprets the data, the ACAS was primarily a study of moderate asymptomatic carotid stenosis. The numbers of patients with severe stenosis were small and not adequate to determine the true natural history of the lesion. From what we know today, if 73% of the patients had less than 80% stenosis, one would expect the rate of stroke to be very low.

Asymptomatic Carotid Surgery Trial (ACST)

Although a greater percentage of patients with high-grade stenosis (917 of 1,560) were in the “deferred surgery” arm, the classification of these patients was based on ultrasound criteria, not angiographic measurements, and many question whether this group was analogous to the ACAS high-grade stenosis population. As in the ACAS, 210 (14%) of those randomized to deferred surgery or the medical arm had CEA earlier than planned: 61 at the request of the patient or surgeon, 31 owing to a “change in the lesion,” and 80 because of TIAs. Because of the absence of angiographic measurements on the patients at the point of entry and the larger number of crossovers, it is difficult to know how many of the patients receiving medical therapy truly had severe (80%) stenosis.

In short, the absence of large numbers of individuals with critical stenosis of their carotid artery very likely led to (1) misinformation regarding the natural history of high-grade stenosis and (2) an underestimate of the overall risk of stroke by diluting the study cohort with individuals at lesser risk (< 80% stenosis). Unfortunately, the results of the ACAS and the ACST have been used by both aggressive interventionalists to justify procedures on patients with moderate stenosis (60–80%) and skeptics to avoid referring individuals with severe asymptomatic stenosis (80–99%) for prophylactic surgery.

The Real Issues Regarding Asymptomatic Carotid Disease

Is there a difference between 60 and 90% asymptomatic stenosis?

Of course there is! As every clinician knows, including Dr. Spence, who sees 900 such patients a year, if a patient presents with a TIA or stroke owing to carotid bifurcation disease, it is extraordinary to find the offending artery to be the one with the lesser degree of stenosis. Assuming that the anatomy was the same the day before the event, when the patient would have been classified as asymptomatic, if the severity did not matter, the chances that either plaque, moderate or severe, caused the symptom would be the same!

Supporting these observations, earlier natural history studies of > 75% stenosis (in the prestatin era) showed that the risk of stroke for those lesions was approximately 5% per year. ^7–9 Other studies have shown that individuals with moderate asymptomatic carotid artery stenosis (50–79%) have a very low risk of stroke, less than 1% per annum. ^7,10 Also, we know from pathologic studies that the more advanced plaques are likely to have ulcerations, intraluminal thrombus formation, and intraplaque hemorrhage, all related to stroke. ¹¹ Finally, it defies logic that the degree of stenosis becomes an important marker for future stroke only after a patient has experienced a TIA or stroke, not before.

Has the increased use of statins reduced the risk of stroke to a point that the beneficial effect of surgery no longer exists?

To support this claim, Dr. Spence cites several studies. One is a literature search that documents the decline in stroke over the past 25 years and attributes this to more aggressive medical therapy without any direct evidence to support the claim. ⁴ The other is a clinical study of 101 asymptomatic patients, all receiving optimal medical therapy. Only 31 patients followed long term had > 70% stenosis. Interestingly, 3 of the 31 (10%) patients had a TIA within 2 years! ⁵ Although Dr. Spence cites this as evidence, to their credit, the authors acknowledge that the study population was too small to draw meaningful conclusions. Other studies mentioned again define “severe stenosis” as anything > 50% reduction and are therefore biased by diluting the population at risk with individuals who are at little or no risk of stroke.

No doubt, better medical therapy will help reduce the risk of many types of stroke and, it is hoped, over time reduce the risk of plaque formation. As yet, however, there is no reliable evidence to support the claim that statins and antihypertensives will reduce the risk for those individuals who have severe carotid stenosis.

Another clinical trial?

It may be time to do a clinical trial to address the issue. Imagine where we would be today had Dr. Barnett and colleagues not organized the NASCET! If clinicians such as Dr. Spence have sufficient doubt regarding the effectiveness of CEA and/or CAS for the prevention of stroke in asymptomatic patients, then perhaps they should lead a modern-day natural history study of the disease without a surgical arm. Such a study, however, should focus on severe stenosis, not those with moderate disease (50–79%), and decide how to account for patients or surgeons who opt for surgery once enrolled.

Better markers?

Clearly, with over 700,000 new strokes every year in the United States, and a large portion of these being due to carotid bifurcation disease, there is a need to identify individuals at risk and treat them before they have a stroke. No doubt, using residual lumen measurements to predict which individuals are most likely to have a stroke is crude and overly sensitive. Unfortunately, it is the only fully validated marker we have at present. It is hoped that advances in imaging, biochemical analysis, genetic profiling, or other tests will help us better differentiate between those who are and those who are not at risk of stroke so that surgical therapy can be better focused.

Who among those with asymptomatic carotid disease should have carotid surgery and who should not?

As a starter, there is little justification for interventions on individuals with < 80% stenosis, or, more appropriately for today's clinical practice, an end-diastolic velocity of < 125 cm/s, unless there are other factors, such as a strong family history for stroke or rapid plaque progression. The risk of stroke for these individuals is < 1% per year. Most experienced clinicians, including the vast majority of vascular surgeons, would agree that these individuals are not at risk for stroke and therefore are not candidates for intervention. Management with aggressive medical therapy and careful monitoring for plaque progression is advised.

For those with > 80% asymptomatic stenosis or end-diastolic velocities > 125 cm/s, the risk of stroke probably increases to 3 to 5% per year. If the patient is in good health and if the procedure can be done with < 2% risk of stroke and death, surgery will be beneficial. Until clinical trials show that CAS can be performed in these individuals with acceptable risk, CEA is the only surgical treatment that can be supported with level I evidence.