Discussion

Article Outline

• Copyright

Dr Pevec (Sacramento, Calif). Captain Arthurs and colleagues have presented a prospective observational study evaluating the association of C-reactive protein with the progression of carotid stenosis. This study, including 271 patients over 3 years, provides valuable information on the association of CRP and inflammatory markers and the progression of atherosclerosis. The authors are to be commended for conducting this longitudinal study. The authors suggest that elevated CRP levels predict progression of carotid artery stenosis and that the measurement of CRP in patients with carotid artery stenosis may be of some clinical benefit.

I have several questions. Patients with CRP in the highest quartile were found to be at higher risk for progression of carotid artery stenosis. The CRP quartiles were assigned retrospectively by taking the average of the CRP measurements obtained from each study patient every 6 months. While this is useful in helping to understand the relationship of CRP with progression of carotid stenosis in general, looking retrospectively at an average of values measured over several years, there is not a benefit in predicting future progression of stenosis in an individual patient. Is there a single cutoff value of CRP that can be used to predict progression of carotid stenosis? Can the authors provide positive- and negative-predictive values for some level of CRP, for instance, greater than 6 mg/dL, which was their level in the fourth quartile?

The authors defined progression of carotid stenosis in two ways. Type A included progression from any category of stenosis to a higher category of stenosis. Type B included only stenosis that progressed to 50% or greater or from 50% to 79% to a greater stenosis. The authors do not state how they define the categories of stenosis, but as the authors are from Washington, one can assume they use the Strandness criteria. If this is true, progression from 0% to 15% to 16% to 49% stenosis is based upon subjective criteria—plaque appearance and presence of spectral broadening—while progression to a greater than 50% stenosis is based on objective velocity criteria. While type B progression based on objective duplex velocity measurements correlated with elevated CRP levels on univariate analysis, this correlation was not significant on multivariate analysis. Only when progression based upon subjective duplex assessment was included, so-called type A progression, were elevated CRP levels predictive on multivariate analysis. A full 41% of study patients had 0% to 15% carotid stenosis at entry. Lack of independent correlation of elevated CRP with objective progression of carotid stenosis seems to greatly weaken the importance of these findings. Can the authors comment on this please?

Finally, is the measurement of CRP of clinical benefit in the management of patients with carotid stenosis? In the paper the authors suggest that decreasing inflammation with aspirin and/or statins may be of benefit in patients with carotid stenosis and elevated CRP. Shouldn’t all patients with carotid stenosis be on aspirin and a statin drug?

I would like to compliment the authors on a well-performed and well-presented study that increases our understanding of the influence of inflammation on progression of atherosclerosis. I would like to thank the authors for providing me with a copy of their manuscript in advance of the meeting, and I would like to thank the program committee for the privilege of discussing this paper. Thank you.

Dr Arthurs. I would like to thank Dr Pevec for his insightful comments. I will start with his second question first as it addresses the matter of our design of the study and the difficulty in doing so from the onset. It comes down to the subjective nature of mild vs moderate disease on duplex and whether that plays a part in the study. It absolutely does. It is extremely subjective, and we attempted to control for that variability. That is why we excluded all patients with less than three examinations, so the patient either had two normal examinations with one abnormal examination on the third study, or the patient had a normal study with two subsequent examinations read out as moderate by the same two vascular technicians. In addition, during this time, the same two vascular staff surgeons read the examinations. Because our ultimate goal with this study was observational evaluation of progression of patients with subclinical disease, we felt that more sensitive imaging modalities would not justify the cost of performing examinations at 6-month intervals.

His second question—ultimately is there a cutoff value for CRP? I think if we look at where the coronary literature has utilized CRP, the CRP values that are abnormal are not far from the definitions utilized for abnormal CRP. In our study, 6 is the cutoff, but ultimately, that just applies to this cohort. I think the message is that we were not trying to identify patients that had high-grade stenosis. We were looking at that population of patients that either came with no stenosis or had a minimal stenosis with minimal to no risk factors. From our study, CRP may be most applicable in those patients without any risk factors and minimal carotid disease who have a CRP level above 6. They appear to be at higher risk for progression of disease based on our findings. We are trying to quantify the idea that there is inflammation as a marker of overall disease risk or patient vulnerability.

That rolls into the third question of shouldn’t all patients who have carotid artery stenosis be on a statin or aspirin. Absolutely. However, a patient has no risk factors, no lesion on examination, but comes into our office and has a CRP level that is persistently elevated, whether it be 4 mg/dL or 6 mg/dL, should there be different goals in medical treatment for that patient? We can’t answer that question with this study, but based on our study, those patients are at a higher risk of carotid stenosis progression. Finally, if a patient has a normal LDL and an elevated CRP, should you treat that patient with a statin or some other anti-inflammatory modality? Thank you.