Implications of a Reliable Method for Quantifying Brain Injury Associated with Repair of Carotid Artery Stenosis
================================================================================================================

* Dixon M. Moody

Surgical intervention involving the left-sided circulation anatomically or functionally proximal to the head is capable of inducing emboli to the brain. Müller et al (page 47) in this issue of the *AJNR* convincingly show new diffusion-weighted imaging lesions associated with carotid endarterectomy (CEA) in 26 of 77 surgeries. Most of these lesions represent infarcts arising from emboli, and most of these infarcts were clinically silent. The incidence of clinically evident perioperative neurologic morbidity for CEA in Müller's series (6%) is similar to that found in the NASCET trial (<7%). I agree with the authors that the incidence of new lesions detected with diffusion-weighted imaging is astonishingly high, but it is predictable. Anyone who has done transcranial Doppler sonography during CEA or cardiac surgery supported by cardiopulmonary bypass (CPB) knows that many emboli can be detected during these procedures. Characterizing the emboli, whether air or particulate, cannot be achieved by means of the sonographic device.

I believe that microembolic signals on Doppler sonograms are important. Müller and colleagues have confirmed in CEA patients what Stump et al (1) observed in CPB patients: greater numbers of detected emboli are associated with poorer outcomes (greater decrement on prospective neurobehavioral testing in Stump's series). Reducing the number of cerebral infarcts associated with CEA is a laudable objective, even if most of them are clinically silent.

In a previous editorial (2), I suggested there was a role for neuroradiologists in making cardiac surgery safer. Investigations have shown that standard T1- and T2-weighted MR images are not sufficiently sensitive for detecting many subtle lesions associated with CPB (2). I hoped that diffusion-weighted imaging, perfusion imaging, or both would be sufficiently sensitive to achieve this (2). Müller and colleagues have made an important contribution to the literature, and their findings need confirmation by researchers at other centers. In future investigations, MR imaging should be performed after diagnostic arteriography and before surgery, followed by MR imaging 1 to 2 days after surgery.

These findings suggest an opportunity for two further lines of investigation. The first is the testing of various neuroprotective interventions, the end point being determination of the number and volume of new lesions revealed by diffusion-weighted imaging. CEA patients are ideal for this type of study, more so than CPB patients, because the research subject has a large brain, the time of the insult is known, and the subject is not too sick for repeat imaging 24 hours after the insult. Approximately half of CPB patients have a temporary pacemaker, and CPB patients are often uncooperative for several days after surgery. If we anticipate that the tested pharmaceutical agent would reduce the incidence of lesions by 50%, and without intervention we expect a 33% incidence of detectable lesions, then a study group composed of 212 patients would give 80% power for rejecting the null hypothesis (that the intervention has no benefit). Therefore, the sensitivity of diffusion-weighted imaging is sufficient to derive meaningful information from a relatively small study sample.

A second, additional line of investigation relates to the considerable current interest in alternative approaches for the treatment of symptomatic and asymptomatic carotid stenosis. Cardiologists and interventional radiologists in some centers routinely employ endovascular stenting. Proponents of these innovative approaches should be encouraged to go head-to-head with surgeons employing the more conventional open-neck, open-vessel surgery in order to determine the safer treatment.

Neuroradiologists are well positioned to assume a pivotal role in such studies. My hope is that members of our profession can find the time and energy to assemble an appropriate team of collaborators and take a leadership position in these important activities.

## References

1.  Stump DA, Brown WR, Moody DM, et al. **Microemboli and neurologic dysfunction after cardiovascular surgery.** Seminars in Cardiothoracic and Vascular Anesthesia 1999;3:47-54
    
    [CrossRef](http://www.ajnr.org/lookup/external-ref?access_num=10.1177/108925329900300108&link_type=DOI) 

2.  Moody DM. **Is there a role for MR in the development of safer cardiac surgery?** AJNR Am J Neuroradiol 1996;17:213-215
    
    [Web of Science](http://www.ajnr.org/lookup/external-ref?access_num=A1996TW23400002&link_type=ISI) 

*   Copyright © American Society of Neuroradiology