Thromboembolic events after carotid surgery : causes and solutions

Introduction: Carotid endarterectomy (CEA) for patients with a significant (>70%) carotid stenosis reduces their risk of stroke over the subsequent years of follow up. This benefit is ameliorated by post-operative strokes and deaths. If the post-operative complication rate can be reduced then it is possible that patients with lesser stenoses may also benefit from CEA. This body of work has concentrated on identifying factors related to the development of strokes secondary to carotid thrombosis following CEA. The number of microembolic signals detected by transcranial Doppler ultrasound arising from the endarterectomised vessel wall has been shown in a number of publications to correlate with risk of post-operative carotid thrombosis. Methods and Results: 1)Patients who undergo two carotid endarterectomies at separate time points have a significant correlation between the number of post-operative emboli seen at the first and second procedures (p=0.038). 2) A randomised controlled trial of 274 patients undergoing closure of the arteriotomy site with either a prosthetic Dacron patch or a vein patch was performed. This showed that whilst the vein patch had fewer emboli overall (median of 3 vs 5, p=0.028), the number of people having high levels of sustained embolisation or needing post-operative Dextran-40 therapy (to prevent carotid thrombosis) was not different between the 2 groups (p=0.62). 3) Laboratory platelet function studies (flow cytometry and aggregometry) showed that patients at increased risk of post-operative carotid thrombosis had platelets that were significantly more responsive to the physiological agonist, ADP (pO.OOOl for flow cytometry and p=0.0012 for aggregometry). The degree of inhibition by aspirin did not have a significant effect. 4) Dextran-40, an effective drug for the prevention of carotid thrombosis, was found to bind to the surface of platelets in a dose dependent manner (p=0.006). Unexpectedly, Dextran-40 increased the rate at which platelets aggregated together following stimulation with ADP (p=0.047). Activated platelets bound significantly more Dextran-40 than resting platelets (p<0.0002). Dextran-40 failed to cause disaggregation of previously aggregated platelets. 5) A platelet receptor polymorphism of the fibrinogen binding site (HPA-3) was found to be associated with both increased pre- and post-operative emboli (p=0.03 and p=0.04 respectively). This polymorphism was also associated with increased fibrinogen binding (p=0.024) and decreased post-operative blood loss (p<0.001). The met+ allele of the collagen receptor (HPA-2) was significantly associated with post-operative restenosis (p=0.01). 6) Studying patient risk factors for post-operative carotid thrombosis identified that women (who are at significantly increased risk of carotid thrombosis) had over 3 times as many emboli as men (p=0.004). Patients who embolised pre-operatively were more likely to do so in the post-operative period (p=0.027). No other risk factors were associated with emboli counts. Patients who had their operation in the morning were significantly more likely to develop high numbers of emboli relative to those performed in the afternoon (p=0.004). Conclusion: The findings of this body of work support the hypothesis that it is the patient who is prothrombotic rather than the procedure per se. It has also identified a number of potential targets and therapeutic strategies that may help to reduce the number of postoperative emboli seen after CEA, and thereby reduce patients? risk of thrombotic stroke. The findings of this study may well have relevance to other areas of cardiovascular intervention.