Background: Independent audits have been proposed to improve carotid endarterectomy (CEA) effectiveness. This study used the online registry Modification of Outcomes by Lowering Ischemic Events after Reconstruction of Extracranial Vessels (MOLIERE) to evaluate the effectiveness of CEA in the Province of Quebec. The concept of MOLIERE is that surgeon involvement in a prospective manner is a prerequisite for them to evaluate, compare, and improve their practice.
Methods: All Quebec surgeons who performed CEA were invited to participate in this study sponsored by the Société des sciences vasculaires du Québec, the Canadian Society for Vascular Surgery and the Association des chirurgiens vasculaires du Québec. Surgeons prospectively entered data for 60 variables relevant to CEA in an online, secure, and confidential database between May 24, 2004, and May 31, 2005. Patient inclusion had to occur before surgery. After the study was completed, charts of all CEAs performed in each participating center were reviewed to validate the results of MOLIERE. For each participating institution, results of CEA that were not entered in the registry were also reviewed.
Results: A total of 279 patients (mean age, 69 years; range, 46-91 years) undergoing a CEA were enrolled in MOLIERE by 23 surgeons from 10 institutions in Quebec; 157 (56%) were symptomatic, and 122 (44%) were asymptomatic. Carotid endarterectomies were performed with patch angioplasty in 252 (89%), primary closure in 24 (9%), and by eversion in six (2%). Follow-up at 30 days was achieved for all patients. The 30-day stroke or death rates for symptomatic and asymptomatic patients were 3.2% (5 of 157, 95% confidence interval [CI], 1.2%-7.4%) and 0%. Validation was excellent for patients who were entered in the registry, with no additional deaths or strokes than those reported by the surgeons. The validation process revealed that participating surgeons entered 66% (279 of 424) of their patients in the registry. Indications and stroke or death rates (SDRs) for those patients who were not entered in the registry were not statistically different (symptomatic, 54% [79 of 145]; SDR of 1.3% [1 of 76] for symptomatic and 1.5% [1 of 66] for asymptomatic). In participating institutions, 11 surgeons did not participate. The SDRs for patients operated on by nonparticipating surgeons were higher but not statistically different than rates for patients operated on by participating surgeons (3.7% [5 of 136] vs 1.7% [7 of 424], P = .16). There was a trend toward higher stroke rate for patients operated on by nonparticipating surgeons (3.7% [5 of 136] vs 1.2% [5 of 424], P = .056). Mean postoperative length of stay was statistically higher for patients operated on by nonparticipating surgeons (4.7 vs 3.4 days, P = .046). The SDRs were adequate for all surgeons in participating centers, with 95% CI within accepted standards for symptomatic and asymptomatic patients.
Conclusion: MOLIERE is the first Canadian online prospective registry allowing surgeons to audit CEA results. The SDRs for participating surgeons were valid and within standards. Scientific vascular societies played a key role in supporting this project. Such audits allow surgeons and medical stroke experts to examine the appropriateness and results of CEAs in their institutions to improve them. The future of MOLIERE is in validation of its concept, increased participation by surgeons, and integration of a multidisciplinary approach.