London, United Kingdom 2013 4 Vascular surgery and neurosurgery 17:00 - 17:10 Is surgery getting safer? An analysis of changes in periprocedural risk of carotid endar-terectomy for symptomatic carotid stenosis over the last 30 years. F. Kennedy1, J. Dobson2, D. Doig3, R.L. Featherstone4, T. Richards5, M.M. Brown6 Institute of Neurology, University College London, London, UNITED KINGDOM1,De-partment of Medical Statistics, London School of Hygiene and Tropical Medicine, London, UNITED KINGDOM2, Institute of Neurology, University College London, London, UNITED KINGDOM3, Institute of Neurology, University College London, London, UNITED KING-DOM4, Institute of Neurology, University College London, London, UNITED KINGDOM5, In-stitute of Neurology, University College London, London, UNITED KINGDOM6 Background: Over the past 30 years several trials showed carotid endarterectomy (CEA) is an effective treatment for symptomatic carotid stenosis, but carries a risk of perioperative stroke or death. There have been improvements in surgical and anaesthetic pathways of care and in medical therapy over this time period. We therefore examined the time-dependent changes in perioperative risk in the trials to examine the hypothesis that risks have fallen over time.Meth-ods: We extracted baseline and early outcome data from all the large randomised clinical trials of CEA conducted between 1981-2010. We determined the rate of any stroke or death within 30 days of CEA for symptomatic stenosis in each trial ‘per-protocol’ and compared this with the mid-point year of recruitment for each trial. Meta-regression analysis was used to assess the strength of the association.Results: Data were available from 8 randomised trials recruiting patients between 1981-2008. Rates of 30-day stroke or death fell from 7% in ECST (mid-point year 1987) to 3.4% in ICSS (mid-point year 2005). Regression analysis showed a reduction of 0.19% (95% CI – 0.02% to 0.40%) in CEA risk per 1 year increase in mid-point year of recruit-ment, but this trend was only borderline significant (p= 0.07, see fig 1). After adjustment for the available baseline characteristics, there remained no statistically significant evidence of an as-sociation. However, there was a significant association between the proportion of patients with raised baseline cholesterol levels in the 6 trials with this data available and a lower risk from CEA (p=0.024, see fig 2).Conclusion: There has only been a modest reduction in the periopera-tive risk of CEA for symptomatic carotid stenosis over the past 30 years. The paradoxical asso-ciation between raised cholesterol and lower risk within the trials might reflect the fact that, in early trials, only patients with very high cholesterol levels received preoperative lipid lowering therapy. Cerebrovasc Dis 2013; 35 (suppl 3)1-854 63 3 Vascular surgery and neurosurgery 16:50 - 17:00 Proportion of patients treated by carotid endarterectomy or stenting who might have had as good an outcome with optimised medical treatment: a modelling study using the Carot-id Artery Risk Score R.L. Featherstone1, F. Kennedy2, D. Doig3, J. Dobson4, M.M. Brown5 ICSS Investigators UCL Institute of Neurology, London, UNITED KINGDOM1,UCL Institute of Neurology, London, UNITED KINGDOM2, UCL Institute of Neurology, London, UNITED KINGDOM3, London School of Hygiene and Tropical Medicine, London, UNITED KINGDOM4, UCL Insti-tute of Neurology, London, UNITED KINGDOM5 Background: Randomised trials established the benefit of carotid endarterectomy (CEA) but these were conducted more than 20 years ago. Medical therapy has improved and CEA may no longer be beneficial in many patients with carotid stenosis. We therefore modeled the like-ly benefits of optimized medical therapy (OMT) using data from a recent randomised trial to determine the proportion of patients who might do as well with OMT.Methods: The baseline risk factors for individual patients with symptomatic carotid stenosis randomised in the Inter-national Carotid Stenting Study (ICSS) were used to calculate their Carotid Artery Risk Score (CAR) i.e. the 5-year risk of ipsilateral stroke, using a predictive algorithm developed in ECST and tested in NASCET. We recalibrated the algorithm to account for the benefit of OMT. Data on plaque morphology was not recorded and therefore we assumed smooth morphology in all patients. We compared CAR scores with actual rates of events. We also calculated CAR assum-ing all patients had irregular or ulcerated plaque. Results: Of 1710 patients randomised in ICSS, 1649 had sufficient baseline medical data to calculate the CAR. Assuming a smooth plaque, 1333 of 1649 (81%) patients had a CAR of 15% or less. In the patients treated by CEA, the mean predicted risk in patients with a CAR of 15% or less was similar to the observed 5-year rate of ipsilateral stroke or perioperative stroke/death (8.6% vs. 6.8% respectively, p=0.27). In patients with a CAR >15%, the predicted risk was significantly higher than observed risk (18.9% vs. 10.3%, p=0.05). If the plaque was assumed to be rough or ulcerated in all cases, 436 of 1639 (26%) had a CAR of 15% or less.Conclusion: As many as 80% of patients referred for CEA might have had as good an outcome if they had received OMT alone. Plaque morphology appears to be a significant factor in determining risk of recurrent stroke and should be incorpo-rated into the assessment of carotid stenosis prior to treatment.
Karger_ESC London_2013
To see the actual publication please follow the link above