Journal of Clinical Neuroscience (1998) 5(4), 417-420 © Harcourt Brace & Co. Ltd 1998 Clinical studies An audit of carotid endarterectomy for internal carotid artery disease at Dunedin Hospital C. R. P. Lind 1, S. N. Bishara 1 MCh FRCS FRACS, A. M. van RJj1 BMedSc MBChB MD FRACS, J. Fulton 2 MBChB FRACR Departments of ~Surgeryand ~Radiology,Dunedin School of Medicine, University of Otago, 201 Great King Street, DuneNn, New Zealand Summary Carotid endarterectomy, mostly for symptomatic internal carotid artery stenosis, has been successfully performed in both the vascular and neurological surgery units at Dunedin Hospital. This study was performed to compare our results with those of the North American Symptomatic Carotid Endarterectomy Trial and the European Carotid Surgery Trial. The 30-day perioperative morbidity and mortality rate was 4.3% (1.2% < 95% CI > 7.4%) which compares favourably with an estimated upper limit of 5.5% based on recent trial reports. The present study highlights the difficulty in modelling local clinical practice on results of major trials when standards of patient evaluation and surgical skill may differ from those of the large studies. To justify generalization of indications for intervention based on the multicentre trials, there, must be continual monitoring of local surgical results, and standardized use of diagnostic investigations. Keywords: carotid endarterectomy, stroke prevention, carotid stenosis, cerebral infarction INTRODUCTION The first carotid endarterectomies (CEAs) were performed in the mid 1950s, 1'2 and by 1959 when the first randomized controlled trial (The Joint Study of Extracranial Arterial Occlusion) was initiated, the operation was favoured as a treatment for carotid stenosis) The results of the study were published in 1970, and showed CEA to have no significant advantage over conservative medical treatment, but this had little effect on the widespread use of the procedure. 4 In the mid-1980s it became apparent that there was wide variability in the success of CEA in different centres. This led to further prospective trials of variable quality. 5 Since then, several well-designed prospective multi-centre trials have been conducted to determine acceptable indications for CEA. Most notable are the European Carotid Surgery Trial (ECST) 6 and the Noah American Symptomatic Carotid Endarterectomy Trial (NASCET), 7'8 which have unequivocally shown CEA to be the treatment of choice for symptomatic patients with high-grade internal carotid artery (ICA) stenosis, provided combined peri- operative neurological morbidity and all-cause mortality rates do not exceed 5.8% (NASCET; 3.3% < 95% CI > 8.3%) or 7.5% (ECST; 5.0% < 95% CI > 10%). Unfortunately a different method for the angiographic measurement of stenosis was used in the ECST than was used in the NASCET (Fig. 1A). The ECST has shown benefit for symptomatic patients with stenosis of greater than 50% if the measurements are converted to NASCET by means of a very approximate linear formula (Fig. 1B). 9 However, this conversion is problematic and does not reliably overcome the difficulty created by use of two different methods of measurement from angiograms. The NASCET showed benefit for patients with at least 70% stenosis. The ECST has also shown that there is no benefit in offering CEA to patients with less than 50% (after formulaic conversion to NASCET) stenosis. 6'1° The NASCET has not yet accumulated enough data to comment conclusively on the 30-69% stenosis bracket, but has shown there is no benefit at less than 30% stenosis. Received 19 February 1997 Accepted 17 March 1997 Correspondence to: C. Lind, Tel: 64 3 474 7952, Fax: 64 3 474 7679, E-mail: crplind @voyager.co.nz In the light of results from the multicentre trials, it has remained for individual centres to conduct their own audits to ensure their surgeons' perioperative morbidity and mortality rates are sufficiently low, and that candidates for surgery meet the specified clinical and radiological criteria. This study was performed to evaluate local surgical results and to assess the need for changes in patient selection. PATIENTS AND METHODS Hospital and surgical unit files were retrospectively evaluated. The audit included all patients undergoing operations between January 1985 and October 1995 in the Neurosurgery Unit; and January 1987 and October 1995 in the Vascular Surgery Unit. The few cases handled by the cardiovascular surgeons involved concurrent cardiac surgery, and were not included in this study. Five out of the 154 CEA recipients were excluded because their files were unavailable. It is known that none of these patients experienced significant morbidity or mortality in the perioperative period. One hundred and sixty-four operations on 149 patients are included in this audit; of the 15 second procedures, 14 were performed on contralateral internal carotid arteries (ICAs), and one was for ipsilateral recurrence. Seventy-nine (48%) CEAs were done in the neurosurgery unit (S.N.B.); and the remaining 85 (52%) were performed by the vascular surgeons. RESULTS Patient selection The Dunedin patient mix was similar to that of both the ECST and NASCET in that the mean age of patients was in the 7th decade (67, 62 and 66 years, respectively), and there were more males than females undergoing CEA. The prevalence of hypertension was similar to that of the NASCET, and that of diabetes mellitus and tobacco use fell between those of the NASCET and ECST. Plasma lipid levels were not recorded in the majority of Dunedin case notes, and the rate of ischaemic heart disease was nearly twice that of the ECST; the NASCET's different classification precluded comparison (Table 1). Ninety-four per cent of the Dunedin operations, as compared to 100% in the NASCET and ECST, were performed on symptomatic 417