HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Correspondence to:

ARTICLES: Dean C.C. Johnston and Larry B. Goldstein Clinical carotid endarterectomy decision making: Noninvasive vascular imaging versus angiography Neurology 2001; 56: 1009-1015 [Abstract] [Full text] [PDF]

Correspondence: Submit a response to this article

Correspondence published:

Clinical carotid endarterectomy decision making: Noninvasive vascular imaging versus angiography

G Young, P Humphrey   (20 August 2001)

Reply to Drs Young and Humphrey

D Johnson, L Goldstein   (20 August 2001)

Clinical carotid endarterectomy decision making: Noninvasive vascular imaging versus angiography 20 August 2001
G Young South Tees Hospitals Middesbrough Cleveland OH, P Humphrey Send Correspondence to journal: Re: Clinical carotid endarterectomy decision making: Noninvasive vascular imaging versus angiography sandi_moriarity{at}urmc.rochester.edu G Young, et al.	The recent article by Drs. Johnston and Goldstein [1] adds to the studies comparing “non-invasive” and “invasive” methods of assessing carotid stenosis. Unfortunately, we believe there are too many flaws in this article to contribute to the debate. Obvious problems relate to the retrospective design, and therefore, non-standardized and non-blinded performance and reporting of the tests. Whether “tandem lesions” should be a contraindication to endarterectomy, and whether the distinction between critical stenosis and occlusion is as important as is suggested demands further scrutiny, given recent evidence from the endarterectomy trials. [2, 3] Perhaps the most important omission from this article, and from the accompanying Editorial [4], relates to observer variability in reporting. When two individuals report a test they will disagree in a proportion of cases. When reporting carotid stenosis, the rate of disagreement will in part depend upon what proportion of individuals studied have stenosis close to the “cut-off” chosen. Therefore, the degree of observer variability is specific to the patient population studied and should be quoted for any method comparison study. When observer variability is considered, it is clear that it accounts for a large proportion of the “disagreements” between different methods of measurement. We found in two separate studies that observer variation in reporting the same conventional catheter angiograms can result in “surgically-significant” disagreements in 3.4% to 7.3% and 3.8% to 12.4% of vessels studied. [5, 6] Compare this figure to the “misclassification rate” of 7.9% reported by Johnston and Goldstein when concordant MRA and ultrasound were compared to catheter angiography. Whatever technique is chosen to measure carotid stenosis, “inappropriate” decisions concerning surgery will be made due to the inevitable differences that occur between individuals (or on different occasions by the same individual) when the measurement is made. Given that there are disagreements when the same technique is assessed on two occasions, it is obvious that comparing any alternative technique will also inevitably produce disagreements, a fact, which should not be as surprising as suggested in the accompanying editorial. [4] In the carotid imaging literature, there tends to be polarization of views with recommendations to rely solely on catheter angiography or solely on non-invasive techniques. Perhaps the pragmatic approach to adopt would be to suggest performing the safest test that gives a reliable answer. In most cases it is possible to identify appropriate individuals with complete confidence by non-invasive techniques, while on some occasions, when non-invasive imaging gives uncertain results, catheter angiography should be performed. References: 1) Johnston DCC, Goldstein LB. Clinical carotid endarterectomy decision- making. Non-invasive vascular imaging versus angiography. Neurology 2001;56:1009-1015. 2) Kappelle LJ, Eliasziw M, Fox AJ, Sharpe BL, Barnett HJM; for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Importance of intracranial atherosclerotic disease in patients with symptomatic stenosis of the internal carotid artery. Stroke 1999;30:282- 286. 3) Rothwell PM, Warlow CP; on behalf of the European Carotid Surgery Trialists’ Collaborative Group. Low risk of ischaemic stroke in patients with reduced internal carotid artery lumen diameter distal to severe symptomatic carotid stenosis: Cerebral protection due to low poststenotic flow? Stroke 2000;31:615-621. 4) Norris JW, Rothwell PM. Noninvasive carotid imaging to select patients for endarterectomy. Neurology 2001;56:990-991. 5) Young GR, Sandercock PAG, Slattery J, Humphrey PRD, Smith ETS, Brock L. Observer variation in the interpretation of intra-arterial angiograms and the risk of inappropriate decisions about carotid endarterecomy. Journal of Neurology, Neurosurgery, and Psychiatry 1996;60:152-157. 6) Young GR, Humphrey PRD, Nixon TE, Smith ETS. Variability in the measurement of extracranial internal carotid artery stenosis as displayed by both digital subtraction and magnetic resonance angiography: An assessment of three calliper techniques as well as the visual impression of stenosis. Stroke 1996;27:467-473.
Reply to Drs Young and Humphrey 20 August 2001
D Johnson Duke University Medical Center Durhan NC, L Goldstein Send Correspondence to journal: Re: Reply to Drs Young and Humphrey djohnstonO508{at}home.com D Johnson, et al.	We thank Drs. Young and Humphrey for their comments and for the opportunity to clarify the points made in our article. [1] The purpose of our study was to evaluate the performance characteristics of noninvasive vascular imaging as reported by experienced radiologists in routine clinical practice in patients being considered for carotid endarterectomy. We sought to determine how frequently decisions to proceed with or defer carotid endarterectomy would be in error using criteria established based on results of randomized, controlled trials. Although a post hoc analysis of data from the NASCET trial suggests that symptomatic patients with mild-moderate intracranial carotid artery stenosis also benefit from carotid endarterectomy [2], patients with severe intracranial stenosis were excluded from NASCET. [6] We considered an intracranial stenosis of >50% as a contraindication to endarterectomy. The cited paper by Rothwell and Warlow [3] does not address the role of surgery in patients with complete carotid occlusion, which was also a contraindication for carotid endarterectomy in the randomized, controlled trials. However, it should be noted that we found 96% of the misclassifications were related to the measurement of stenosis of the extracranial carotid artery and not these other potential causes of error. [1] Drs. Young and Humphrey’s main concern focuses on inter-observer variability in interpreting vascular imaging test results. We acknowledge that it is important to assess interobserver variability in traditional studies of test performance. However, as single radiologists provide official reports for individual tests, this type of variability could not be assessed in our study, and is not relevant to routine clinical practice. Regardless of the source of error, a test that disagrees with an acknowledged gold standard (in this case catheter angiography) 18-28% of the time should be of concern to clinicians relying on the results for clinical decision-making. We agree wholeheartedly that “performing the safest test that gives a reliable answer” is the best approach. Our study reinforces the point that it is critical for clinicians to know the performance characteristics of noninvasive tests as performed at their own institutions as this can be highly variable and error rates in routine practice may be substantial. [7, 8] References 1) North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N.Engl.J.Med 1991;325:445-453. 2) Howard, Chambless LE, Baker WH et al. A multicenter validation study of Doppler ultrasound versus angiography. J.Stroke Cerebrovasc.Dis. 1991;1:166-173. 3) Howard G, Baker WH, Chambless LE, Howard VJ, Jones AM, Toole JF. An approach for the use of Doppler ultrasound as a screening tool for hemodynamically significant stenosis (despite heterogeneity of doppler performance). A multicenter experience. Stroke. 1996;27:1951-1957.