Cerebral Angiography Practices at US Teaching Hospitals
Implications for Carotid Endarterectomy
Background and Purpose Although several clinical trials of carotid endarterectomy (CE) have been carried out in the last decade, the methods for angiographic measurement of carotid stenosis have not been standardized. How one measures carotid stenosis may affect the applicability of clinical trial results. We sought to obtain information on cerebral angiography practices at teaching hospitals in the United States.
Methods We surveyed hospitals with an accredited radiology residency program.
Results Of the 200 radiology program directors who were sent the survey, 97 responded. The angiographic complication rate was known in 68 of 97 medical centers and averaged 0.6%. The most common method being used for measurement of carotid stenosis is the NASCET method (70%). Forty-two of 97 program directors reported a decrease in the volume of angiography being performed. Of these 42, one third reported that CE was commonly being performed on the basis of noninvasive tests alone.
Conclusions The angiographic complication rate at American teaching hospitals is within the “acceptable” range. The NASCET method of stenosis measurement is the most popular among academic radiologists. The volume of cerebral angiography appears to be decreasing. How these data compare with community hospitals without an accredited radiology residency program warrants further study.
Extracranial atherosclerotic disease of the internal carotid artery is a major cause of ischemic stroke. The role of CE in the management of this condition has been the subject of several clinical trials conducted in the last two decades.1 2 3 4 Despite these large studies, there is still considerable uncertainty regarding the appropriate use of CE in patients with carotid stenosis, especially asymptomatic stenosis.5 6
The radiological evaluation of extracranial carotid stenosis has also been plagued by controversy.7 8 9 The controversy and confusion have arisen partly because of different imaging modalities available to assess carotid artery disease (eg, ultrasound, MR angiography, conventional angiography) and partly because of different methods by which angiographic stenosis is quantified.
We undertook a study to ascertain practices of academic radiologists in the evaluation of carotid artery stenosis. We were specifically interested in determining the following: (1) the major complication rate of cerebral angiography; (2) methods used for angiographic measurement of stenosis; and (3) whether CE was being done without cerebral angiography in a significant number of medical centers.
A questionnaire was sent to 200 radiology program directors at medical centers in the United States with an accredited radiology residency as listed in the 1995/1996 Graduate Medical Education directory.10 The survey was sent in May 1996, and responses were analyzed 4 months later. There was no attempt made to resubmit the questionnaire to initial nonresponders.
It was requested that the program director have the survey completed by the individual with the greatest knowledge of the subject matter (eg, neuroradiology section head).
The questionnaire was returned by 97 program directors (49%). All responding medical centers indicated that cerebral angiography was being performed. Results of specific questions were as follows:
Do you monitor the stroke and death rate from angiography at your institution?
Ninety-six of 97 programs (99%) reported that the angiographic complication rate was being monitored in their medical center. Stroke was defined as a focal neurological deficit of vascular origin lasting more than 24 hours.
Is the complication rate known for your institution?
Sixty-eight of 97 program directors (70%) were able to give a specific figure in response to this question. The median major complication rate for stroke and death was 0.6% (range, 0.01% to 2%) in these institutions.
Which method do you use for measurement of internal carotid artery stenosis?
Responses to this question are indicated in Table 1⇓. The NASCET method is defined as the ratio of the linear diameter at the point of greatest stenosis divided by the linear diameter of the distal, disease-free portion of the internal carotid artery. The ECST method is defined as the linear diameter at the point of greatest stenosis divided by the estimated diameter of the normal carotid bulb. None of the centers reported using the common carotid artery method or estimation of the residual lumen.
Have you experienced any increase or decrease in the volume of angiography being requested in the last 5 years?
Answers to this question can be found in Table 2⇓. One particularly noteworthy response from a radiologist at a medical center with a declining rate of angiography was that fewer than five angiograms were being done per year compared with over 1000 ultrasound and MRA examinations.
For those institutions reporting a decrease in the volume of cerebral angiography, do the surgeons in your hospital commonly or uncommonly operate on the basis of duplex and/or MRA?
In 42 medical centers that reported a decrease in the volume of angiography, 14 (33%) indicated that CE without angiography was common in their institution. In these hospitals, 79% employed both duplex and MRA preoperatively and 21% resorted to ultrasound alone. Two medical centers reported that computed tomography angiography was being used increasingly as a second noninvasive imaging option. An interesting observation from one radiologist was that vascular surgeons commonly operated without angiography but that this strategy was uncommonly used by neurosurgeons.
Fourteen of the 42 medical centers (33%) stated that CE without angiography was uncommon, and another one third of programs were uncertain. Overall, 14 of 97 medical centers (14%) responded that CE without angiography was a common practice.
As in any study in which a survey methodology is used, there are limitations to our project. First, the data set is incomplete, with approximately a 49% response rate. Second, we have no way of independently verifying information such as the neurological complication rates of angiography that were provided to us. Recently, concern has been expressed regarding the reliability of such self-reported data.11
Despite these limitations, we believe the study provides important information pertaining to practices of academic radiologists in the evaluation of carotid artery disease. With regard to the issue of complications of cerebral angiography, published studies have indicated a range varying from a less than 1% major complication rate to a 3% to 4% rate of permanent stroke and/or death.12 13 The angiographic morbidity and mortality rate of 0.6% reported by medical centers in this study is comparable to what can be achieved at major medical centers. In NASCET, an international, multicenter trial, the angiographic stroke rate in over 2200 patients has been approximately 0.7%.14 Therefore, although one would ideally like to have no complications from a diagnostic procedure, the angiographic complication rate reported in this study is within the “acceptable” range.
In close to one third of the medical centers, the angiographic complication rate was unknown. This is not entirely surprising, since in a recent study we demonstrated that the morbidity and mortality from CE is unknown in over 50% of academic medical centers.15 Clearly, there needs to be greater quality control and accountability among hospitals and physicians in terms of preoperative and postoperative evaluations of patients with carotid stenosis.
According to recent systematic reviews, perioperative rates for stroke and death are higher in symptomatic CE patients than in asymptomatic CE patients.16 17 Therefore, the impact of angiographic complications is especially important for the patient with asymptomatic carotid stenosis. In general, the risk-benefit ratio of CE for this population is much narrower than for a patient with high-grade, recently symptomatic stenosis. Thus, a patient with severe, symptomatic stenosis may have 10 times the risk of stroke from CE compared with the angiogram, whereas in asymptomatic patients, the ratio of stroke from CE compared with angiography is closer to 4:1 or 5:1. In the Asymptomatic Carotid Atherosclerosis Study, the risks of perioperative stroke were almost equivalent to the risks of angiographic stroke. In any case, there is a necessity for each medical center where CE is undertaken to strictly audit and record the angiographic complication rate, and this is especially significant for patients with asymptomatic stenosis.
With regard to methods of carotid artery stenosis measurement, the main modalities used in randomized clinical trials have been the NASCET method and the ECST method. Others have advocated using the distal common carotid artery as the reference point18 or utilizing estimation of the residual lumen.19 Our study demonstrates that the NASCET method either alone or in combination with “eyeballing” is overwhelmingly favored by American academic radiologists. Whether consumers of radiology services, such as neurologists and surgeons, strictly adhere to the NASCET method of stenosis measurement when reviewing angiograms is uncertain. However, some surgical services have found the NASCET method of stenosis measurement to be reliable and reproducible.20
As a result of the potential morbidity of cerebral angiography, some clinicians and imagers have suggested that noninvasive techniques, either alone or in a battery, may suffice as a preoperative radiological strategy.21 22 There may be other incentives for CE without angiography, including the possibility of reduced costs, which carries a certain appeal with third party payers. These points are somewhat at odds, however, with the fact that all the randomized clinical trials of CE have used angiography as the basis for stenosis grading. It is interesting to note that a substantial proportion of medical centers in our group (43%) reported a decline in the volume of cerebral angiography being performed. This may be due to a greater reliance on noninvasive imaging, although only a minority of the centers in our study (14%) are commonly performing CE without angiography. This figure is comparable to the rate of CE without angiography being observed in the Aspirin and Carotid Endarterectomy Trial of 13% (W. Taylor, poster reprint, 1997).
One further note is the fact that the questionnaire in this study was sent only to medical centers with an accredited residency. How these data compare with community hospitals without an accredited radiology residency is unknown at this juncture. If one speculates that a community hospital radiology department does less cerebral angiography than a “major academic medical center,” then perhaps the angiographic complication rates are higher at smaller community facilities. Data from other interventional procedures such as cardiac angioplasty have shown that procedure volume is linked to the complication rate.23
In summary, our study suggests that cerebral angiography is still an integral part of the preoperative imaging strategy at major medical centers in the United States and that it is being performed with acceptable levels of morbidity. Future studies should address the practices of neurologists and surgeons in carotid artery stenosis assessment, as well as the impact of managed care on perioperative imaging trends.
Selected Abbreviations and Acronyms
|ECST||=||European Carotid Surgery Trial|
|MRA||=||magnetic resonance angiography|
|NASCET||=||North American Symptomatic Carotid Endarterectomy Trial|
The authors would like to thank Dr Allan Fox for reviewing the manuscript.
- Received June 2, 1997.
- Revision received July 9, 1997.
- Accepted July 22, 1997.
- Copyright © 1997 by American Heart Association
Barnett HJM, Eliasziw M, Meldrum HE, Taylor DW. Do the facts and figures warrant a 10-fold increase in the performance of carotid endarterectomy on asymptomatic patients? Neurology. 1996;46:603-608.
Rothwell PM, Warlow CP. Making sense of the measurement of carotid stenosis. Cerebrovasc Dis. 1996;6:54-58.
Graduate Medical Education Directory 1995-96. Chicago, Ill: American Medical Association; 1996.
Davies KN, Humphrey PR. Complications of cerebral angiography in patients with symptomatic carotid territory ischemia screened by carotid ultrasound. J Neurol Neurosurg Psychiatry. 1993;56:967-972.
Dion JE, Gates PC, Fox AJ, Barnett HJM, Blom RJ. Clinical events following neuroangiography: a prospective study. Stroke. 1987;18:997-1004.
Eliasziw M, Rankin RN, Fox AJ, Haynes RB, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Accuracy and prognostic consequences of ultrasonography in identifying severe carotid artery stenosis. Stroke. 1995;26:1747-1752.
Chaturvedi S, Femino L. Are carotid endarterectomy complication rates being monitored? Stroke. 1997;28:245. Abstract.
Rothwell PM, Slattery J, Warlow CP. A systematic review of the risks of stroke and death due to endarterectomy for symptomatic carotid stenosis. Stroke. 1996;27:260-265.
Rothwell PM, Slattery J, Warlow CP. A systematic comparison of the risks of stroke and death due to carotid endarterectomy for symptomatic and asymptomatic stenosis. Stroke. 1996;27:266-269.
Bladin CF, Alexandrov AV, Norris JW. Carotid endarterectomy and the measurement of stenosis. Stroke. 1994;25:709-710.
Suwanwela N, Can U, Furie KL, Southern JF, Macdonald NR, Ogilvy CS, Hansen CJ, Buonanno FS, Abbott WM, Koroshetz WJ, Kistler JP. Carotid Doppler ultrasound criteria for internal carotid artery stenosis based on residual lumen diameter calculated from en bloc carotid endarterectomy specimens. Stroke. 1996;27:1965-1969.