Cost-efficient carotid surgery: A comprehensive evaluation

Cost-efficient carotid surgery: A comprehensive evaluation Alex D. Ammar , M_D, Wichita, Kan .

Purpose: This study was performed to determine whether comprehensive cost-cutting strategies adversely affect the outcome in patients undergoing carotid endarterectomy. Methods: From December 1994 to December 1995, 237 consecutive patients undergoing 260 carotid endarterectomies were prospectively studied. The following variables were assessed: carotid arteriography, preoperative laboratory tests, electrocardiograms and chest x-ray films, use of carotid shunts during operation, use of pathology department, intensive care, oxygen therapy, telemetry, and hospital stay. In addition, complications were tabulated. Results: Previously, all variables evaluated were routinely ordered. Subsequent to initiat- ing the cost-containment strategies, the following results were achieved: arteriography in 52 (22%) of 237 patients, preoperative complete blood cell count and SMA-7 in 161 (62%) of 260 cases, preoperative electrocardiograms in 185 (71%) of 260 cases, preoperative chest x-ray films in 190 (73%) of 260 cases, carotid shunts in 83 (32%) of 260 cases, disease in no cases (0%), intensive care in 29 (11%) of 260 cases, oxygen therapy in 34 (13%) of 260 cases, telemetry in 17 (7%) of 260 cases, and hospital stay was decreased from an average of 2.6 to 1.3 days. Total savings based on average hospital and physician charges was $2.3 million. Complications included four strokes, one myocardial infarction, and no deaths. No patient required readmission. No recurrent or new neurologic or cardiac findings were identified clinically in follow-up at 1 and 4 weeks after surgery. Conclusions: The results dearly demonstrate that comprehensive cost-cutting strategies can reduce charges significantly while maintaining patient safety. (J Vasc Surg 1996;24: 1050-6.)

Cost in health care is currently an important issue. Over the years the author has been impressed with the use o f extensive testing, monitoring, and invasive procedures in patients undergoing surgery, all o f which may not significantly alter outcome. In addition, providers have historically had little incentive to lower costs. Presently, efforts are being made by businesses, third-party payers, legislators, and providers to cut costs; h9wever , this is acceptable only if patient safety is not sacrificed.

Approximately 100,000 carotid endarterectomies (CEAs) are performed annually in the United States at a cost o f $1.2 billion. 1,2 Many recent articles have addressed cost-reducing strategies in carotid surgery. s-2s Each study has dealt mainly with one or two aspects of cost reduction such as reduction in inten-

From the Department of Surgery, The University of Kansas School of Medicine.

Reprint requests: Alex D. Ammar, MD, 818 North Emporia, Ste 200, Wichita, KS 67214.

Copyright @ 1996 by The Society for Vascular Surgery and Inter- national Society for Cardiovascular Surgery, North American Chapter.

0741-5214/96/$5.00 + 0 24/1/74526

1050

sive care use, reduction in hospital stay, or reduction in the use o f arteriography, all o f which have not adversely affected patient safety. Review of a recent hospital bill after uneventful CEA from our institu- tion was appalling. Consequently the author em- barked on a study to determine whether comprehensive cost-cutting strategies would adversely affect the outcome in patients undergoing CEA. The author specifically quantified carotid arteriography, preoperative laboratory tests, electrocardiograms (ECG) and chest x-ray films, use of carotid shunts during operation, use of the pathology department, intensive care, oxygen therapy, telemetry, and hospital stay. To evaluate patient safety, complications, particularly stroke morbidity and death, were tabulated.

M E T H O D S

From December 1994 to December 1995 a per- sonal series of 237 consecutive patients admitted for 260 CEAs was prospectively studied by the author in two large community hospitals affiliated with The University of Kansas School of Medicine-Wichita. Staged bilateral operation was performed on 23 patients, and 17 redo operations were performed in this

)'OURNAL OF VASCULAR SURGERY Volume 24, Number 6 A m m a r 1 0 5 1

Tab le I . Recent savings strategies (260 CEAs)

Use before Use after study study Average

instituted instituted Year Average hospital physician Total savings Item (%) (%) instituted charge 1995 charge 1995

Outpatient 100 22 1994 $6500 $ 1 6 0 0 $i,643,760]" arteriography (arch & bilateral selective carotid)*

Preoperative laboratory 100 62 1993 $ 140 $ 13,860 (cBc, SMA-7)

Preoperative 100 71 i993 $ 150 $ 35 $ i3,875 electrocardiogram

Preoperative chest X- 100 73 I993 $ 185 $ 40 $ 15,750 ray evaluation

Carotid shunts 100 32 1991 $ 80 $ 14,160 (A~gyle)

Pathology (gross & 100 0 1994 $ 125 $ 125 $ 65,000 microscopic)

Intensive care 100 11 1994 $1500 per dayz~§ $ 229,000 Oxygen 100 I3 1994 $ 250 per day $ 56,500 Teleme~ onward 100 7 1994 $ 350 per day $ 76,300 Hospital stay 2.6 days 1.3 days 1991 $ 500 per day~: $ 169,000

$2,297,205"

*Includes technical charge for bilateral selective carotid catheterization and supervision and interpretation charge for arch, bilateral carotid common, internal, cerebral), bilateral external carotid, bilateral vertebral (cervical, intracranial). 1"Hospital and physician charges for repeated duplex scans are subtracted from total arteriography savings. :~Room and board only. §Regular room rate subtracted from intensive care unit rate.

group. O f the entire patient population two under- wcnt concomitant carotid-subclavian bypass for proximal disease, four underwent repair o f carotid kink by resection and end- to-end anastomosis of the common carotid artery, two had dilatation for fibro- muscular dysplasia with an additional kink repair by internal carotid to external carotid transposition, and one had concomitant vertebral transposition to the common carotid artery. The average patient age was 76 years. The patient population consisted of 154 men. Demographic data were typical o f elderly patients: hypertension 185 (78%), diabetes mellitus 62 (26%), coronary artery disease by history or electrocardiogram 149 (63%), coronary revascularization by coronary artery bypass or percutaneous transluminal angioplasty 73 (31%), previous stroke 33 (14%), and continuance o f smoking 161 (68%). CEAs were performed in 16t (62%) patients with typical carotid territory symptoms and appropriate carotid lesions and in 99 (38%) in patients who had no symptoms or had nonhemispheric symptoms and stenosis o f 80% or greater.

Hospital and physician charge information is ex- tremely difficult to obtain from our institutions. Ac- tual hospital cost has been impossible to obtain. For the purpose of calculating charge savings in Table I and charge information in Tables I I and I I I , the

finance department personnel at our two hospitals were contacted. After multiple calls and several weeks, a range of charges for each item was received and an average determined. Hospital-based patholo- gist and radiologist charges were also obtained from the hospitals and from one radiology group's office manager, and an average from a range o f charges was obtained.

In the 216 (91%) patients referred by noncardiologists, preoperative cardiac screening tests other than routine preoperative ECG were performed only when clinically indicated, that is, for unstable, recent onset, or undiagnosed angina, significant dyspnea unrelated to chronic obstructive lung disease, and new arrhythmias.

Carotid duplex scans were routinely reviewed by the author. Many of these studies were ordered by primary care physicians who frequently practice m rural areas throughout the state. The reliability of these duplex scans, especially those performed m rural areas, was unk_nown to the author. I f any question arose as m the technical adequacy, interpretation, or reliability o f the duplex scan, particularly In patients with no symptoms, these were repeated m the vascular laboratory located in either o f the two institutions where the operations were performed. The vascular v :chnicians in these two laboratories are

JOURNAL OF VASCULAR SURGERY 1052 Ammar December 1996

Table II. Average hospital and physician charges for selected tests to evaluate coronary artery disease and cardiac function

Hospital Physician Evaluative test charge charge

Color Doppler echocardiogram $1180 $175 Treadmill stress test $ 380 $140 Stress dobutamine echocardiogram $ 850 $200 Thallium treadmill stress test $1350 $200 Dipyridamole thallium stress test $I780 $200 Coronary arteriogram with left $7000 $850

ventriculogram

certified. Overall accuracy of the laboratories is peri- odically monitored by peer review consisting of a vascular technician, a vascular surgeon, and a radiologist.

The criteria for obtaining arteriography included the following: difficult to scan vessels, distal cervical internal carotid artery disease or presence of a loop or kink, which may alter flow dynamics, question of inflow disease, stenosis less than 50% in patients with symptoms, presence of recurrent or severe bilateral disease or occlusion (no longer definite indications for arteriography), and disagreement between duplex scans.

Most of our patients who undergo arteriography have intraarterial digital subtraction arch and unilat- eral or bilateral selective carotid arteriography without selective subclavian or vertebral views unless clinically warranted. Consequently, in determining the average hospital and radiologist (technical, supervision, interpretation) charges seen in Table I, the average charge for arch and bilateral selective carotid arteriography is used.

Complete blood count (CBC), electrolytes, glu- cose, and creatinine (SMA-7) were ordered on admis- sion if not done within the past 30 days before surgery and if ECG and chest-x-ray evaluation were not done within the past 90 days, unless clinically warranted. Patients were questioned regarding previous testing, and reports were obtained to avoid rep- etition. Patients were instructed to bring any home medications to the hospital whenever possible to avoid the use of hospital pharmacies.

The number of and indications for computed tomography (CT) brain scans, magnetic resonance imaging (MRI) brain scans, and magnetic resonance arteriograms (MRA) ordered by referring physicians were tabulated.

Patients were admitted the same day of operation. Operation was performed with the patients under

general endotracheal anesthesia. Low molecular weight dextran was started at the beginning of each procedure at an initial rate of 70 to 100 m l / h r for the first hour and was continued at a rate of 25 to 40 m l / h r until 500 ml was infused. The rate was depen- dent on the age, weight, and cardiac status of the patient. Heparin, 1.5 mg/kg , was given several min- utes before carotid clamping was performed. Sys- temic blood pressure was maintained at the upper end of each patient's physiologic range and was measured by continuous arterial pressure monitoring with a radial arterial line. Shunts (Argyle; Sherwood Medi- cal, St. Louis, Mo.) were opened only when re- quested by the operating surgeon; however, the cir- culating nurse was instructed before each case to have them readily available. The three indications for shunt placement include history of ipsilateral stroke, contralateral internal carotid artery occlusion, or nonpalpable internal carotid stump pulse. 2. Standard endarterectomy was meticulously completed. Speci- mens were not sent to the pathology department for either microscopic or gross inspection. Intraoperative imaging was not performed. Patching was performed infrequently, except in those patients who underwent redo operation or in those with small internal carotid arteries such as young patients with premature ath- erosclerosis. Protamine sulfate was given to reverse one half the heparin previously administered. The lowest cost anesthetic drugs that accomplished the desired end point were used but not evaluated in this study. None of the patients had placement of pulmo- nary arterial catheters. Drains were rarely used.

Patients were tal~en to the recovery room for 11/2 to 21/2 hours of observation. The decision was made to trfinsfer the patient to a regular surgical floor or to the intensive care unit based primarily on the stability of the blood pressure. I f the blood pressure was stable for 1 hour without medication, the patient was transferred to the regular surgical floor. I f the blood pressure was instable, either excessively high or low, the patient was transferred to the intensive care unit. The usual acceptable range of blood pressure was a systolic of 100 to 170 mm Hg. Other reasons for transfer to the intensive care unit included pharyngeal dismotility (one patient), perioperative arrhythmias (three patients), or a history of severe cardiac disease (three patients).

Oxygen saturations were performed in the recovery room before transfer. I f the oxygen saturation was greater than 90% on room air, oxygen was discontinued. Telemetry was not ordered on transfer to the regular surgical floor unless significant arrhythmias were present during the operation or in the recovery

JOURNAL OF VASCUIdKR SURGERY Volume 24, Number 6 Ammar 1053

room, or occasionally if the patient had a history of severe heart disease.

Patients received oral intake the evening of oper- at_ion after the presence of cervical hematoma and neurologic complications was excluded. Intravenous fluids were discontinued as soon as the patient could adequately take oral fluids.

Patients were dismissed the day after surgery whenever possible. Patients were clinically monitored for wound complications, clinically significant peripheral nerve deficits that affected function, central neurologic events, cardiac events, and death. Patients were seen in the office of the author during the first and fourth postoperative weeks and were questioned again regarding their cardiac and neurologic status.

RE S U L T S (TABLE I)

In calculating percentages and savings, patients undergoing bilateral CEA were counted twice except in calculations involving arteriography and duplex scans, because these studies would not have been repeated for the second operation. All patients were initially screened with carotid duplex scans. Carotid arteriography was obtained in 52 (22%) patients. The reasons for obtaining arteriography were the following: distal cervical internal carotid artery disease or presence o f a loop or kink (4), proximal arterial disease (7), presence of recurrent or severe bilateral disease or occlusion (9), stenosis less than 50% in patients with symptoms (10), disagreement between two duplex scans (4), during evaluation of other areas such as coronary or peripheral arteries (4) and ordered by other physicians before referral (14). Based on review of clinical history and duplex scan findings by the author, seven arteriograms were not needed in the group ordered by other physicians. Nevertheless, these were not excluded from calculations of total savings. Duplex scans were repeated 72 times: and the extra charge was deducted from the total savings. To our knowledge no patient was inappropriately treated either by proceeding with operation (gross findings at surgery correlated with preoperative imaging) or withholding operation (no new neurologic deficits reported) based on the preoperative imaging performed. No complications of arteriography occurred.

The following items were ordered in the patients studied (percentages in parenthesis) preoperative CBC and SMA-7 in 161 (62%), preoperative ECG in 185 (71%), preoperative chest x-ray evaluation in 190 (73%), carotid shunts in 83 (32%), intensive care unit in 29 (11%), oxygen in34 (13%), and telemetry in 17 (7%). One patient who had a postoperative stroke

Table III . Average hospital and physician charges for selected tests to evaluate cerebrovascular disease

Hospital Physician Evaluative res~ charge charge

Duplex carotid scan $ 420 8; 150 CT* head without and with contrast $1,050 $ 175 MRI t head-MRA:~ head and neck $1,600 $ 500

without and with contrast Arch and bilateral selective carotid $6,500 $1,600

arteriogram

*Computed tomography. tMagnefic resonance image. :[:Magnefic resonance arteriography.

remained in the intensive care unit for 3 days, and two patients remained for 2 days, one who had pharyngeal dismotility and one with severe hypertension; all other patients stayed 1 day. No panent transferred to the regular surgical floor required subsequent transfer to the intensive care unit. Thirteen patients were discharged from the intensive care unit and from the hospital on the first postoperative day.

Complications included the following: two cervical hematomas requiring reoperation, one glossopha- ryngeal nerve injury resulting in pharyngeal dismotility, four strokes, one myocardial infarction, and one congestive heart failure without myocardial infarction. All of the previously mentioned patients recovered well. The patient who had pharyngeal dismotil- ivy required Dobhofftube feedings for 6 weeks before swallowing returned to normal. The strokes were moderate in seventy, and all of these patients returned home, three with outpatient physical therapy and one after 4 weeks of therapy in a rehabilitation hospital. Of the four patients who had strokes, two were evaluated before surgery with arteriography and two with duplex scan only. One patient who had a postoperative stroke underwent immediate reoperation after duplex scan and was found to have platelet aggregates; the other three had normal postoperative duplex scans. The patient who had a myocardial infarction and the one with congestive heart failure recovered uneventfully. No complications occurred in the redo carotid surgery group or in those who underwent concomitant procedures. No deaths occurred in this series.

Hospital rooms are charged at midnight each day. Consequently, patients admitted for 1 day and discharged the next clay are charged for one day. On the first postoperative day 215 (83%) patients were dismissed, 37 (14%) on the second day, two on the third day, and six for periods greater than 3 days. The 17

JOURNAL OF VASCULAR SURGERY 1 0 5 4 Ammar" December 1996

patients who underwent redo CEA and the two patients who underwent concomitant carotid subclavian bypass were dismissed on the first postoperative day. The patient who had concomitant vertebral transposition was dismissed on the second postoperative day. The patient who had pharyngeal dismotility remained in the hospital for 5 days, two patients who had strokes remained for 5 days and two for 7 days, the patient who had myocardial infarction stayed for 6 days, and the patient who had congestive heart failure stayed 2 days, as did the patients who underwent reoperation for cervical hematomas. The average hospital stay was 1.3 days (as compared with our study published in 1991 when the average hospital stay was 2.6 days). 25 No patient required readmission.

All patients were seen in the office of the author at 1 and 4 weeks after surgery. No recurrent or new neurologic findings were found. Likewise, no new cardiac symptoms were identified clinically.

DISCUSSION

The tests and patient care items listed in Table I were routine before the onset of this series of patients. The results clearly demonstrate that comprehensive cost-cutting strategies have not sacrificed patient safety. The stroke rate of 1.5% and death rate of 0 compare favorably with the literature. Specifically, they have not changed in the author's practice compared with results published in 1991 before most of these cost-containing endeavors were instituted. 25 With the use of the savings strategies outlined , the hospital charge per patient undergoing evaluation and operation can be reduced from approximately $23,000 to $13,000 (including duplex scan, laboratory, ECG, chest x-ray evaluation, operating room, recovery room, intravenous fluids, anesthetic and other drugs, i-day room and board; and excluding arteriography, pathology, intensive care, oxygen therapy, telemetry). Radiologists' professional charge is decreased from $1800 (when chest x-ray evaluation, duplex scan, and arteriography is performed) to $200 (duplex scan and chest x-ray film interpretation only). It was surprising that the hospital and radiologist charge for arch and bilateral selective carotid arteriography was comparable to the hospital and surgeon charge for CEA. In one surgeon's practice more than 2 million dollars of savings have been easily achieved (Tabl e I). Because charges were used to calculate savings, and! because charges are more than cost or reimbursement, the true savings wil ! actually be less than the figure stated. Unfortunately, hospital

cost information is particularly difficult to obtain from the involved institutions. Nonetheless the savings are substantial.

To substitute duplex scanning for arteriography in the preoperative assessment of patients being con- sidered for CEA, the accuracy of each vascular laboratory must be clearly established. During a previous audit comparing duplex scans with intraarterial digital subtraction arteriography, duplex scans were diagnostic or within 10% stenosis in 87 (97%) of 90 vessels evaluated. No carotid bifurcations that were normal by arteriography were identified as diseased by duplex scan. All nine occluded internal carotid arteries were accurately identified by duplex scan. Two proximal flow-reducing lesions limited the ability to evaluate the carotid bifurcation, and one distal internal carotid artery kink was suggested by duplex findings. Arte- riography confirmed the proximal common carotid lesions associated with one high-grade and one moderate bifurcation stenosis; the distal kink was associated with high-grade stenosis.

Seventy-two percent of the charge savings were derived from elimination of arteriography. The average hospital charge of $6500 and radiologist charge of $1600 for arch and bilateral selective carotid arteriography seem excessive. This may reflect the additional charges incurred by selective catheterization of both carotid arteries, which would be more costly than nonselective arch and carotid arteriography. In addition, this may reflect the low penetrance of managed care (<15%) in our health care market. Two insurance company executives have stated that hospital charges for cardiovascular procedures in our community are above the national average. In one of the more recent articles regarding cost-effective CEA, Kraiss et a1.16 reported charges in excess of $4000 for arteriography in Seattle, a market which has a significant presence of managed care. The same authors commented that they are aware of procedural charges exceeding $7000 for carotid arteriography in some cities.

A clear charge profile for arteriography is exceed- ingly difficult to obtain from those who actually code the charges, because so many variables exist and are not consistent from patient to patient. For example, hospital charges for arteriography include intravenous fluids, medications, contrast media, guide wires, catheters, catheter placement fees for each vessel evaluated, dye injection fees for each vessel evaluated, and recovery room. Radiologist charges include catheter placement within each vessel studied, supervision of the procedure, and interpretation of data. It is important to realize that multiple charges can be generated

IOURNAL OF VASCULAR SURGERY Volume 24, Number 6 Ammar 1055

depending on how an arteriogram is ordered. Each maneuver adds charges by the hospital and radiologist. Therefore only those images that are necessary for good patient care should be ordered. This decision can be based on clinical and duplex findings.

Historically most or all specimens removed from a patient at operation are sent to the pathology department for evaluation and confirmation: However, because patient care is not altered after pathologic review and because the presence of an arterial lesion is verified on preoperative imaging studies such as duplex scans and arteriograms, our medical staff execu- tive committee approved that atheromas be ex- empted from pathologic evaluation.

Other strategies were not evaluated for the purpose of calculating savings in this study because of involvement of other physicians, difficulty in quanti- fying the savings, or lack of change in the author's practice career. Nonetheless these can also be signif- ican t For instance, cardiac testing can be very expensive (Table II) and in our experience is infrequently necessary in preparing a patient for CEA. Although a selection bias exists in that one third of our patients undergoing CEA have previously undergone some type o f coronary revascularization, only 4 (1.9%) of 216 patients referred by noncardiologists underwent Preoperative cardiac testing (including two coronary arteriograms ) other than routine ECG. Mason et al.26 recently reported that vascular surgery without preoperative coronaw arteriography generally leads to better outcomes and that coronary arteriography should be reserved for patients whose estimated mortality from vascular surgery is substantially higher ~an average,

F i T , seven CT brain scans (36 for nonfixed deficits), 14 M ~ scans (8 for nonfixed deficits), and 9 MRAs were ordered by referring physicians. CT scanning r~ely, if ever, alters surgical decision mak- ing[7'2Z '28 Since 1982 the author has obtained CT brain scans only in patients with acute fixed neuro- 10Nc deficits to exclude intracranial hemorrhage, tu- mors, and other unusual anatomic abnormalities. MRI is m0re expensive and like CT scanning would rarely help select patients for CEA Likewise, MRA is not only costly but rarely helpful and did not influ- ence patient treatment in this series. This modality does not adequately evaluate the proximal arch branches and adds little to the information obtained from u!trason0graphy regarding ~ e cervical carotid artery, and the findings of intracranial disease rarely change the m~agement recommended for bifurcation disease The use of these expensive scans (Table III) shoed be selective, and it is hoped that ongoing

educational meetings will further curb the use of these tests. Specifically, radiologists should be aggres- sively encouraged to exclude reporting the need for other imaging studies such as CT scans, MRI, MRA, and digital arteriography because of medicolegal im- plications, if the tests are not ordered, and because many inexperienced physicians will order the test when they are not necessary.

In-hospital pharmacy use should be kept to a minimum, because hospital charges for drugs are significantly more expensive than outside pharmacy charges. Patients are instructed before surgery to bring their medications with them to the hospital, and their own medications are used to save not only the cost of the medication but also the cost of admin- istering the drug, which is usually far more than the drug itself. For example, the hospital charge for aspi- rin (325 mg tablet) is only $0.01; however, the hospital charge for the administration is $5.05. Along this line, the author is in discussion with anesthesiol- ogists regarding anesthetic drugs. Although not addressed in this article, attempts are being made to use the most cost-effective drugs that offer the desired end point of anesthesia. Without the use of intraoperative monitoring of cerebral perfusion and with markedly reduced use of intensive care and hospital stay, the author believes that the cost of general anesthesia is comparable to that of regional anesthesia. 8

The savings shown in carotid surgery can proba- bly be extended to many other procedures. This was shown with the cell-saver autotransfusion device, which is unnecessary in most patients undergoing infrarenal abdominal aortic surgery. 29 It was also shown when aggressive discussion with manufactu> ers reduced the cost of Dacron grafts in our institu~ tion. a° Additional savings can be accomplished. We are negotiating with hospital administrators to un- bundle operating room charges so that each individ- ual surgeon can be credited with his or her efforts to cut costs. Furthermore outpatient surgery centers (with facilities to observe patients overnight) can be used, because they are much more cost-effective than hospitals. Unfortunately, the author has learned Medicare will not reimburse for CEA or many other procedures done in outpatient surgery centers, and discussion is beginning with our state and federal legislators in this regard. Last, efforts to directly con- tract with employers, thereby eliminating middle men or third-party payers, could drastically reduce red tape and administrative costs.

In conclusion, the previous model of cost-efficient carotid surgery has significantly reduced charges

JOURNAL OF VASCULAR SURGERY 1056 Ammar December 1996

without adversely affecting patient outcome. Inci- dentally, this was accomplished without fancy and costly computers, software, clinical pathways or pro- tocols, or consultants. Simple common sense is sufficient. Further savings can be attained by critically analyzing each test ordered, each drug or material used, and each service provided. Because the physician is ultimately responsible for most health care expenditure, all o f us must participate by action and by education of medical students, residents, nurses and other health care providers, and most important, health care consumers (patients).

REFERENCES

1. Pokras R, Dyken M. Dramatic changes in the performance of endarterectomy for diseases of the extracranial arteries of the head. Stroke 1988;19:1289-90.

2. Ernst CB, Rutl~ow IM, Clevelon RJ, et al. Vascular surgery in the United States. Report of the Joint Society for Vascular Surgery-International Society for Cardiovascular Surgery, Committee on Vascular Surgery Manpower. J Vasc Surg 1987;6:611-21.

3. Akers D, Bell W, Kerstein M. Does intracranial dye study contribute to evaluation of carotid artery disease? Am J Surg 1988;156:87-90.

4. Blackshear W, Connar tk Carotid endarterectomy without angiography. Am J Surg 1982;23:477-82.

5. Collier PE. Carotid endarterectomy: a safe cost-efficient ap- proach, J Vasc Surg 1992;16:926-33.

6. Crew JR, Dean M, Johnson JM. Carotid surgery without angiography. Am J Surg 1984;148:217-20.

7. Dawson DL, Zierler RE, Strandness DE Jr, Clowes AW, Kohler TtL The role of duplex scanning and arteriography before carotid endarterectomy: a prospective study. J Vasc Surg 1993;18:673-83.

8. Friedman SG, Tortolani AJ. Reduced length of stay following carotid endarterectomy under general anesthesia. Am J Surg 1995;170:235-6.

9. Gelbart HA, Moore WS. Carotid endarterectomy without angiography. Surg Clin North Am 1990;70:213-23.

10. Gertler JP, Cambria RP, Kistler JP, et al. Carotid surgery without arteriography: noninvasive selection of patients. Ann Vas Surg 1991;5:253-6.

11. Geuder J, Lamparello P, Riles T, Giangola G, Imaparto A. Is duplex scanning sufficient evaluation before carotid endarterectomy? J Vasc Surg 1989;9:193-201.

12. Goodson S, Flanigan D, Bishara R, Schuler J, Kikta M, Meyer J. Can carotid duplex scanning supplant arteriography in patients with focal carotid territory symptoms? J Vasc Surg 1987;5:551-7.

13. Hill JC, Carbonnean K, Baliga PK, Akers DL, Bell WH III, Kerstein MD. Safe extracranial vascular evaluation and surgery without preoperative arteriography. Ann Vasc Surg 1990;4: 34-8.

14. Hoyle RM, Jenldns JM, Edwards WH Sr, Edwards WH Jr,

Morten RS III, Mulherin JL Jr. Case management in cerebral revascularization. J Vasc Surg 1994;20:396-402.

15. Kent KC, Kuntz I(19I, Patel MR, et al. Perioperative imaging strategies for carotid endarterectomy: an analysis of morbidity and cost-effectiveness in symptomatic patients. JAMA 1995; 274:888-93.

16. Kraiss LW, Kilberg L, Critch S, Johansen KH. Short-stay carotid endarterectomy is safe and cost-effective. Am J Surg 1995;169:512-5.

17. Mani BS, Mullins TF, Catlin JC, O'Hara P. Carotid endarterectomy: a ten-year analysis of outcome and cost of treatment. J Vasc Surg 1990;12:732-39.

18. Moore WS, Ziomek S, Quinones-Baldrich WJ, Machleder HI, Busuttil RW, Baker JD. Can clinical evaluation and noninvasive testing substitute for arteriography in the evaluation of carotid artery disease? Ann Surg 1988;208:91-4.

19. Nicholas GG, Osborne MA, Jaffe JW, Reed IF III. Carotid artery stenosis: preoperative noninvasive evaluation in a com- mtmity hospital. J Vasc Surg 1995;22:9-16.

20. O'Btien MS, Ricotta JJ. Conserving resources after carotid endarterectomy: selective use of the intensive care unit. J Vasc Surg 1991;14:796-802.

21. Ranaboldo C, Davies J, Chant A. Duplex scanning alone before carotid endarterectomy: a 5-year experience. Eur J Vasc Surg 1991;5:415-9.

22. Sumner D, Russell J, Miles tL Are noninvasive test sufficiently accurate to identify patients in need of carotid arteriography? Surgery 1982;91:700-6.

23. Wagner WH, Treiman RL, Cossman DV, Foran RF, Levin PM, Cohen JL. The diminishing role of diagnostic arteriography in carotid artery disease; duplex scanning as definitive preoperative study. Ann Vasc Surg I991;5:105-10.

24. Ammar AD, Pauls DG. Correlation of carotid artery stump pressure with a palpable carotid artery pulse. J Cardiovasc Surg 1992;33:59-61.

25. Ammar AD, Hanosh JJ. Carotid endarterectomy: immediate morbidity and mortality in 337 patients. Kansas Med 1991; 92:197-98.

26. Mason JJ, Owens DK, Harris RA, Cooke JP, Hlatky MA. The role of coronary angiography and coronary revascularization before noncardiac vascular surgery. JAMA 1995;273:1919- 25.

27. Martin JD, Valentine RI, Myers SI, Rossi MB, Patterson LB, Clagett GP. Is routine CT scanning necessary in the preoperative evaluation of patients undergoing carotid endarterectomy? J Vasc Surg 1991;14:267-70.

28. StreetD, O'BrienM, Ricotta J, et al. Observations on cerebral computed tomography in patients having carotid endarterectomy. J Vasc Surg 1988;7:798-801.

29. Kelley-Patteson CL, Ammar AD, Kelley H. Should the cell saver autotransfusion device be used routinely in all infrarenal abdominal aortic bypass operations? J Vasc Surg 1993;18: 261-5.

30. Ammar AD: Cost containment in health care (letter to the editor). I Vasc Surg 1994;19:366-7.

Submitted Jan. 29, 1996; accepted Apr. 25, 1996.

Documents

Cost-efficient carotid surgery: A comprehensive evaluation