The current surgical management of carotid body paragangliomas Glenn M. LaMuraglia, M D , Richard L. Fabian, M D , David C. Brewster, M D , J o h n Pile-Spellman, M D , R. Clement Darling, M D , Richard P. Cambria, M D , and William M. Abbott, M D , Boston, Mass. To determine if recent trends in evaluation and therapy have contributed to the successful surgical management of carotid body paragangliomas, we reviewed our experience over the past decade. Nineteen carotid body paragangliomas were identified in 17 patients. Eleven patients underwent complete, preoperative embolization of their afferent arteries with one complication. Calculated carotid body paragangliomas surface areas did not differ between the embolized 64.6 -+ 43.3 cm 2 and nonembolized 63.0 -+ 57.9 cm2 lesions. Intraoperative blood loss was lower (p = 0.02) in the patients treated with embolization (372 -+ 213 ml) compared with their cohorts (609 -+ 564 ml). However, the operative times were equivalent 4.1 hours versus 4.5 hours in both groups. Intraoperative electroencephalographic (EEG) monitoring was used in 10 patients; in one patient the EEG indicated intraoperative thrombosis of the carotid artery, which was successfully treated by thrombectomy without complications. Two patients required carotid bifurcation resection and vascular reconstruction to remove the entire tumor; a late stroke manifested by contralateral hand weakness developed in one of these patients. The incidence of cranial nerve injury was low at 16%, with one transient ramus mandibularis paresis and two instances of vocal cord dysfimction. Two additional patients had a postoperative Hornet's syndrome. We conclude that by diminishing in~aoperative blood loss through complete and careful preoperative embolization and use of intraoperative EEG monitoring along with careful surgical technique, the complications associated with this challenging operation are facilitated and diminished. (J VASC SURG 1992;15: 1038-45.)

Arising in the carotid bifurcation, carotid body paragangliomas (CBPs) are usually slow-growing, asymptomatic masses that are incidentally identified on physical examination. Once attaining a large size, these neoplasms can occur with symptoms relating to compression of adjacent anatomic structures, but malignant local behavior or metastasis is infrequent. 1-4 The major technical challenge of CBP is resection of a highly vascular, bulky lesion that is densely adherent to the splayed bifurcation of the carotid artery. The proximity of the cranial nerves in a From the Division of Vascular Surgeryand the Head and Neck Serviceof the GeneralSurgicalServices,and the Neuroradiology Sectionof the Departmentof Radiology,MassachusettsGeneral Hospital and MassachusettsEye and Ear Infirmary,Harvard Medical School,Boston. Presented at the EighteenthAnnualMeetingof the New England Society for Vascular Surgery, Quebec, Canada, Sept. 26-27, 1991. Reprint requests: Glenn M. LaMuraglia,MD, Department of Surgery,MassachusettsGeneralHospital,Fruit St., Boston,MA 02114. 24/6/35505 1038

commonly bloody operative field can add considerably to the surgical morbidity. Surgical resection of CBPs has had a historically high mortality rate (29%) and complication (40%) rate, 5 but long-term results remained unsatisfactory with conservative management. s7 Advances in surgical techniques, preoperative angiography, anesthesia management, and blood product transfusion have helped decrease the mortality rates significantly, but the morbidity, especially in regard to postoperative cranial nerve dysfunction, remains high. 8-12 To determine whether improvements in preoperative evaluation and management including preoperative embolization, and the use of various intraoperative adjunctive measures could help decrease the high reported complication rate, we reviewed our experience over the past decade in the operative management of CBPs. PATIENTS AND METHODS During the 10-year period 1982 to 1991, all patients undergoing primary surgical resection of

Volume 15 Number 6 June 1992

Current surgical management of carotid body tumors 1039

CBPs at the Massachusetts General Hospital and Massachusetts Eye and Ear Infirmary were identified. Patients with other abnormalities such as glomus tumors that may have involved the carotid arteries, or cervical paragangliomas that did not involve the carotid bifurcation were excluded from this series. Nineteen CBPs were identified in 17 patients.

Demographic data and evaluation Five men and 12 women comprised the patient group for a male :female ratio of 1 : 2.4. The patient ages ranged from 23 to 78 years, with an average of 44 years. The CBPs were found equally between the left (10) and right (9) side of the neck. Two male patients were found to have bilateral lesions and -~nderwent staged surgical resection. The most common clinical presentation was an asymptomatic midcervical mass (68%), which was identified during routine physical examination. Three (16%) symptomatic patients were admitted with local discomfort or dysphagia, whereas another 16% CBPs were identified by careful evaluation of the contralateral carotid bifurcation or the family members of probands. 1a-is

Preoperative angiography and embolization Seventeen CBPs (89%) underwent preoperative diagnostic angiographic evaluation. In the two patients that had no angiograms, the correct preoperative diagnosis was not established, but on finding the CBPs, resection was undertaken. Eleven (58%) CBPs underwent preoperative, complete, selective angiographic embolization of afferent arteries that supplied these hypervascular rumors (Fig. 1) in an attempt to decrease operative blood loss and facilitate surgical tumor extirpation. The technique requires highly selective catheter cannulation of the arteries feeding the tumor. A standard angiogram is performed followed by superselective angiography with use of a microcatheter introduced into a feeder artery. Collaterals from the feeding vessels that anastomose to the vertebral or internal carotid artery and thus provide direct circulation to the brain, called "dangerous anastomoses," are also carefully identified (Fig. 2). Catheter positioning is appropriately established for injection of the embolic material. To avoid inadvertent embolization of dangerous anastomoses and vessels that perfuse cranial nerves, an angiogram with opacified xylocaine is performed. A neurologic examination before and after this injection helps identify situations at high risk for nerve or cerebral injury. This sequence is undertaken multiple times during the procedure as

Fig. 1. Digital subtraction angiograrn of a CBP before embolization demonstrates a large, very vascular mass splaying the carotid bifurcation. Note the overhanging nature of the mass around the arteries. the flow dynamics of the tumor change. Embolization is performed by the introduction of 150 to 300 txm polyvinyl alcohol beads (Ivalon) into the microvasculature of the CBPs and gelfoam into the artery branch itself to totally obliterate arterial flow (Fig. 3). Most of these afferent tumor arteries arise from branches of the external carotid artery, the most prominent of which is the ascending pharyngeal artery. The ascending cervical branches of the thyrocervical trunk and vertebral artery also need to be carefully examined. Meticulous technique is essential, especially during catheter manipulations and embolic material injections, to avoid particle reflt~x into the ophthalmic and cerebral circulations as discussed in more detail elsewhere. 16 Temporary balloon occlusion of the carotid artery can bc performed as part of the angiographic procedure to assess collateral cerebral circulation across the Circle of Willis. This is performed with the patient awake, and vasodilators are used to lower systemic blood pressure and determine the blood pressure threshold at which the patient develops unilateral neurologic symptoms.

Journal of VASCULAR SURGERY

1040 LaMuraglia et al.

Fig. 2. Digital subtraction, superselective angiogram into the ascending pharyngeal artery which is an afferent feeding artery into the CBP. Note the "dangerous anastomosis" (arrow) of an arterial collateral from the feeder vessel into the vertebral artery.

Fig. 3. Digital subtraction angiograrn of a CBP after complete embolization of the tumor and afferent arteries. Note the decrease in size and the absence of the tumor blush noted in Fig. 1. extent of surgical dissection required for CBP removal and should correlate most closely with the operative blood loss.

Operative management Although some patients were operated on up to a week after embolization, it is our preference to proceed 1 to 2 days after embolization to operation to minimize edema or local inflammatory changes. The surgical approach was through oblique neck incisions anterior to the sternocleidomastoid muscle, but larger tumors were approached through Y-shaped incisions in two cases. Early vascular control was obtained when possible. Resection of the external carotid artery was performed primarily to facilitate tumor removal. Intraoperative electroencephalographic (EEG) monitoring was used at the discretion of the surgeon. It was helpful in assessing cerebral circulation if carotid clamping or replacement became necessary. Operative blood loss was documented from the anesthetic or operative record. Tumor surface areas were calculated from the dimension measurements of the pathology reports, since they best reflect the

Statistics Statistical analysis was performed by means of Student's t test comparisons of unpaired data, and analysis of covariance (ANCOVA) was used to determine if there were relationships between factors examined.

RESULTS There were no complications from diagnostic angiography and one complication (9%) from preoperative embolization. In the one patient in whom problems developed, the radiologists terminated the procedure early because the extreme tormosity of the carotid and feeder arteries made stable catheter positioning very difficult. This resulted in an unacceptably high risk of reflux embolization into the cerebral circulation. Nevertheless, the next morning a transient episode of aphasia developed in the patient lasting approximately 24 hours. The CBP, analyzed

Volume 15 Number 6 June 1992

with cases that were not embolized, underwent uneventful resection during the same admission. Other than the one attempted CBP embolization that could not be performed, the remainder of cases were considered successful in obliterating major, afferentartery, blood flow to the CBP. The proximal external carotid artery was divided to facilitate CBP resection in seven (37%) patients, four of whom had previously undergone angiographic embolization. Two (11%) additional patients required resection of the entire carotid bifurcation with the specimen to adequately remove the tumors. One patient had reconstruction with a saphenous vein interposition graft, and the other had a primary anastomosis of the common to internal carotid artery. There were four (21%) additional carotid artery injuries incurred during tumor dissection that required suture repair. Preoperative embolization diminished intraoperative blood loss (Fig. 4), 373 -+ 213 ml versus 609 +- 564 ml, even with comparable surface area measurements in both groups (64.6 + 43.3 cm 2 vs 63.0 + 57.9 cruZ). When plotting operative blood loss versus surface area, and comparing the linear regression of the embolized and nonembolized patients (Fig. 4), ANCOVA provides a statistically significant difference (p = 0.02) between the two slopes. With use of this method of analysis, a difference is noted between the blood loss in the cases of embolized and nonembolized CBPs when taking into account the tamaor surface areas. There were no differences in operative times for CBP resection between the embolized (4.1 hours) and the nonembolized (4.5 hours) cases. Electroencephalographic monitoring was performed in 10 of the 19 patients. One of the patients demonstrated positive changes of ipsilateral slowing during the operation. On further investigation the internal carotid artery was found to be thrombosed. Thrombectomy restored patency, normalized the EEG, and the patient awakened without a neurologic deficit. No perioperative central neurologic problems were noted in patients in which the EEG was not used. All patients survived the surgical resection of the CBP. Pathologic examination of all specimens revealed benign histology of paraganglioma, although one patient requiring carotid resection had nontransmural tumor invasion of the artery. No patients had evidence of metastatic disease, and no known recurrences were noted in the patients' records. There was one (5%) systemic complication, postoperative respiratory failure from an aspiration

Current surgical management of carotid body tumors 1041

pneumonia, requiring prolonged ventilatory support. No wotmd complications were noted. One patient had a stroke, manifested by contralateral hand weakness, 10 days after CBP resection during repair of a clamp injury to the internal carotid artery. Cranial nerve injury occurred in three (16%) cases: one ramus mandibularis branch of the facial nerve paresis, thought to be traction related, which resolved within 4 months; and two tenth nerve injuries resulting in vocal cord paresis, only one of which eventually required polytetrafluoroethylene injections. The facial nerve injury occurred in a patient that had preoperative embolization, and the two tenth nerve injuries were in nonembolized CBPs. Permanent Homer's syndrome developed in two (11%) additional patients after operation. DISCUSSION Surgical resection has been well documented as the only means of obtaining definitive cure of CBP. 8,12 Although the high perioperative mortality rate has been successfully reduced to a negligible rate, 1,8,12 the postoperative morbidity has remained high. 8 Complications from surgical resection of these tumors results from the formidable challenge they pose to the surgeon because of their highly vascular nature, their intimate adherence to the carotid adventitia, their location at the carotid bifurcation which they usually splay and overhang in a dumbbell fashion, and their proximity or involvement with the cranial nerves. Consequently, the problems of excessive blood loss, stroke, or other neurologic deficits, and significant postoperative disability caused by cranial nerve dysfunction remain important risks to the surgical treatment of CBP. Our low incidence of neurologic morbidity concurs with other recent published series, which have noted a decrease in the historically high incidence of stroke.4'8'12 Previously reported series and our recent 10-year experience suggest that several features are contributing to improved surgical outcome. 1,8,10,17-19 Careful periadventitial carotid dissection is important to avoid injury to the arterial wall, which can precipitate serious bleeding and necessitate arterial repair. Resection of the external carotid artery can be useful in large tumors to facilitate the medial dissection, especially to the posterior aspect of the carotid bifurcation where the tumors are usually very adherent, or to identify the vagus nerve posteriorly. Carotid resection and vascular reconstruction have occasionally been indicated when the internal carotid artery is essentially enveloped by the CBP or when the tumor appears to be invading the arterial wall,

Journal of VASCULAR SURGERY

1042 LaMuraglia et al.

1500

[]

A

[]

E

1000

(~

All

--1 0 a

O O

.-1

500

0

100

200

SURFACE AREA ( e r a 2 )

Fig. 4. Linear plot of operative blood loss versus tumor surface area identifies the preoperative non-embolized (F), linear regression line (A); and embolized (zx), linear regression line (B), CBP. A significant difference exists between these slopes (ANCOVA p = 0.02), indicating that preoperative embolization does decrease operative bleeding when turnor surface area is taken into account. and dissection of the t u m o r from the artery cannot be performed. Since this possibility cannot be predicted before operation, it is recommended that the legs be prepared for operation in all cases should the saphenous vein be required. Test occlusion of the common carotid artery has been most recently performed with an intraluminal balloon catheter during preoperative angiography. This procedure has been found to be safe at our institution and may predict the need for carotid shunting should carotid clamping be necessary during subsequent CBP resection. Electroencephalographic monitoring, although used in only 5 5% of our cases, can prove very valuable with intraoperative management. This was demonstrated by the one case where the EEG was the first indication of unsuspected carotid artery thrombosis. Expeditious recognition and operative correction of this problem prevented postoperative complications. The EEG can also indicate the need for temporary intraluminal shunting should carotid clamping or a vascular reconstruction be necessary. Some authors have suggested early clamping of the common carotid artery and the insertion of a temporary intraluminal shunt to minimize blood loss during the snrgery. 2°,2I Although there can be the

need for temporary damping of the carotid artery, such as for a suture repair, we do not advocate its routine use. Carotid artery clamping decreases the arterial pressure, and the carotid arteries become soft and easily collapsible. This makes identification of the proper dissection planes between the CBP and artery more difficult. In addition, carotid clamping would require systemic heparin anticoagulation, and would not diminish the frequently substantial retrograde blood flow through the external carotid artery to the tumor.

Other specific surgical techniques may also have contributed to improving the surgical outcomes. Papaverine or nitroglycerine can be topically used to reverse internal carotid artery spasm. Thrombinsoaked, hemostatic pads can be applied to the dissected surface of the CBP during local retraction to diminish bleeding. The bipolar cautery can be very useful in minimizing the risk of electrical or thermal injury to the artery or contiguous cranial nerves. Careful attention to the anatomy and proximity of the cranial nerves in these tumors is imperative to minimize their injury, which remains the highest morbidity of this operation. For surgeons not familiar with the distorted anatomic relationships around this uncommon tumor or the vascular tech-

Volume 15 Number 6 June 1992

niques associated with safe, effective carotid artery surgery, a multidisciplinary approach involving both a head and neck surgeon and vascular surgeon can be advantageous to optimize the outcome of the patient.4,19 Besides the recognition of the important technical aspects of surgical management, the other major advance in the last 10 years that may have had an impacton facilitating surgical extirpation of CBPs is preoperative embolization of large tumors. This technique, refined by radiologists at our and other institutions, is usually performed at a single session and is facilitated by the presence of dilated vessels, which result from the high blood flow through the CBPs (Fig. 1). It is also important to document whether bilateral CBPs are present. Although there /lave been reports in the literature of preoperative embolization of CBPs, which reflect our successful experience with this technique, 1,1719 they have been primarily case reports or small series without a comparable cohort of patients that have not undergone embolization. In this study, the significant difference in the regression slopes of blood loss versus tumor size between CBPs treated with or without embolization (Fig. 4), clearly demonstrates that this technique reduces intraoperative blood loss, making the surgical field easier to work in. On the basis of the linear regressions, one would expect that the greatest benefit would be achieved in large tumors. Around the linear intercepts, with approximate CBPs surface areas of 35 cm 2 or diameters of 3 cm, the benefit of preoperative embolization in diminishing blood loss would not appear as great. However, the numbers in this series are not large and over-interpretation of the data should be avoided. The complication rate from preoperative embolization of these tumors is low in this series. The experience and judgment of the interventional radiologist is crucial to determine the safety of the injections of embolic material. When the risks are deemed high, as in the patient whose procedure was terminated because the catheter position could not be maintained, it is important not to continue the embolization. Careful attention to the dangerous anastomoses and blood supply to the cranial nerves, which can be identified with opacified xylocaine injections, is also critical, and embolizing material should not be introduced in equivocal situations. Because of the careful review of films and repeated examinations of the patients, these procedures are time consuming (4 to 6 hours) and costly to undertake. Since these are involved, difficult procedures that have the potential for serious complica-

Current surgical management of carotid body tumors 1043

tions, some authors have been reluctant to recommend their use. s Our data support the tenet that preoperative embolization does considerably facilitate the surgery by decreasing blood loss and may be an important factor in improving surgical outcomes. The authors would recommend preoperative embolization only by an experienced interventional radiologist, in CBPs that had readily identified afferent arteries amenable to embolization, in lesions at least of 3 cm in size, and that the procedure be undertaken within a few days before surgery. The authors thank Mr. Gil L'Italien for assistance with the statistical analysis and Ms. Feroline Laughlin for assistance in preparation of the manuscript. REFERENCES

1. Smith RF, ShettyPC, ReddyDJ. Surgicaltreatrnent of carotid paragangliomas presenting unusual technical difficulties.The value of preoperative embolization. J VASC SURG 1988;7: 631-7. 2. Shamblin WR, Re,Mine WH, Sheps SG, Harrison EG. Carotid body tumor (chemodectoma). Clinicopathologyanalysis of ninety cases. Am J Surg 1971;122:735-9. 3. Dent TL, Thompson NW, Fry WJ. Carotid body tumors. Surgery 1976;80:365-72. 4. Lees CD, Levine HL, Beven EG, Tucker HM. Tumors of the carotid body. Experiencewith 41 operative cases. Am J Surg 1981;142:362-5. 5. Monro RS. The natural history of carotid body tumours and their diagnosis and treatment with a report of five cases. Br J Surg 1950;37:445-53. 6. Irons GB, Weiland LH, Brown WL. Paragangliomas of the neck: clinicaland pathological analysisof 116 cases. Surg Clin North Am 1977;57:575-83. 7. Mitchell DC, Clyne CAC. Chemodectomas of the neck: the response to radiotherapy. Br J Surg 1985;72:903-5. 8. Hallett JW, Nora JD, Hollier LH, Cherry KJ, Pairolero PC. Trends in neurovasoalar complications of surgical management for carotid body and cervicalparangliomas: a fifty-year experience with 153 tumors. J VAsc SURG 1988;7:284-91. 9. Padberg FT, Cady B, Persson AV. Carotid body tumor: the Lahey Clinic Experience. Am J Surg 1983;145:526-8. 10. McPherson CAD, Halliday AW, MansfieldAO. Carotid body tttmours and other cervical paragangliomas: diagnosis and management in 25 patients. Br J Surg 1989;76:33-6. 11. DeBoer FRA, Terpstra IL, Vink M. Diagnosis, treatment and operative complications of carotid body tumors. Br J Surg 1981;68:433-8. 12. Krupski WC, EffeneyDJ, Ehrenfeld WI(, Stoney RJ. Cervical chemodectoma: technical considerations and management options. Am J Surg 1982;144:215-20. 13. Sprong DH, Kirby FG. Familialcarotid body tumors. Report of nine cases in eleven siblings. Ann West Med Surg 1949;3:241-2. 14. SheddDP, AriasJD, Glunk RP. Familialoccurrenceof carotid body tumors. Head Neck 1988;12:496-9. 15. DuBois J, KellyW, McMenarnin P, Macbeth GA. Bilateral carotid body tumors managed with preoperative embolization: a case report and review. J VASC SURG 1987;5:648-50.

1044

Journal of VASCULAR SURGERY

LaMuraglia et al.

16. Brismar J, Cronqvist S. Therapeutic embolization in the external carotid artery region. Acta Radiol Diag 1978;19: 715-31. 17. Schick PM, Hieshima GB, White RA, et aL Arterial catheter embolization followed by surgery for large chemodectoma. Surgery 1980;84:459-64. 18. Borges LF, Heros RC, DeBrun G. Carotid body tumors managed with preoperative emboIization. J Neurosurg 1983; 59:867-70. 19. Robison JG, Shagets FW, Beckett WC, Spies JB. A muir]dis-

ciplinary approach to reducing morbidity and operative blood loss during resection of carotid body rumor. Surg Gyneco! Obstet 1989;168:166-70. 20. Javid H, Chawla SK, Dye WS, er al. Carotid body tumor: resection or reflection. Arch Surg 1976;111:344-7. 21. Amir-Jahed AK. A simple method for resection of tumors of the carotid body. Surg Gynecol Obstet 1976;143:807-9. Submitted Oct. 4, 1991; accepted Dec. 6, 1991.

DISCUSSION

Dr. Thomas O ' D o n n e l l (Boston, Mass.). The authors have described their excellent results with resection of carotid paragangliomas during the 10-year period 1982 to 1991, a decade that embodies all of the potential modern methods for treating this rare, but highly demanding neck minor. The Surgical mortality rate for this tumor was 30% in the 1950s. Like today's report, the mortality rate has dropped to near zero in most modern series. The stroke rate, on the other hand, has declined from 22% as reported by Shamblin from the Mayo Clinic in the 1970s to less than 5% as shown in the present study. I would like to focus briefly on cranial nerve injuries associated with resection of this minor, because they generally represent an unsolved problem until the present series. You have reported a low 16% incidence of nerve injury, which, I might say, represents the lowest incidence of cranial nerve injury reported in the literature. Cranial nerve injury is due to two possible factors: (1) tumor invasion of the nerves requiring nerve dissection and sometimes actual resection and (2) technical misadventure. The present series' low incidence of nerve injury may be attributed to two factors. One, the use of selective preoperative angiographic embolization of the afferent arteries, which not only would decrease blood loss but, more important, provide a dry field in which to dissect. Could you comment on how long it takes your neuroradiologist to perform the embolization procedure? The second factor that accounts for this very low cranial nerve injury is the extent of tumor. The present series did not present the traditional Shamblin classification of tumor size and extent as follows: type I, where the tumor is fairly localized and not invading the arterial wall; type II, where there is some invasion of the adventitia; and type III, a more extensive tumor where there is transmural arterial invasion. Obviously, if a series has a high percentage o f Shamblin type I or even of type II tumors a low incidence of nerve injury should be expected. With these two types, the surgeon will not be dissecting in zone 3, the dangerous area. By contrast, if a series has a high incidence of type III tumors, extensive resection and usually venous interposition grafts must be used. Have you examined your data with regard to Shamblin's classification? I would suspect

that you had a low incidence of type III tumors, because only two patients in this series required bypass. Dr. David B. Pilcher (Burlington, Vt.). Ten years ago, we presented to this Society seven carotid body tumors that were resected without embolization. Only one of those had a greater than 6 0 0 ml blood loss. N o b o d y received a transfusion. It looks as if you had the same results because you had only two cases that had greater than the 600 ml. I would propose that small carotid body tumors can be resected safely with other techniques that you have described as magnification and the bipolar cautery, without the embolization. You are probably going to get more complications embolizing the small tumors than you are resecting them. Dr. Thomas J. D o n o v a n (Manchester, Conn.). I want to add a caveat for the embolization procedure: although it is very effective, it has to be done by experts as Dr. LaMuraglia has pointed out. We had a case at the Hartford Hospital with an arteriovenous malformation of the left ear that was being prepared for embolization with an arteriogram, and the patient had some vague symptoms they thought were anxiety. They proceeded with the second injection and she had a temporary quadroparesis and respiratory arrest after they injected the ascending pharangeal. She eventually recovered with about a year of physical therapy, but a suit to the radiology department ensued. And so I think it is extremely important to have the embolization done where the experience is good and the patients are protected. Dr. Alan D. Drezner (Hartford, Conn.). You mention the need for EEG monitoring. I am curious to know what the supporting data for that conclusion is. Dr. Glenn LaMuraglia. Dr. O'Donnell, it is true that the embolization procedures are not inexpensive and take a fair amount of time. On the average they take 4 to 6 hours a procedure. The radiologists like to have anesthesia monitoring even though the procedures are performed while the patient is awake. However, I agree with Dr. Donovan that the expertise of the individual doing this procedure is imperative. In the one case in which we had an embolizatiou complication, albeit small, I think this could have turned into a major disaster had the neurora-

Volume 15 Number 6 June 1992

Current surgical management of carotid body tumors 1045

diologist not terminated the procedure and continued to attempt to embolize the patient in which the risks were much too high. I agree with Dr. Pitcher that small rumors may not need embolization. In our slide of blood loss versus tumor size, the linear regression lines of nonembolized tumors and embolized tumors actually intercept each other. At that point when you convert the surface area to diameter of the rumor yields approximately 3 cm. So, therefore, I agree that somewhere around 2 to 3 cm there is really no need to embolize these patients because of the potential added risk of the procedure. In these smaller tumors, the involvement of cranial nerves and the difficulty of the dissection is not quite as it is with the larger lesions. Although we did not retrospectively categorize all these tumors by the Shamblin criteria of I, II, and III, if you examine the surface areas and convert them to size I would ray that there were a few Shamblin type I, most of the

patients were Shamblin type II, and there were approximately three or four of them that were Shamblin type III, I believe this is the approximate distribution in most of the series that have reported according to that criteria. Dr. Drezner, regarding the question of EEG monitoring, we use EEG monitoring for all of our carotid endarterectomies and every time we expect to clamp the carotid artery. As in the one patient who had an unsuspected internal carotid artery thrombosis during the procedure, this was expeditiously recognized by EEG and corrected before a stroke or other major complication could have arisen. This reinforces our experience and our recommendation that EEG be used in these circumstances. However, I agree there may be an occasional patient who may have thrombosis of their internal carotid artery, and because of good collateral circulation across the Circle of Willis, it may not be detected on the EEG monitor.

LIEBIG FOUNDATION AWARD FOR VASCULAR SURGICAL RESEARCH, 1993 The Liebig Foundation announces the eleventh annual competitive award of $10,000 for the best essay on a problem in general vascular surgery. The investigative work shall be: 1. Clinical or experimental research 2. Original and unpublished (not submitted elsewhere for publication) 3. Performed by a house officer in the United States, Canada, or Mexico with senior collaborators acting in a consultive capacity 4. Submitted in English (6 copies of typed manuscript and 6 copies of glossy prints of illustrations) 5. Accompanied by a signed letter from the essayist's superior confirming the status of the essayist and complying with "Instructions to Authors" of the JouI~AL The submitted manuscripts will be evaluated by a select committee of vascular surgeons. The manuscript judged best will be submitted to the Program Committee of the Southern Association for Vascular Surgery for consideration for inclusion on its 1993 scientific program and publication in the JOURNALOF VASCULARSURGERY. Further inquiries may be directed to the same address to which the essays must be sent, postmarked no later than Sept. 15, 1992: Dr. Richard Turner Award Committee Secretary 112 Bauer Drive Oakland, NJ 07436 USA (201) 337-6126