Surgical management of carotid body tumor – Is Shamblin classification sufficient to predict surgical outcome?

Abstract

Background

The study aims to conduct a review of the surgical management of carotid body tumor.

Methods

Consecutive patients with CBT who received surgical interventions from January 1994 to January 2014 at our institution were reviewed. Clinical, operative, pathological and follow up information were reported.

Results

Twenty patients (four males; median age was 36) with 21 CBT operations were recorded during the period. One patient undertook sequential operations for bilateral CBTs. Patients had 19 neck mass, 1 incidental finding and 1 facial nerve palsy. Six CBTs (28.6%) were Shamblin class I, ten (47.6%) were class II and five (23.8%) were class III. Nine CBTs had preoperative conjunctive embolization. Two operations required internal carotid artery resection and reconstruction. Four patients received subtotal resections, while 17 achieved complete resection. Complications included two major strokes, three hoarse voice and two Horner’s syndrome. Shamblin class was significant predictor of operative time, blood loss, and whether complete resection accomplished, but could not predict postoperative complication. With median follow up period of 94 months, there was no tumor recurrence found in those had complete resection.

Conclusions

This small cohort showed that Shamblin class was significant in predicting technical difficulties but could not predict occurrence of complications.

Keywords

Carotid body tumor Shamblin classification operative outcomes predictors stroke internal carotid artery reconstruction recurrence

Introduction

Carotid body tumor (CBT) is a rare disease. Surgical excision is the only curative treatment, although some physicians may advocate radiotherapy^1–3 or conservative approach,⁴ if the tumor is very big or when surgical expertise is not available. As paragangliomas are histologically benign tumors, the criterion of malignancy as defined by presence of metastases to cervical lymph nodes or distant sites rather than the histologic appearance, malignant behavior has been noted in approximately 6% of carotid body paragangliomas.⁵

Since Shamblin et al.⁶ proposed the classification of CBTs based on the involvement of carotid vessels in 1971, it has been widely used to describe CBTs with the hope to predicting surgical difficulties and the need of intraoperative vessel replacement. Shamblin grade I is defined as small tumor with minimal attachment to carotid vessels. Grade II is defined as large tumor with some arterial attachment but does not entirely encase the carotid vessels. Grade III is defined as large tumor that encases the carotid vessels.

The aim of this study was to review the surgical outcomes of patients who had surgical excision of CBTs in our institution.

Methods

For the period of January 1994 to January 2014, all the patients with diagnosis of CBT who underwent surgical resection at our tertiary referral center were retrospectively reviewed. Data was collated into a predefined database.

Their demographics and presenting symptoms were recorded. Preoperative workups included genetic study for those with family history, functional study for hypertensive patients, and those with symptoms of raised catecholamine level, for example, flushing or tachycardia and imaging. Cross sectional imaging, either a computer tomography (CT) or magnetic resonance imaging (MRI), was used to measure the size of tumor and graded according to Shamblin classification.⁶ Other imaging modalities including duplex ultrasound, digital subtraction angiography, 123I-metaiodobenzylguanidine (MIBG) scintigraphy, octreotide scintigraphy, positron emission tomography (PET) or 18F-DOPA PET scan (FDOPA) scan may be used in selected cases. The diagnosis was made by a hypervascular tumor in typical location with characteristic splaying of the carotid bifurcation. Biopsy was not mandatory but may be indicated in uncertain situation and obvious unresectable cases. It could be in the form of fine needle aspiration or open. Trucut biopsy was rarely performed for fear of uncontrolled bleeding. Long axis and short axis of the tumor was measured by cross sectional imaging. Volume of tumor was approximated by 4/3 × π × (long axis) × (short axis)².

Preoperative embolization or radiotherapy may reduce the vascularity of the tumor and thereby minimize intraoperative blood loss. Embolization should be performed within 48 h of surgery; otherwise revascularization edema combined with local inflammatory process could create difficulty in dissection. All the operations were performed under general anesthesia. Standard linear neck incision was made anterior to the sternocleidomastoid muscle. Cephalic extension could be made posterior to the ear. The sternocleidomastoid muscle was pushed laterally and the common facial vein was divided to achieve satisfactory exposure. After carotid sheath was entered, careful control of the carotid vessels was made. Care was taken to avoid injury to glossopharyngeal nerve (cranial nerve ninth), vagal nerve (cranial nerve tenth) and hypoglossal nerve (cranial nerve twelfth). A systematic dissection around the tumor can minimize the risk of cranial nerve injury.⁷ Bipolar electro-cautery should also be available. Tumor was carefully dissected in the capsular-subadventitial plane. The external carotid artery could be sacrificed. Occasionally it was necessary to resected the tumor en-bloc with the carotid bifurcation with arterial continuity restored with either with primary anastomosis (with or without patch-plasty) or interposition graft, preferably with autologous vein conduit. Preoperative marking of the great saphaneous vein and preparation of the groin should be arranged if vascular reconstruction was anticipated. Adjunct of endovascular carotid stenting was a relatively new technique. Covered stent could provide vascular exclusion of external carotid artery feeding branches to the tumor. This technique was rarely performed and only appeared in limited case reports.

Completeness of resection, operative details and perioperative mortality and morbidity were recorded, with comparison of different Shamblin grading. Results of the final pathology, as well as long-term follow up information were given. Difference between groups was tested with chi square test for categorical variables and with Student’s t-test or ANOVA for continuous variables. All statistical analysis was performed using SPSS version 22.0. P value of less than 0.05 was accepted as significant.

Results

Patient demographics

Twenty patients (4 males and 16 females) with 21 CBT operations were recorded during the 20 years period. One patient undertook sequential operations for bilateral CBTs. The median age was 36 (range 21–58).

All except two presented as painless neck mass. One of the CBTs was an incidental finding of contralateral asymptomatic tumor during CT scan assessment. Another patient presented as facial nerve palsy involving the mandibular branch (Table 1). Eleven CBTs were over left side, 8 over right, and 1 bilateral.

Table 1.

Clinical presentation.

	Number (n = 21)	Percentage
Painless neck mass	19	90
Incidental finding on imaging	1	5
Cranial nerve palsy	1	5

Diagnostic details

Among 21 CBTs, all had preoperative cross sectional imaging of either CT scan (15 out of 21, 71%) or MRI scan (11 out of 21, 52%). Digital subtraction angiography was also frequently employed and was done on 12 CBTs (57%) (Table 2). The average diameter (long axis) measured on cross sectional imaging was 3.8 cm (range 2.0–8.0), with an averaged tumor volume of approximately 184 cm³ (range 10–1239 cm³). Preoperative functional study was carried out in two hypertensive patients in the form of serum or urine catecholamine level, all of which were within normal range. No patient received genetic study. FNACs were performed in six CBTs and open biopsies in three (Table 2).

Table 2.

Investigations used for diagnosis.

	Number (n = 21)	Percentage
Duplex ultrasound	8	38
CT/CTA	15	71
MRI/MRA	11	52
DSA	12	57
123I MIBG scan	0	0
Octreotide scintigraphy	0	0
PET/F-DOPA PET	1	5
FNAC	6	29
Open biopsy	3	14

Preoperative details

Six CBTs (28.6%) were Shamblin class I, 10 (47.6%) were class II, and 5 (23.8%) were class III. Eight CBTs had preoperative embolization and 1 CBT had both preoperative radiotherapy and embolization. Ascending pharyngeal artery was the target vessel in all except one of those received embolization; while the remaining one had embolization of descending branch of ophthalmic artery and muscular branch of vertebral artery. There was no external carotid artery covered stent inserted in our cohort. The average time from onset of symptoms to operation was 129 weeks, and 13 patients were assessed to be American physical status classification (ASA) grade I, and 7 patients were grade II. All patients were relatively young and healthy preoperatively.

Operative details

The mean operative time was 267 min (range 90–630), with mean blood loss of 606 ml (range 20–2000). The mean number of units of blood transfusion was 0.6 (range 0–4). Two operations required ICA resection and reconstruction to facilitate tumor removal. One used autologous vein graft and another used prosthetic graft. Four patients had subtotal resections, while 17 achieved complete resections.

Postoperative details

Two patients had postoperative major stroke: one frontal hemorrhage on contralateral side and one ipsilateral middle cerebral artery infarct. Other minor complications included three patients with hoarseness of voice and two patients had temporary Horner’s syndromes (Table 3).

Table 3.

Postoperative complications.

	Number (n = 21)	Percentage
Vascular/neurological complication
ICA occlusion with massive MCA infarct	1	5
Hemorrhagic stroke	1	5
Nerve injury
Hoarseness of voice	3	14
Temporary Horner syndrome	2	10

Note: ICA: internal carotid artery, MCA: middle cerebral artery.

Shamblin class was found to be significant in predicting operative time (mean operative time 139, 279, and 396 min for class I, II, and III, respectively, ANOVA P = 0.04), blood loss (Mean blood loss 53, 478, and 1500 ml for class I, II, and III, respectively, ANOVA P < 0.001), units of blood transfusion (0, 0.2, and 2 units for class I, II, and III, respectively, ANOVA P = 0.001), whether complete resection accomplished (6/6(100%), 9/10 (90%), and 2/5 (40%) achieved complete resection in class I, II, and III, respectively, chi-square P = 0.025), and whether vascular reconstruction necessary (vascular reconstruction needed in 0/6(0%), 0/10(0%), and 2/5 (40%) in class I, II, and III respectively, chi-square P = 0.029). However, it could not predict postoperative stroke (major stroke occurred in 1/6 (17%), 1/10 (10%), and 0/5 (0%) of class I, II, and III, chi square, P = 0.64), or overall complications (complications occurred in 2/6 (33%), 4/10 (40%), and 1/5 (20%) of class I, II, and III, chi square P = 0.88) (Table 4).

Table 4.

Operative outcomes in different Shamblin class.

	All (n = 21)	Shamblin I (n = 6)	Shamblin II (n = 10)	Shamblin III (n = 5)
	Mean (±standard deviation, range)	Mean (±standard deviation, range)	Mean (±standard deviation, range)	Mean (±standard deviation, range)	P value
Maximal diameter, cm	3.8 (±1.5, 2.0–8.0)	4.2 (±2.3, 2.0–8.0)	3.2 (±0.7, 2.2–4.8)	4.7 (±1.5, 3.3–7.0)	0.164^a
Volume, cm³	184 (±273, 19–1239)	183 (±200, 19–536)	76 (±36, 41–160)	400 (±483, 53–1239)	0.090^a
Operative time, min	267 (±141, 90–630)	139 (±31, 90–180)	279 (±112, 121–528)	396 (±156, 270–630)	0.04^a
Operative blood loss, ml	606 (±646, 20–2000)	53 (±32, 20–100)	478 (±435, 50–1500)	1500 (±324, 1100–2000)	<0.001^a
Units of blood transfusion	0.6 (±1.1, 0–4)	0.0 (±0.0, 0–0)	0.2 (±0.6, 0–2)	2 (±1.4, 0–4)	0.001^a
	No (%)	No (%)	No (%)	No (%)
Preoperative embolization	9/21 (43)	2/6 (33)	5/10 (50)	2/5 (40)	0.80^b
Complete resection achieved	17/21 (81)	6/6 (100)	9/10 (90)	2/5 (40)	0.025^b
Vascular reconstruction	2/21 (10)	0/6 (0)	0/10 (0)	2/5 (40)	0.029^b
Major stroke	2/21 (10)	1/6 (17)	1/10 (10)	0/5 (0)	0.64^b
Overall complication	7/21 (33)	2/6 (33)	4/10 (40)	1/5 (20)	0.74^b

By ANOVA.

By chi-square test comparing among Shamblin I, II, and III.

Preoperative embolization did not significantly reduce operating time or blood loss, nor did it predict if complete resection achieved, need of vascular reconstruction and complication occurrence (Table 5).

Table 5.

Operative outcomes with or without preoperative embolization.

	All (n = 21)	Preoperative embolization (n = 9)	No embolization (n = 12)
	Mean (±standard deviation, range)	Mean (±standard deviation, range)	Mean (±standard deviation, range)	P value
Maximal diameter, cm	3.8 (±1.5, 2.0–8.0)	3.8 (±1.5, 2.0–7.0)	3.9 (±1.6, 2.2–8.0)	0.959^a
Volume, cm³	184 (±273, 19–1239)	227 (±387, 19–1239)	151 (±155, 41–536)	0.542^a
Operative time, min	267 (±141, 90–630)	264 (±161, 120–630)	269 (±132, 90–528)	0.949^a
Operative blood loss, ml	606 (±646, 20–2000)	530 (±697, 40–2000)	667 (±628, 20–1500)	0.648^a
Units of blood transfusion	0.6 (±1.1, 0–4)	0.7 (±1.4, 0–4)	0.5 (±0.9, 0–2)	0.745^a
	No (%)	No (%)	No (%)
Shamblin class III	5/21 (24)	2/9 (22)	3/12 (25)	0.800^b
Complete resection achieved	17/21 (81)	9/9 (100)	8/12 (67)	0.083^b
Vascular reconstruction	2/21 (10)	2/9 (22)	0/12 (0)	0.171^b
Major stroke	2/21 (10)	2/9 (22)	0/12 (0)	0.171^b
Overall complications	7/21 (33)	3/9 (33)	4/12 (33)	0.681^b

By Student’s t-test.

By chi-square test comparing between preoperative embolization versus no embolization.

Pathology

Pathology showed 18 benign paraganglioma, 1 malignant paraganglioma and 2 schwanomma. All resected lymph node did not reveal metastatic CBT. Resected lymph node of one of the patient showed metastatic papillary thyroid carcinoma. Total thyroidectomy was subsequently performed.

Follow up details

With median follow-up period of 94 months, there were no early or long-term mortality and no tumor recurrence in all of the 16 patients whom complete resection achieved. Three residual tumors including the one with malignant paraganglioma received postoperative radiotherapy while one residual tumor was managed conservatively. All the patients were alive up to the present moment.

Discussion

Shamblin classification⁶ was first proposed in year 1971 with the hope to predict the need of intraoperative vessel replacement. Subsequent series of surgical management of CBTs did support its predictive power of operative difficulties.^8–14 However, some questioned its usage in predicting postoperative complications.^8,10 Our study showed that Shamblin classification was significant in predicting operative time, amount of blood loss, units of blood transfusion, whether complete resection can be accomplished and whether major vascular reconstruction was necessary. However, we could not demonstrate that Shamblin classification was significant in predicting major operative strokes or overall complications.

Pre-existing carotid atherosclerosis should be rare with this disease of young age. In addition, carotid vessels of our patients imaged preoperatively were all normal. The two operations that required vascular reconstruction did not result in stroke. On the other hand, two cases of postoperative stroke occurred unexpectedly in one Shamblin class I and another class II tumor, respectively. The aetiology of strokes in CBT resection may be more complex than vascular clamping and resection. Some hints could be discovered if the two cases of stroke were studied carefully.

Our first patient who developed postoperative stroke was a 30 years old lady who had a right-sided Shamblin class I 2.6 cm × 1.6 cm CBT. She had uneventful left CBT excision the year before. Preoperative functional test was not performed. Preoperative embolization of the ascending pharyngeal artery was arranged one day prior to surgery. Operation of right CBT was smooth with operating time 150 min and blood loss 80 ml. No vascular reconstruction was required. She had headache and uncontrolled hypertension immediate after operation. Initial urgent CT brain showed no abnormality. Twenty-four-hour urine catecholamine level later came back as elevated. She developed convulsion on postoperative day 6. Repeated CT brain showed left frontal hematoma with extension to ventricle. The hemorrhagic stroke was managed conservatively. She was discharged three weeks post excision with minimal neurological deficit. This might be a case of missed functional tumor. Successful operation for functional CBT required careful preoperative evaluation, alpha and beta-adrenergic blockage and gentle intraoperative manipulation and intensive postoperative care.¹⁵

Our second patient who had postoperative neurological complication was a 28 years old lady with Shamblin class II 4 cm × 2.5 cm right CBT. Tumor was embolized two days before operation. Intra-operatively, troublesome bleeding was encountered during dissection of cephalic end of tumor due to dense adhesion. Operating time was 352 min and blood loss 300 ml. Vascular reconstruction or clamping was not required. She complaint sudden onset of slurring of speech and left sided weakness 12 h post surgery. Urgent CT scan of brain showed middle cerebral artery thrombosis with infract. Emergency endovascular thrombectomy was attempted, with angiogram showing right distal common carotid artery occlusion with suspected dissection. The true lumen of right internal carotid artery was successfully cannulated and thrombectomy performed with Penumbra suction system (Penumbra Inc., CA, USA). Completion angiogram showed ICA was revascularized up to cavernous segment. However, extravasation of contrast was noticed in posterior middle cerebral artery territory. Decompressive craniectomy was necessary due to right temporoparietal intracerebral hemorrhage with mass effect noticed on Dyna CT scan. She was discharged after prolonged rehabilitation for over half a year. Unfortunately, she had permanent neurological damage rendering her wheelchair-bound (Figure 1). The dissection of the internal carotid artery could conceivably be caused by traction or diathermy thermal injury while bleeding encountered, and en-bloc carotid vessel resection might avoid the unintentional injury.¹⁶ Prophylactic carotid stent placement is a relatively novel technique to facilitate tumor removal. With protective stenting of the internal carotid artery, the vessel wall was reinforced allowing safe subadventitial dissection.^17,18

Figure 1.

(a) Angiogram showed occluded right distal common carotid artery (CCA). (b) Right internal carotid artery (ICA) was cannulated. Filling defects was treated with aspiration thrombectomy. (c) Contrast extravasation was noticed at posterior part of middle cerebral artery (MCA). (d) Dyna CT scan showed temporo-parietal lobe intracerebral hemorrhage with mass effect. SA: subclavian artery; VA: vertebral artery.

There was no operative mortality in our cohort. Perioperative morbidities were still substantial considering patients’ age and premorbid, with stroke rate up to 10% and nerve injury rate 24%. These figures were comparable to published series.^10–13

Controversies still existed for the effectiveness of preoperative embolization.^19–21 In our series, no difference was found in operative time, blood loss, units of blood transfusion and whether curative resection achieved between those received preoperative embolization and those without.

Conclusion

Close proximity of the tumor to major carotid vessels and nerves incurred significant operative morbidities in these relatively young fit patients. Our study showed that Shamblin classification was significant in predicting difficulty of the operation but could not predict the occurrence of postoperative complications.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Hinerman

Amdur

Morris

. Definitive radiotherapy in the management of paragangliomas arising in the head and neck: a 35-year experience. Head Neck 2008; 30: 1431–1438.

Haddad

. Carotid body tumors: radiotherapy as an alternative approach. Acta Med Iran 2012; 50: 359–360.

Dupin

Lang

Dessard-Diana

. Treatment of head and neck paragangliomas with external beam radiation therapy. Int J Radiat Oncol Biol Phys 2014; 89: 353–359.

Langerman

Athavale

Rangarajan

. Natural history of cervical paragangliomas: outcomes of observation of 43 patients. Arch Otolaryngol Head Neck Surg 2012; 138: 341–345.

Pellitteri

Rinaldo

Myssiorek

. Paragangliomas of the head and neck. Oral Oncol 2004; 40: 563–575.

Shamblin

ReMine

Sheps

. Carotid body tumor (chemodectoma). Clinicopathologic analysis of ninety cases. Am J Surg 1971; 122: 732–739.

Hallett

Jr Nora

Hollier

. Trends in neurovascular complications of surgical management for carotid body and cervical paragangliomas: a fifty-year experience with 153 tumors. J Vasc Surg 1988; 7: 284–291.

Smith

Passman

Dattilo

. Carotid body tumor resection: does the need for vascular reconstruction worsen outcome? Ann Vasc Surg 2006; 20: 435–439.

Plukker

Brongers

Vermey

. Outcome of surgical treatment for carotid body paraganglioma. Br J Surg 2001; 88: 1382–1386.

10.

Luna-Ortiz

Rascon-Ortiz

Villavicencio-Valencia

. Does Shamblin’s classification predict postoperative morbidity in carotid body tumors? A proposal to modify Shamblin’s classification. Eur Arch Otorhinolaryngol 2006; 263: 171–175.

11.

O’Neill

O’Donnell

Harkin

. A 22-year Northern Irish experience of carotid body tumours. Ulster Med J 2011; 80: 133–140.

12.

Sen

Stephen

Malepathi

. Neurological complications in carotid body tumors: a 6-year single-center experience. J Vasc Surg 2013; 57(2 Suppl.): 64S–68S.

13.

Makeieff

Raingeard

Alric

. Surgical management of carotid body tumors. Ann Surg Oncol 2008; 15: 2180–2186.

14.

Amato

Bianco

Compagna

. Surgical resection of carotid body paragangliomas: 10 years of experience. Am J Surg 2014; 207: 293–298.

15.

Zeng

Feng

Zhao

. Clinical characteristics and strategy for treatment of functional carotid body tumours. Int J Oral Maxillofac Surg 2013; 42: 436–439.

16.

Sharma

Massey

. Avoiding pitfalls in surgery of the neck, parapharyngeal space, and infratemporal fossa. Otolaryngol Clin North Am 2005; 38: 795–808.

17.

Piazza

Di Lella

Bacciu

. Preoperative protective stenting of the internal carotid artery in the management of complex head and neck paragangliomas: long-term results. Audiol Neurootol 2013; 18(6): 345–352.

18.

McDougall

Liu

Chow

. Covered carotid stents as an adjunct in the surgical treatment of carotid body tumors: a report of 2 cases and a review of the literature. Neurosurgery 2012; 71(1 Suppl Operative): 182–184. discussion 185.

19.

Farago

Castellani

Ponzi

. Preoperative embolization of carotid chemodectoma: a technical challenge that can be customized according to angioarchitecture. Illustrative cases. Neuroradiol J 2013; 26: 678–682.

20.

Zeitler

Glick

Har-El

. Preoperative embolization in carotid body tumor surgery: is it required? Ann Otol Rhinol Laryngol 2010; 119: 279–283.

21.

Power

Bower

Kasperbauer

. Impact of preoperative embolization on outcomes of carotid body tumor resections. J Vasc Surg 2012; 56: 979–989.