Microsurgery in carotid body paraganglioma| ACTA Otorhinolaryngologica Italica

Abstract

Objectives. In carotid paraganglioma surgery, magnification is crucial to properly evaluate the anatomical relationships between mass, carotid wall, cranial nerves, tumour vascular supply and fascial envelope. The aims of this study are to describe the microsurgical technique, along with the underlying microsurgical anatomy, and to assess outcomes in terms of disease control, complications and functional results.
Methods. Twenty-six patients, accounting for 29 carotid paragangliomas, treated with microsurgery by the same senior surgeon over a 35-year period, were included.
Results. No carotid injury requiring repair, nor peri- or post-operative stroke occurred in this series. No surgical injury of the main trunk of VII to XII cranial nerves occurred. Complete excision was obtained in all cases and no recurrence was observed during follow-up.
Conclusions. The small study size and its retrospective nature suggests caution; however, our results show that microsurgery can allow a safe and precise dissection of the carotids and nerves.

Introduction

Carotid paraganglioma is a rare tumour accounting for 0.6% of head and neck tumours and 65% of head and neck paragangliomas ^1-3, which originates from the paraganglia, a cluster of neuroendocrine cells, located on the postero-medial side of the common carotid bifurcation ^4,5. Treatment choices include surgical excision, radiation therapy, or a conservative wait and see approach, depending on parameters such as patient’s age, comorbidities, surgical risk, size of the tumour, tumour extension, localisation, and multicentricity ⁶.

Surgical excision currently represents the only option for definitive treatment ⁷. However, surgery is challenging and faces two major technical issues, separation of the tumour from the carotid wall and preservation of the cranial nerves, which are directly related to the functional outcome. Surgical resection can be associated with important morbidities, including massive bleeding, perioperative stroke and nerve loss ^3,8-12.

Performing a safe and effective surgical approach to carotid paragangliomas involves a thorough understanding of the anatomical relationship between the carotid adventitia, tumour, cranial nerves and the fascial apparatus wrapping around the tumour-carotid complex. The paraganglion lies on the carotid wall adventitia, under the cervical fascial sheet that individually invests carotids, internal jugular vein, and vagal nerve ^13,14. The tumour originates from the paraganglion and grows on the wall of the artery. It appears as a unique mass with the carotids, as the cervical fascia invests both tumour and carotid arteries in a single envelop and often fuses with the deep cervical fascia on the tumour’s posterior side. Here it is thick and highly vascular, as it carries most of the blood supply to the tumour. The carotids have a simple contact with the tumour, but, as the latter grows, they become lodged in a groove on the tumour. Growth and depth of the groove increase simultaneously up to the point that the groove borders fuse around the artery and, through further growth and fusion, they achieve complete encasement. A thin layer of loose connective tissue remains between tumour and artery. The blood supply is simple and obvious in a small tumour, while it remains basically the same, but richer, in a large tumour. In a small tumour there are two hila. The lower one carries the glomic artery originating from the dorsal aspect of the carotid bifurcation and running across connective tissue it ramifies to the tumour’s lower pole ^15,16. The cranial pole of the tumour receives branches from the ascending pharyngeal and external carotid artery ¹⁵. As the tumour grows cranially, the superior vessels take on most of the blood supply. These vessels reach the dorsal and, to a lesser extent, medial aspects of the tumour, ramifying on the fascial surface and thereby entering the tumour. In large tumours, such a distribution of tumour vessels on the fascial layer may be visualised at imaging as a thick, contrast enhanced, hyper-vascular halo.

Intra-operatively, the use of magnification tools is key to properly evaluate the complex anatomical relationships between tumour, carotid wall, cranial nerves, tumour vascular supply and fascial envelope. Recently, a few reports regarding the use of microsurgical dissection in carotid paraganglioma have appeared in the literature, a single case report by Ceccato et al. 2019 ¹⁷, and a series of 31 cases by Degollado-García et al. ¹⁸. However, a thorough discussion of the technical issues and anatomical considerations related to the microsurgical approach to carotid paraganglioma is still needed.

The aims of this study are: i) to describe the microsurgical technique, along with the underlying microsurgical anatomy; ii) to assess the outcomes in terms of disease control, complications and functional results.

Materials and methods

Patients

A cohort of 26 patients, accounting for 29 consecutive carotid paragangliomas, was selected out of the overall series of 228 cases of head and neck paragangliomas (175 jugular, 29 carotid and 24 vagal paragangliomas), that were surgically treated by the same senior surgeon (AM) over a 35-year period (1985 to 2020).

Each patient had undergone preoperative imaging with contrast-enhanced CT and, since 1990, with gadolinium-enhanced MRI. Tumours were radiologically classified according to Shamblin’s classes, and imaging-based classification was confirmed intra-operatively. When there was a discrepancy between the pre- and intraoperative appearance, Shamblin’s class was assigned according to intraoperative findings. Genetic assessment was systematically performed on each patient since 2005, and ⁶⁸Ga-DOTATATE PET imaging was added, to rule out multiple or metachronous tumours.

A clinical evaluation of cranial nerves VII to XII was performed in each case. No patient underwent preoperative tumour embolisation of a carotid tumour.

The clinical outcomes considered were disease control (including local recurrence and distant metastasis) and postoperative complications including early/late death, stroke, newly onset permanent cranial nerves loss, Horner’s syndrome, bleeding and cervical haematoma.

Clinical and radiological follow-up was scheduled as follows: clinical examination and head and neck contrast-enhanced MRI every year for the first five years, then every two years until the 10^th year for sporadic tumours, then every 5 years during the following years for familial or syndromic cases ¹⁹. In these latter cases of familial or multiple tumours, follow-up was completed with chest x-ray, contrast-enhanced total-body CT scan and, more recently, ⁶⁸Ga-DOTATATE PET.

Follow-up ranged from 2 to 23 years, mean 11 years. All but one case had more than 5 years of follow-up.

Surgical technique

Several microsurgical steps of resection were common in head and neck paragangliomas. Our experience, gained with tympano-jugular (TJ) paraganglioma (156 surgical cases with concomitant internal carotid artery involvement), was applied to the present series of carotid body paraganglioma.

The surgical microscope was set up with a 250 mm front lens. The microsurgical instrumentation included a Freer-type elevator with semi-sharp straight and curved ends, bipolar coagulation forceps and two-way suction-irrigation tubes (Fig. 1).

The patient was positioned supine, with the head contralaterally turned by 30°. Level II-III cervical lymph nodes were removed to rule out the presence of metastasis. Initial steps involved skin incision on the sternocleidomastoid area, raising of the skin-platysma flap, exposure of carotid arteries, internal jugular vein and cranial nerves X to XII, as well as facial nerve main trunk in the case of a high-lying tumour. If in contact with the tumour, the vagal and hypoglossal nerves were separated, and their fascial sheet preserved.

Resection of the carotid paraganglioma involved opening the common fascia (tumour-carotid fascia), dissecting fascia from tumour, dissecting tumour from carotid(s), dissecting lower pole of tumour from carotid bifurcation and coagulating the glomic artery. In detail, the surgical procedure was performed as follows.

Figures 2 and 3 show an exemplary case, describing surgical details.

The body of the tumour-carotid complex, along with its common fascia envelop (Fig. 3A), was exposed on its anterior, medial, and lateral sides. The superior pole of the tumour, previously freed from the hypoglossal nerve, was followed up to the transition to the dorsal side. In Shamblin 2-3 paragangliomas, while dissecting the dorsal side of the tumour, a thick, tight, highly vascularised tissue can be encountered and dissected, as allowed by a safe carotid retraction manoeuver. Sometimes such a dissection might be stopped and left for a later step. The vessels branching on the exposed surface of the common fascia were coagulated.

The anterior side of the fascia was opened vertically (Fig. 3B) and the cleavage plane between fascia and tumour was developed with further coagulation on bleeding spots between fascia and tumour. Intra-fascial dissection reached the internal (Fig. 3C,D) and external (Fig. 3E) carotids, where the tumour-artery interface was clearly visible. This interface was dissected with the Freer elevator in a cranio-caudal direction, also along its dorsal aspect (Fig. 3F), and stopped at the tumour’s lower pole to avoid injury to the glomic artery. Dissecting the lower pole of the tumour at the carotid bifurcation may be a difficult step of resection due to the risk of bleeding and lack of a landmark of the glomic artery. Bleeding may arise from the supplying arteries on the tumour’s posterior side as well as the glomic artery. Cranial displacement of the freed tumour, to reach the glomic artery, involved the risk of tearing its origin, thus causing a carotid fenestration. Bipolar coagulation was feasible, as the adventitia was preserved (Fig. 3G,H). A better option was to gain a good control of the glomic artery root area by fully freeing both carotids. Gently retracting external carotid and tumour revealed the tumour-carotid interface and allowed dissecting strictly on the tumour surface. Once the lower pole had been dissected, the tumour could be separated on its dorsal side all along the thick, vascular fascia, which was sometimes fused with the deep cervical fascia.

Resecting Shamblin 3 paragangliomas involved similar but more extensive steps, except for the technique to free the carotids from the preoperatively assessed encasement. After haemostasis on the common fascia and its opening, the dissection proceeded towards either the cranial or caudal pole of the tumour encasement, where the carotid entered the tumour. There, the curved dissector was run into the tumour-artery interface and lifted off the tumour by a few millimeters. The dissected bridge of the tumour was then coagulated and cut. The separation-coagulation-section manoeuver was performed in several steps until the encasement had been completely opened. The tumour was thus opened in two halves carrying the artery in a groove, from which it could be lifted.

Cleavage between tumour and the carotid was done carefully across the thin connective layers with respect to the artery adventitia. The bleeding spots encountered at this level belonged to the tumour, rather than to the artery, except for the glomic artery that was handled as reported above. The condition of “encasement” occurred both partially and totally in different areas of the tumour.

The external carotid artery was managed with the same technique as for the internal. In the case of a tumour impacting on the cranial base, the microscopic cervico-parotid approach allowed control of the upper aspect of the tumour, as well as the carotid and jugular foramina at the skull base. Wide variations on fasciae, connective tissue and vasculature were observed and related both to tumour size and to individual cases. Cleavage planes, as well as fasciae, were thick or thin, tight or loose, strong or weak. However, in every condition, reaching a cleavage was crucial for a safe resection. The variable tumour impact along the dissection planes and at the carotids interface, especially in Shamblin 3, required flexibility in the order of operative steps so as to perform the carotid dissection in the most favourable conditions and with the least injury.

Results

Study population

A total of 26 consecutive patients (11 males, and 15 females), accounting for 29 carotid paragangliomas were included in this study. All cases were treated with microsurgical excision. Of the 29 paragangliomas, 11 tumours were unilateral, while 18 in 14 patients were multiple, either unilateral or bilateral tumours. Three patients, belonging to the same family (two females and one male), suffered from familial paragangliomatosis, with bilateral or multiple unilateral tumours. One case had bilateral carotid paragangliomas (surgically treated on both sides), and a concomitant vagal one (which was followed-up).

The other case had a carotid paraganglioma, which was surgically treated, and a concomitant carotid and vagal paraganglioma on the other side, which was observed.

The third case of the family, not enrolled in the present surgical series, was a right carotid paraganglioma (PGL) operated on with artery removal in another institution, and a right operated vagal tumour. A right TJ PGL was observed and treated with radiation therapy at the other institution, as well as a right vocal cord PGL. Left vagal and carotid paraganglioma were observed and followed.

Tumours were classified as follows: 9 cases Shamblin 1 (30%), 17 Shamblin 2 (59%), and 3 Shamblin 3 (11%). Three cases were preoperatively over-staged class 3, but were intraoperatively assessed as 2, because the carotid artery was only partially encased by the tumour and a thick, hyper-vascular fascia connected the two close edges of the tumour simulating complete encasement. In one case of multiple unilateral paraganglioma (tympano-jugular, carotid and vagal PGLs, that appeared as a single mass preoperatively), a lymph node micrometastasis was evident at frozen sections and confirmed at definitive pathology. The tumours were removed in a single procedure. This patient died without evidence of disease after 20 years of follow-up. No genetic test was available at the time.

Clinical and functional results

No major intra- or peri-operative events were recorded. No carotid injury requiring repair and no peri- or postoperative stroke occurred in this series. The tumour resection time ranged from 30 minutes in Shamblin 1 to 180 minutes in Shamblin III cases.

No surgical injury or postoperative-onset functional damage of the main trunk of VII to XII cranial nerves occurred except for the lateral branch of the superior laryngeal nerve of the vagal and the descending branch of the hypoglossal, which were injured or at risk of undetected injury, in Shamblin 2 and 3. Complete excision was obtained in 100% of the cases and no recurrence was observed during follow-up in this cohort.

Discussion

Surgery of carotid body paragangliomas involves potential risks of intra-, peri-, and post-operative complications. The most frequently reported are stroke and new cranial nerve losses. In a systematic review on carotid PGL surgery, Suarez et al. ⁹ examined 2175 cases in 67 articles with 26% Shamblin 1, 44% Shamblin 2 and 30% Shamblin 3 cases. In the high classes, the common or internal carotid was resected in 12.5% of cases due to injury or difficult dissection and was reconstructed in 9.7%. Peri- or post-operative stroke with permanent sequelae occurred in 3%, and new postoperative permanent cranial nerve loss in 22.2%. In a review with meta-analysis on 4743 cases, Robertson et al.⁸ reported 2.3% of 30-day stroke-death for carotid injury, 30-day stroke of 3.5% and death-stroke rate of 4.1%. The rate was lower for Shamblin 1 and increased with the higher classes. New postoperative cranial nerve losses were associated to increasing Shamblin class, with an overall rate of 11.1%. By decreasing rate, the hypoglossal, vagal, sympathetic, glossopharyngeal, accessory, and facial were the affected nerves. Although the overall trend in risk of complication has shown a continuous decrease during the past decades, a small, but non-negligible, residual risk still remains ^20,21. Robertson et al. ⁸ outlined the possibility that the current risk of complication might be underestimated due to a possible publication bias, in terms of not reporting poor outcomes. The postoperative course of our series was uneventful. No carotid artery complications, in terms of stroke, arterial injury requiring repair, graft, or bypass, were encountered. Separation of the cranial nerves from the tumour was also uneventful, except for the lateral branch of the superior laryngeal nerve of the vagal and the descending branch of the hypoglossal in Shamblin 2 and 3 cases. To better characterise outcomes and the safety profile of our microsurgical technique, a larger prospective series with more cases is needed. Nonetheless, it may be reasonable to hypothesise that, applying the same microsurgical principles to the neck employed in other surgical fields, such as the skull base, a safer handling of vessels and connective tissue interfaces might be achieved by using microsurgery. In this sense, our preliminary results seem to be promising, outlining good disease control and a very favourable safety profile.

Microsurgery on carotid PGL was initiated by our group years after operating on jugular ^22-24 and vagal PGL ²⁵ applying in the neck the same principles of dissection on tumour, vessels and nerves, haemostasis with bipolar coagulation under suction-irrigation, periadventitial carotid dissection (which is different from the frequently reported sub-adventitial dissection) and development of a tumour-carotid interface. Preservation of the adventitia allowed the carotid root of the glomic artery to be coagulated when the latter was torn off the common carotid.

Handling of the hypertrophic, hyper-vascular fascia investing the complex tumour-carotid was a crucial step. It involved the preliminary exposure of the full tumour body from the lateral to the posterior side, coagulation of the vessels all along the fascial surface and, after opening of the fascial envelop with intra-fascial dissection, the bleeders on the fascia-tumour interface. This allowed both an ischaemic tumour and a clear view of the transition area between tumour and carotid. Separation of the carotid by a semi-sharp dissector was also safely performed in Shamblin 3 cases, but in these larger tumours more evidence is required to assess whether the microsurgical technique is effective and uneventful as shown in smaller tumours. Microsurgery on large tumours such as Shamblin 2-3 classes reported in the literature ^17-19 and in our series amount to a total of 50 cases with favourable results.

The issue of timing of surgery in multiple carotid PGL is still debated. Factors influencing the choice are patient age, comorbidities, life expectancy, genetics, vagal nerve function, PGL size and Shamblin class, as well as carotid atherosclerosis. Our current strategy prioritises the chance of recovery, preventing neurovascular injuries at the same time. In bilateral carotid tumours, the Shamblin 1 tumour is removed first, shortly followed by removal of the larger tumour.

In bilateral Shamblin 2-3, the smaller tumour is removed first. In equal sized tumours, there is no general criterion. Multiple unilateral lesions are removed in the same procedure.

Two aspects deserve mention, the handling of the tumour’s lower pole and stenting of the carotid artery. The lower pole impacting on carotid bifurcation involved an exacting dissection in a large tumour due to a combination of such factors as adhesion, glomic artery and bleeding from arteries of the tumour’s posterior aspect. Injury of the carotid artery was also a risk despite the dissection steps suggested above and implied vascular expertise for repair. Stenting the internal carotid artery ^26,27 never came into question, since the compensatory mechanism was pre-operatively assessed, and the artery’s sacrifice was to be left to intraoperative adverse circumstances that did not occur. The absence of clinically evident complications does not exclude the need to preoperatively assess both the efficacy of the compensatory intracranial circle with the balloon occlusion test (BOT) and the condition of the vessel wall. Stenting is prescribed if BOT is positive ^26,27. If BOT is negative, our experience suggests that the carotid is left unstented and to be repaired or closed in the infrequent, exceptional case of injury. Dissecting tumour from carotid involves mechanical trauma to the artery wall with the inherent, potential morbidity of a thrombo-embolic event, especially in the atheromatous vessel. Preoperative assessment of the artery condition is mandatory, and the uneventful management of the carotid remains a prior concern.

Conclusions

Our microsurgical approach to carotid PGL developed from the experience gained in the PGL of the jugular foramen extending to the neck and the vagal PGL. The crucial points of such technique consist of intra-fascial dissection of the tumour and periadventitial dissection of the carotids. In particular, the steps to specifically face carotid PGL include preservation of the cervical fascia investing carotid and PGL in a single mass and carrying part of the blood supply to the tumour, haemostasis directly on the fascia, intrafascial dissection, identification of the tumour-carotids interface and safe dissection with preservation of the adventitia.

The outcomes of our series were favorable in terms of disease control, functional preservation, absence of peri-operative or postoperative complications. Overall, the limited size of our series allows only for cautious statements. However, the established use of microsurgery in other areas of the head and neck suggests that it may be a useful tool also in the treatment of carotid PGL, as it allows a potential reduction in morbidity, along with an improvement of intraoperative precision and safety in dissecting tumour from carotid and nerves.

Acknowledgements

The Authors are grateful to Mrs. Alison Garside for editing the English version of this manuscript.

Conflict of interest statement

The authors declare no conflict of interest.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

AM: conceptualisation, design, conduct, manuscript drafting, supervision; LF: conceptualisation, design, conduct, manuscript drafting, supervision; EZ: conceptualisation, design, conduct, manuscript drafting, supervision.

Ethical consideration

The research was conducted ethically, with all study procedures being performed in accordance with the requirements of the World Medical Association’s Declaration of Helsinki.

All patients signed an informed consent form regarding the processing and publication of their data, which were examined in agreement with the Italian privacy and sensitive data laws, and the internal regulations of the University Hospital of Padua.

Figures and tables

Figure 1.Microsurgical instrumentation: A) bipolar coagulation forceps; B) Freer-type elevators with semi-sharp straight and curved ends; C) two-way suction-irrigation tube.

Figure 2.Pre-operative MRI of a left-sided carotid body tumour: A) contrast enhanced axial T1 sequence (arrow: external carotid artery; arrowhead: internal carotid artery); B) axial T2 sequence (arrow: external carotid artery; arrowhead: internal carotid artery).

Figure 3.Surgical steps of carotid dissection from paraganglioma (PGL) (same case of left-sided carotid body tumour): A) initial view of the tumuor (T), enveloped by common tumour-carotid fascia (IC: internal carotid; V: vagus nerve); B) opening of common fascia (*common fascia; T: tumour); C) elevator on tumour-carotid interface; D) internal carotid artery freed from tumour on antero-lateral aspect; E) dissection of external carotid (EC) completes separation of the ventral part of tumour; F) dorsal aspect of internal carotid freed from tumour (T: tumour; IC: internal carotid; V: vagus nerve); G) blood spilling from origin of glomic artery on the posterior wall of carotid bifurcation, where the common carotid (CC) divides into external (EC) and internal carotid (IC) arteries; H) the same after bipolar coagulation.

References

1. Bryant JP, Wang S, Niazi T. Carotid body tumor microenvironment. Adv Exp Med Biol. 2020; 1296:151-162. DOI
2. Céruse P, Ambrun A, Cosmidis A. Paragangliomes latérocervicaux. EMC Oto-rhino-larayngologie. 2013; 8:1-15.
3. Gonzalez-Urquijo M, Castro-Varela A, Barrios-Ruiz A. Current trends in carotid body tumors: comprehensive review. Head Neck. 2022; 44:2316-2332. DOI
4. El-Naggar AK, Chan JKC, Grandis JR. WHO Classification of Head and Neck Tumours. WHO: Geneve; 2017.
5. Cleere EF, Martin-Grace J, Gendre A. Contemporary management of paragangliomas of the head and neck. Laryngoscope Investig Otolaryngol. 2022; 7:93-107. DOI
6. Van den Berg R. Imaging and management of head and neck paragangliomas. Eur Radiol. 2005; 15:1310-1318. DOI
7. Davila VJ, Chang JM, Stone WM. Current surgical management of carotid body tumors. J Vasc Surg. 2016; 64:1703-1710. DOI
8. Robertson V, Poli F, Hobson B. A systematic review and meta-analysis of the presentation and surgical management of patients with carotid body tumours. Eur J Vasc Endovasc Surg. 2019; 57:477-486. DOI
9. Suárez C, Rodrigo JP, Mendenhall WM. Carotid body paragangliomas: a systematic study on management with surgery and radiotherapy. Eur Arch Otorhinolaryngol. 2014; 271:23-34. DOI
10. Darouassi Y, Alaoui M, Mliha Touati M. Carotid body tumors: a case series and review of the literature. Ann Vasc Surg. 2017; 43:265-271. DOI
11. Lozano FS, Muñoz A, de Las Heras JA. Simple and complex carotid paragangliomas. Three decades of experience and literature review. Head Neck. 2020; 42:3538-3550. DOI
12. La Muraglia GM, Fabian RL, Brewster DC. The current surgical management of carotid body paragangliomas. J Vasc Surg. 1992; 15:1035-1038. DOI
13. Grodinski M, Holyoke EA. The fasciae and fascial spaces of the head, neck and adjacent regions. Am J Anat. 1938; 63:367-408.
14. Hayashi S. Histology of the human carotid sheath revisited. Okajimas Folia Anat Jpn. 2007; 84:49-60.
15. Padget DH. The development of the cranial arteries in the human embryo. Contrib Embryol Carneg Inst. 1948; 32:205-261.
16. Telischak N, Gross BA, Zeng Y. The glomic artery supply of carotid body tumors and implications for embolization. J Clin Neurosci. 2014; 21:1176-1179. DOI
17. Ceccato GHW, Rassi MS, Borba LAB. Microsurgical resection of multiple giant glomus tumors. J Neurol Surg B Skull Base. 2019; 80:S385-S388. DOI
18. Degollado-García J, Medina-Pizarro M, Cano-Velazquez G. Microsurgical treatment of carotid body tumors using periadventitial dissection: analysis of outcomes and prognostic factors in a neurological referral center. Surg Neurol Int. 2022; 13:487. DOI
19. Janakiram TN, Nadakkavukaran J, Bhatia Sharma S. A hybrid approach towards successful resection of a huge carotid body paraganglioma using coil embolization and traditional surgical techniques. Indian J Otolaryngol Head Neck Surg. 2019; 71:136-139. DOI
20. Parisien-La Salle S, Chbat J, Lacroix A. Post-operative recurrences in patients operated for pheochromocytomas and paragangliomas: new data supporting lifelong surveillance. Cancers (Basel). 2022; 14:2942. DOI
21. Hallett JW, Nora JD, Hollier LH. 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.
22. Lim J, Kim J, Kim S. Surgical treatment of carotid body paragangliomas: outcomes and complications according to the Shamblin classification. Clin Exp Otorhinolaryngol. 2010; 3:91. DOI
23. Mazzoni A, Zanoletti E. Observation and partial targeted surgery in the management of tympano-jugular paraganglioma: a contribution to the multioptional treatment. Eur Arch Otorhinolaryngol. 2016; 273:635-642. DOI
24. Mazzoni A. The petro-occipital trans-sigmoid approach for lesions of the jugular foramen. Skull Base. 2009; 19:48-56. DOI
25. Zanoletti E, Mazzoni A. Vagal paraganglioma. Skull Base. 2006; 16:161-167. DOI
26. Prasad SC, Laus M, Al-Ghamdi S. Update in the classification and the role of intra-arterial stenting in the management of carotid body paragangliomas. Head Neck. 2019; 41:1379-1386. DOI
27. Abu-Ghanem S, Yehuda M, Carmel N. Impact of preoperative embolization on the outcomes of carotid body tumor surgery: a meta-analysis and review of the literature. Head Neck. 2016; 38:E2386-E2394. DOI