surgical outcome of carotid body tumour resection after … · 2013. 2. 11. · hypoglossal nerve...

of 5

Original research study

Licensee OA Publishing London 2013. Creative Commons Attribution License (CC-BY)

Com

petin

g in

tere

sts:

non

e de

clar

ed. C

onfli

ct o

f int

eres

ts: n

one

decl

ared

.A

ll au

thor

s co

ntrib

uted

to th

e co

ncep

tion,

des

ign,

and

pre

para

tion

of th

e m

anus

crip

t, a

s w

ell a

s re

ad a

nd a

ppro

ved

the

final

man

uscr

ipt.

All

auth

ors

abid

e by

the

Ass

ocia

tion

for M

edic

al E

thic

s (A

ME)

eth

ical

rule

s of

dis

clos

ure.

For citation purposes: Abdel-Aziz T, Lehmann M, Dietrich U, Ebmeyer J, Sudhoff H. Surgical outcome of carotid body tumour resection after percutaneous embolizationusing Onyx®, an ethylene-vinyl alcohol copolymer. Head Neck Oncol. 2013 Jan 04;5(1):5.

Surgical outcome of carotid body tumour resection after percutaneous embolizationusing Onyx®, an ethylene-vinyl

alcohol copolymer†

T Abdel-Aziz1*, M Lehmann1, U Dietrich2, J Ebmeyer1, H Sudhoff1

AbstractIntroductionCarotid body tumours (CBTs) are highly vascularised tumours that can only be treated curatively by com-plete surgical resection. Preoperative embolization is widely used to reduce intraoperative bleeding and minimise surgical complications.ObjectiveTo assess the surgical outcome of CBT resection after percutaneous embolization using Onyx®.MethodsRetrospective analysis of five consecu-tive patients with CBTs was performed from August, 2008 until December, 2011 after embolization via percuta-neous intratumoural injection using Onyx®, and the results were compared with those available in the literature.ResultsIn four of five patients, complete devas-cularisation was achieved. No compli-cations occurred during embolization. All CBTs were resected completely. Duration of surgery varied between 35 and 289 min, with an average of 148 min. Mean estimated intraoper-ative blood loss was 327 mL (range, 40–1400 mL). We experienced one

Horner’s syndrome, one temporary hypoglossal nerve paresis and a per-manent paresis of the recurrent laryngeal nerve.ConclusionPreoperative embolization reduces intraoperative blood loss, improves operative field visualisation and con-secutively diminishes damage to the nerval and vascular structures. Percu-taneous embolization via direct intra-tumoural puncture appears to achieve a higher degree of devascularisation by affecting arterial vessels, capillary bed and venous drainage. Existing data suggest a decrease in intraoperative blood loss and fluoroscopy times over transarterial embolization.

IntroductionCarotid body tumours (CBTs) are highly vascularised tumours and belong to a group of paragangliomas. They are considered a rarity, with a 0.001% overall incidence and 0.6% proportion of all head and neck tumours1. CBTs have a female to male ratio of 3:12. Further, 90% of all CBTs occur sporadically, whereas 10% are of familial origin. To date, several mutations of succinate dehydrogenase subunits have been identified. Although CBTs can arise at any age, incidence increases from the 4th decade in spo-radic and between the 2nd and 3rd decade in hereditary CBTs1. Bilaterality is seen in 5% of sporadic cases and up to 30% of familial CBT cases3,4.

The carotid body is located in the posteromedial portion of the carotid bifurcation within the adven-titia. It serves as a vascular chemore-ceptor, detecting changes in O2- and CO2-partial pressure as well as changes

in acid–base balance. Innervation is supplied by the glossopharyngeal nerve2. Although vascular supply of the carotid body occurs via the carotid body artery arising off the carotid bifurcation5, CBTs are most commonly supplied by branches of the external carotid artery, especially the ascending pharyngeal and occipital arteries. Additional vascular supply may also occur via branches of the internal carotid or vertebral arteries2.

Clinically, most CBTs present as slow growing, painless masses in the lateral neck. Of 75% CBTs that are inci-dentally diagnosed, only 25% cause medical conditions such as a thrill, dys-phagia, cranial nerve deficits, carotid sinus syndrome or cerebrovascular complications.

Most CBTs are hormonally inactive2. Further, 94% CBTs are benign; only 6% CBTs undergo malignant transforma-tion6. However, even benign tumours have the potential to locally invade and destruct adjacent cranial nerves, leading to permanent nerve palsies.

Complete surgical resection is the only curative therapy, and therefore, it is the therapeutic gold standard. Radia-tion therapy is considered a treatment option only in elderly patients and in patients with vital con traindications to surgical intervention7.

While intraoperative mortality has declined below 1%2, the greatest peri-operative risk for the patient is exten-sive blood loss, which also impairs adequate visualisation of the surgical field, thus increasing the risk for nerval injury. However, it is important to empha sise that due to potential tumour invasion of nerval structures, sometimes the sacrifice of nerves,

* Corresponding authorEmail: [email protected]† Results of the article were presented, in

part, as an oral presentation at the 83rd Annual Meeting of the German Society of Otorhinolaryngology, Head and Neck Surgery on 17 May 2012.

1 Department of Otorhinolaryngology, Head and Neck Surgery, Klinikum Bielefeld, Academic Teaching Hospital, University of Münster, Bielefeld, Germany

2 Department of Diagnostic and Interventional Radiology and Neuroradiology, Evangelic Hospital, Bielefeld, Germany

of 5



Com

petin

g in

tere

sts:

non

e de

clar

ed. C

onfli

ct o

f int

eres

ts: n

one

decl

ared

.A

ll au

thor

s co

ntrib

uted

to th

e co

ncep

tion,

des

ign,

and

pre

para

tion

of th

e m

anus

crip

t, a

s w

ell a

s re

ad a

nd a

ppro

ved

the

final

man

uscr

ipt.

All

auth

ors

abid

e by

the

Ass

ocia

tion

for M

edic

al E

thic

s (A

ME)

eth

ical

rule

s of

dis

clos

ure.


regardless of surgical field visibility, is inevitable in order to achieve complete tumour resection.

To minimise intraoperative blood loss (IBL) and its complications, many surgeons prefer preoperative tumour embolization.

Two different strategies of emboli-zation have been proposed: transarte-rial, first described by Schick et al. in 19808,9 and direct percutaneous intratumoural puncture, first described by Casasco et al. in 19949,10.

Comparison of both routes of embolization reveals several impor-tant advantages for the percutaneous approach. While transarterial emboli-zation only affects the arterial blood supply, direct intratumoural applica-tion of the embolic agent is able to occlude arterial supply, the capillary bed and the venous vessels, resulting in an overall higher degree of devas-cularisation11–13. Furthermore, tortuous vascular architecture can make micro-catheterisation of feeding vessels very difficult or even impossible12. Addi-tionally, in case of multiple feeding vessels, time-consuming microcathe-terisation of every vessel is required. These difficulties are bypassed in direct

intratumoural embolization, leading to reduced fluoroscopy times12.

For these reasons, at our institution, we prefer preoperative embolization of CBTs via direct intratumoural puncture.

Most reports of direct intratum-oural embolization have used n-butyl cyanoacrylate (n-BCA) or polyvinyl alcohol12. In the past several years, a few reports have been published prop-agating the use of Onyx® (ev3, Irvine, CA, USA), a non-adhesive liquid embolic agent. It consists of ethylene-vinyl alcohol copolymer (EVOH) dissolved in dimethyl sulphoxide (DMSO) and suspended micronized tantalum pow-der as a contrast agent for visualisa-tion during fluoroscopy. On contact with blood, DMSO dissipates, leading to precipitation of EVOH (Figure 1).

Several advantages of Onyx® over n-BCA have been proposed. Onyx® pol-ymerises slower than n-BCA, allowing intratumoural application in a more controlled and precise manner. Addi-tionally, injection can be halted if needed to evaluate the achieved degree of devascularisation or to repo-sition the injection needle for larger sized tumours7.

For these reasons, we considered the use of Onyx® for preoperative percutaneous tumour embolization as favourable.

Methods and materialIn our setting, percutaneous emboli-zation was performed in collaboration with the Department of Diagnostic and Interventional Radiology and Neu-roradiology at the Evangelic Hospital in Bielefeld, Germany. Generally, embo-lization is performed 24 h prior to surgery under general anaesthesia. The detailed procedural description has been previously published14. After angiographic display of the common carotid artery (Figure 2A), a protective microballoon is inserted into the inter-nal carotid artery until the level of the observed tumour blush. The microb-alloon can be temporarily inflated to avoid retrograde penetration of the embolic agent into the carotid artery14. Subsequently, the tumour is percuta-neously punctured using the road-map technique and intratumoural access is confirmed using contrast dye injec-tion (Figure 2B). Finally, percutane-ous injection of Onyx® was performed under fluoroscopic roadmap guidance. Intraprocedural angiography was used to control the achieved degree of devas-cularisation (Figure 2C).

ResultsFrom August, 2008 to December, 2011, five CBTs were preoperatively embo-lized and subsequently resected com-pletely by subadventitial tumour dissection. All patients were female with a mean age of 58.2 (32–78) years. All CBTs were benign and occurred sporadically. No CBTs displayed endo-crine activity (Table 1).

An average of 4.4 (1–9) mL of Onyx® was needed for devascularisation. In four of five cases, complete devascu-larisation could be achieved; in one case devascularisation was nearly complete, as controlled by angiogra-phy. No complications were observed during embolization (Table 1).

Figure 1: Histologic image of embolized CBT. Blood vessels filled with Onyx®, arrow (H/E staining, 40×).

of 5



Com

petin

g in

tere

sts:

non

e de

clar

ed. C

onfli

ct o

f int

eres

ts: n

one

decl

ared

.A

ll au

thor

s co

ntrib

uted

to th

e co

ncep

tion,

des

ign,

and

pre

para

tion

of th

e m

anus

crip

t, a

s w

ell a

s re

ad a

nd a

ppro

ved

the

final

man

uscr

ipt.

All

auth

ors

abid

e by

the

Ass

ocia

tion

for M

edic

al E

thic

s (A

ME)

eth

ical

rule

s of

dis

clos

ure.


Duration of surgery varied between 35 and 289 min, with an average of 148 min. Mean estimated IBL was 327 mL (range, 40 to 1400 mL; Table 2). In the case with 1400 mL blood loss, surgical resection proved exquisitely difficult due to an unusual cranial localisation of the carotid bifurcation. Additionally, complete preoperative embolization had not been achieved with feeding branches of the carotid bifurcation in remaining patients (Table 1). Excluding this case as a runaway value yielded

an average IBL of 54 mL. Carotid arteries were not injured in any case, and no vascular reconstruction was required. Postoperatively, one patient suffered from Horner’s syndrome, and the patient with the unusual anatomy suffered from a permanent paresis of the recurrent laryngeal nerve and a temporary paresis of the hypoglossal nerve that resolved com-pletely within 9 months after surgery. The nerves themselves had not been sacrificed during surgery as confirmed by intraoperative nerve stimulation.

DiscussionOnly few reports exist on the surgical outcome after preoperative direct intratumoural embolization of CBTs using Onyx®. They all cover only small case series, but results seem to resemble the ones obtained at our institution.

Shah et al.15 retrospectively reviewed a consecutive series of seven CBTs that had undergone preoperative direct percutaneous Onyx® emboliza-tion. An average of 8.7 (4.2–18) mL of Onyx was used. Average estimated IBL was 55 (15–80) mL. Complica-tions during embolization did not oc-cur, surgical complications were not addressed.

Wanke et al.14 in a case series of four patients with six CBTs required an average of 4.75 (3–9.1) mL Onyx for complete devascularisation. Mean surgical duration was 151 (80–280) min. In two patients with bilat eral CBTs, surgical resection was per-formed one at a time. No complica-tions occurred during embolization or surgical resection.

Compared with older literature reports in which CBTs were resected either without prior embolization or after transarterial embolization, the data available so far indicate a benefit of the aforementioned preoperative intratumoural Onyx® embolization over such methods. LaMuraglia et al.16 retrospectively investigated a series of 17 patients with 19 CBTs treated in a 10-year period from 1982 to 1991. In 11 CBTs, complete preoperative tran-sarterial embolization of all afferent vessels was performed using polyvinyl alcohol beads and gel foam. Average IBL differed significantly (P < 0.02) among both cohorts, with 373 ± 213 mL in the embolized and 609 ± 564 mL in the non-embolized cohort. Both cohorts did not differ significantly in tumour surface area. Surgical time did not differ significantly, with 4.1 h in the embolized and 4.5 h in the non-embolized group. One attempt of preoperative embolization had to be terminated due to extreme tortuos-ity of feeding vessels, making stable catheter positioning impossible. This patient developed a transient episode of aphasia. Regarding surgical compli-cations, two cases of Horner’s syn-drome and two cases of tenth nerve injuries resulting in vocal cord pare-sis were reported.

CBTs are highly vascularised tumours that require surgical excision for curative treatment. Preoperative embolization has proved to signif-icantly reduce IBL and therefore is helpful in improving visualisation of the operative field. This reduces the likeliness of damage to nerval and vascular structures15–17.

Figure 2A: Angiography of the com-mon carotid artery. AP view (Case 5) shows a large CBT splaying internal and external carotid arteries.

Figure 2B: Percutaneous needle punc-ture of the tumour and injection of contrast dye. Intra-arterial catheter and balloon catheter are placed medially in the internal carotid artery.

Figure 2C: Postembolization angiogra-phy demonstrates almost total devas-culari sation of the tumour with a small remnant of contrast enhancement at the medial border.

of 5



Com

petin

g in

tere

sts:

non

e de

clar

ed. C

onfli

ct o

f int

eres

ts: n

one

decl

ared

.A

ll au

thor

s co

ntrib

uted

to th

e co

ncep

tion,

des

ign,

and

pre

para

tion

of th

e m

anus

crip

t, a

s w

ell a

s re

ad a

nd a

ppro

ved

the

final

man

uscr

ipt.

All

auth

ors

abid

e by

the

Ass

ocia

tion

for M

edic

al E

thic

s (A

ME)

eth

ical

rule

s of

dis

clos

ure.


Percutaneous embolization via direct intratumoural puncture appears to achieve a higher degree of devascu-larisation by affecting arterial vessels, capillary bed and venous drainage. Additionally, it shortens fluoroscopy time and under certain conditions, even impossible microcatheterisation of small tortuous feeding vessels is no longer necessary.

Although our group, similar to Shah et al.15 and Wanke et al.14, could not observe any complications during per-cutaneous Onyx® embolization, one

interesting complication was reported by Wiegand et al.11 in 2010. After embolization of a large CBT measuring 7 × 3 × 4 cm and necessitating the use of 20 mL Onyx®, postembolic tumour swelling was observed that caused a permanent Horner’s syndrome and transient paresis of the hypoglossal and glossopharyngeal nerves, which recovered within 7 days after tumour resection.

One case of embolization of the superior sagittal sinus with subse-quent haemorrhagic venous infarct

requiring emergency decompression craniectomy and frontal lobectomy after Onyx® embolization of an intrac-ranial frontal parasagittal meningioma raised the question of a possible increased risk for venous spread of the embolic agent when compared with the transarterial approach12.

Only few case series have investi-gated the efficacy of preoperative per-cutaneous Onyx® embolization, and existing data suggest a decrease in IBL and fluoroscopy times over transarte-rial embolization.

Reduced blood loss is a prerequi-site to facilitate visualisation of the surgical field and avoid unintended injury of adjacent nerval and vascular structures. It shortens the duration of surgery and consequently reduces perioperative morbidity, leading to an accelerated postoperative recovery. However, it is important to emphasise that due to the possible tumour inva-sion of nerves, nerval sacrifice some-times is inevitable to achieve complete tumour resection.

The few existing case series only comprise a small number of patients. To date, there is no study from a sin-gle institution directly comparing the two techniques of embolization, tran-sarterial vs. percutaneous. Similarly, among the different agents applicable for percutaneous embolization, such as n-BCA and Onyx®, no comparative clin-ical trial from a single institution exists. Blinded studies will hardly be accom-plishable due to the characteristic garlic-like breath and taste of Onyx®11. However, more and larger case series are required to definitely determine the value of percutaneous preopera-tive Onyx® embolization of CBTs.

References1. Schipper J, Boedeker CC, Maier W, Neumann HP. Paragangliomas in the head-/neck region. I: Classification and diagnosis. Head Neck Oncol. 2004 Jun; 52(6):569–74.2. Sezer Ö, Tjan TDT, Scheld HH, Hoffmeier A, Welp H. Glomustumore. Z Herz- Thorax- Gefäßchir. 2009;23:131–9.

Table 1 Patients and embolization results

Patient/age (years)/sex (M, F)

CBT size (cm) Onyx (mL) Devascularisation Complications

1/32/F 2.9 × 3.9 × 2.2 9 Complete No

2/41/F 2.3 × 1.9 × 2.8 4 Complete No

3/35/F 2.5 × 2.5 × 3.5 4 Complete No

4/78/F 3.2 × 2.0 × 1.4 1 Complete No

5/48/Fa 2.8 × 2.0 × 1.5 4 Almost complete

No

aDifficult anatomy due to unusual cranial location of the carotid bifurcation.

Table 2 Surgical results

Patient Duration of surgery (min)

IBL (mL) RBC transfusion

Surgical complications

Surgical removal

1 289 50 0 None Complete

2 133 75 0 Horner’s syndrome

Complete



5a 176 1400 2 Recurrent laryngeal nerve (permanent)

Complete

Hypoglossal nerve (temporary)

AVG 148 327 (54b) Complete

AVG, average (where applicable); RBC, red blood cell.aDifficult anatomy due to unusual cranial location of the carotid bifurcation.bAverage IBL with 1400 mL considered a runaway value.

of 5



Com

petin

g in

tere

sts:

non

e de

clar

ed. C

onfli

ct o

f int

eres

ts: n

one

decl

ared

.A

ll au

thor

s co

ntrib

uted

to th

e co

ncep

tion,

des

ign,

and

pre

para

tion

of th

e m

anus

crip

t, a

s w

ell a

s re

ad a

nd a

ppro

ved

the

final

man

uscr

ipt.

All

auth

ors

abid

e by

the

Ass

ocia

tion

for M

edic

al E

thic

s (A

ME)

eth

ical

rule

s of

dis

clos

ure.


3. Bernard RP. Carotid body tumors. Am J Surg. 1992 May;163(5):494–6.4. Kafie FE, Freischlag JA. Carotid body tumors: the role of preoperative emboli-zation. Ann Vasc Surg. 2001 Mar;15(2): 237–42.5. Kummer W. Paraganglien. In: Benning-hoff A, Drenckhahn D, editors. Anatomie. 16th ed. Munich, Germany: Elsevier; 2002:p6176. Lee JH, Barich F, Karnell LH, Robinson RA, Zhen WK, Gantz BJ, et al. American College of Surgeons Commission on Can-cer; American Cancer Society. National Cancer Data Base report on malignant paragangliomas of the head and neck. Cancer. 2002 Feb 1;94(3):730–7.7. Schipper J, Boedeker CC, Maier W, Neumann HP. Paragangliomas of the head and neck. Part 2: Therapy and follow-up. Head Neck Oncol. 2004 Jul;52(7):651–60.8. Schick PM, Hieshima GB, White RA, Fiaschetti FL, Mehringer CM, Grinnell VS, et al. Arterial catheter embolization

followed by surgery for large chemodec-toma. Surgery. 1980 Apr;87(4):459–64.9. Elhammady MS, Farhat H, Ziayee H, Aziz-Sultan MA. Direct percutaneous embolization of a carotid body tumor with Onyx. J Neurosurg. 2009 Jan;110(1): 124–7.10. Casasco A, Herbreteau D, Houdart E, George B, Tran Ba Huy P, Deffresne D, et al. Devascularization of craniofacial tumors by percutaneous tumor puncture. AJNR Am J Neuroradiol. 1994 Aug;15(7):1233–9.11. Wiegand S, Kureck I, Chapot R, Sesterhenn AM, Bien S, Werner JA. Early side effects after embolization of a carotid body tumor using Onyx. J Vasc Surg. 2010 Sep;52(3):742–5.12. Elhammady MS, Peterson EC, Johnson JN, Aziz-Sultan MA. Preoperative Onyx embolization of vascular head and neck tumors by direct puncture. World Neuro-surg. 2012 May–Jun ;77(5–6):725–30.13. Elhammady MS, Wolfe SQ, Ashour R, Farhat H, Moftakhar R, Lieber BB, et al. Safety and efficacy of vascular tumor

embolization using Onyx: is angiographic devascularization sufficient? J Neurosurg. 2010 May;112(5):1039–45.14. Wanke I, Jäckel MC, Goericke S, Panagiotopoulos V, Dietrich U, Forsting M. Percutaneous embolization of carotid paragangliomas using solely Onyx. AJNR Am J Neuroradiol. 2009 Sep;30(8):1594–7.15. Shah HM, Gemmete JJ, Chaudhary N, Pandey AS, Ansari SA. Preliminary experi-ence with the percutaneous embolization of paragangliomas at the carotid bifurca-tion using only ethylene vinyl alcohol copolymer (EVOH) Onyx. J Neurointerv Surg. 2012 Mar 1;4(2):125–9.16. LaMuraglia GM, Fabian RL, Brewster DC, Pile-Spellman J, Darling RC, Cambria RP, et al. The current surgical manage-ment of carotid body paragangliomas. J Vasc Surg. 1992 Jun;15(6):1038–44.17. Vogel TR, Mousa AY, Dombrovskiy VY, Haser PB, Graham AM. Carotid body tumour surgery: management and out-comes in the nation. Vasc Endovasc Surg. 2009 Oct–Nov;43(5):457–61.

surgical outcome of carotid body tumour resection after … · 2013. 2. 11. · hypoglossal nerve...

Documents