Annals of Vascular Surgery
Volume 20, Issue 4 , Pages 435-439, July 2006

Carotid Body Tumor Resection: Does the Need for Vascular Reconstruction Worsen Outcome?

  • J. Joshua Smith, MD

      Affiliations

    • Division of Vascular Surgery, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
    • ,
  • Marc A. Passman, MD

      Affiliations

    • Division of Vascular Surgery, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
    • Corresponding Author InformationCorrespondence to: Marc A. Passman, MD, Section of Vascular Surgery, University of Alabama at Birmingham, BDB 503 1808, 7th Avenue South, Birmingham, AL, 35294-0012, USA
    • ,
  • Jeffery B. Dattilo, MD

      Affiliations

    • Division of Vascular Surgery, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
    • ,
  • Raul J. Guzman, MD

      Affiliations

    • Division of Vascular Surgery, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
    • ,
  • Thomas C. Naslund, MD

      Affiliations

    • Division of Vascular Surgery, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN, USA
    • ,
  • James L. Netterville, MD

      Affiliations

    • Department of Otolaryngology, Vanderbilt University Medical Center, Nashville, TN, USA

Article Outline

We evaluated outcomes after carotid body tumor resection (CBR) requiring vascular reconstruction. Patients undergoing CBR at an academic medical center between 1990 and 2005 were identified. Medical records were retrospectively reviewed for clinical data, operative details, Shamblin's classification, tumor pathology, complications, and mortality. Comparisons were performed between those undergoing CBR alone and CBR requiring vascular reconstruction (CBR-VASC). Of the 71 CBRs performed in 62 patients, 16 required vascular reconstruction (23%). Although there was no difference in mean tumor size (CBR 29.1 ± 11.9 mm, CBR-VASC 32.5 ± 9.9 mm; p = 0.133), carotid body tumors were more commonly Shamblin's I when CBR was performed alone (CBR 53% vs. CBR-VASC 25%, p = 0.045) and Shamblin's II/III when vascular reconstruction was required (CBR 47% vs. CBR-VASC 75%, p = 0.045). There was also a significant difference in malignant tumor pathology when vascular reconstruction was required (CBR 4.4% vs. CBR-VASC 25%, p = 0.034). Cranial nerve dysfunction was higher in patients requiring vascular repair (CBR 27% vs. CBR-VASC 63%, p = 0.012), but there was no difference in baroreflex failure (CBR 7.27% vs. CBR-VASC 0%, p = 0.351), Horner's syndrome (CBR 5.5% vs. CBR-VASC 6.25%, p = 0.783), or first bite syndrome (CBR 7.27% vs. CBR-VASC 12.5%, p = 0.877). There were no perioperative strokes in either group, and one death was unrelated to operation. When required, carotid artery reconstruction at the time of CBR can be performed safely. Although cranial nerve dysfunction is more common when vascular repair is required, this is more likely related to locally advanced disease and tumor pathology rather than operative techniques.

 

Back to Article Outline

INTRODUCTION 

Carotid body tumors (CBTs) are rare and usually benign neoplasms, but because of unpredictable malignant potential, early and complete surgical resection is often recommended. Many groups have reported low perioperative mortality and varying degrees of postoperative neurological morbidity with complete resection.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 Even with advances in earlier diagnosis, preoperative embo-lization, improved intraoperative management, and modern vascular techniques, the rate of postoperative cranial nerve deficits remains relatively high. Although some of these studies have included information regarding adjunctive vascular repair, specific information regarding added morbidity from operations requiring vascular reconstruction is lacking. The purpose of this study was to evaluate outcomes after CBT resection (CBR) requiring vascular reconstruction.

Back to Article Outline

METHODS 

All patients undergoing CBR between 1990 and 2005 at Vanderbilt University Medical Center were identified. Those patients requiring additional vascular reconstruction (CBR-VASC) were also identified for subgroup analysis.

Medical records were retrospectively reviewed for clinical data, operative details, tumor pathology, complications,13 and mortality. Shamblin's classification was used for tumor stratification.14 Shamblin's type I is defined as minimal displacement of the internal and external carotid arteries by the tumor. Shamblin's type II is defined as tumor indentation by the external and internal carotid arteries making a deep groove within the tumor, with the hypoglossal and superior laryngeal nerves across the tumor surface. Shamblin's type III involves encasement of the carotid artery and nerves by the tumor.

The operative technique for CBR is well described.10, 15 The need for preoperative tumor embolization was determined selectively at the discretion of the operating surgeon (J.L.N.). The decision for vascular repair was made intraoperatively, with various techniques employed at the discretion of the vascular surgeons including primary repair, patch repair, ligation, and bypass (M.A.P., J.B.D., R.J.G., T.C.N.).

Statistical comparisons were performed between the CBR alone and the CBR-VASC groups. Data were analyzed using the Fisher exact method, Wilson's confidence intervals (CIs), and t-tests for comparison of means.

Back to Article Outline

RESULTS 

Over a 15-year period, 71 CBRs were performed in 62 patients, with 16 requiring vascular reconstruction (23%). The types of vascular repairs (n = 16) included interposition bypass with reversed saphenous vein graft (n = 9), primary repair (n = 4), primary repair with external carotid artery ligation (n = 1), patch repair (n = I), and endar-terectomy with patch repair (n = 1).

Demographics are shown in Table I, with no difference in patient populations with regard to CBT-related factors. Table II shows the tumor characteristics. There was no difference in mean tumor size (CBR 29.1 ± 11.9 mm vs. CBR-VASC 32.5 ± 9.9 mm, p = 0.133); however, there was a significant difference in malignant tumor histology (CBR 4.4% vs. CBR-VASC 25%, p = 0.034). Distribution based on Shamblin's classification is shown in Table III. CBTs were more commonly Shamblin's class I when CBR was performed alone (CBR 53% vs. CBR-VASC 25%, p = 0.045) and Shamblin's class II/III when vascular reconstruction was required (CBR 47% vs. CBR-VASC 75%, p = 0.045).

Table I. Comparison of demographics of patients with CBTs undergoing complete resection alone (CBR) and with vascular reconstruction (CBR-VASC)
DemographicsCBR (n = 46)CBR-VASC (n = 16)pWilson CI
Mean age (years)42.9 ± 12.945 ± 18.90.343aN/A
Gender
Men36%60%0.092bCBR 0.24–0.50
Women64%40% CBR-VASC 0.36–0.80
Hypertension19.1%20%0.684bCBR 0.10–0.33
CBR-VASC 0.07–0.45
Family history of CBT22%12.5%0.343bCBR 0.12–0.36
CBR-VASC 0.03–0.36
Tobacco abuse26%25%0.61bCBR 0.16–0.40
CBR-VASC 0.10–0.49
Peripheral vascular disease or prior vascular operation8.69%18.8%0.933bCBR 0.03–0.20
CBR-VASC 0.07–0.43
Bilateral tumor61%37.5%0.092bCBR 0.46–0.74
CBR-VASC 0.18–0.61

a t-test.

b Fisher's exact method and Wilson's CI.

Table II. Comparison of tumor pathology in patients with CBTs undergoing complete resection alone (CBR) and with vascular reconstruction (CBR-VASC)
CBR (n = 46)CBR-VASC (n = 16)pWilson CI
Tumor size (mm)29.1 ± 11.932.5 ± 9.90.133aN/A
Malignant/metastatic4.4%25%0.034bCBR 0.01–0.14
CBR-VASC 0.10–0.49

a t-test.

b Fisher's exact method and Wilson's CI.

Table III. Comparison of Shamblin's tumor type in patients with CBTs undergoing complete resection alone (CBR) and with vascular reconstruction (CBR-VASC)
Shamblin's typeCBR (n = 55)CBR-VASC (n = 16)paWilson CI
I53%25%0.045CBR 0.39–0.65
CBR-VASC 0.10–0.49
II/III47%75%0.045CBR 0.35–0.60
CBR-VASC 0.51–0.89

a Fisher's exact method and Wilson's CI.

Preoperative tumor embolization was performed in 40.1% (n = 71) of all resections (CBR 29.6% [n = 21] vs. CBR-VASC 11.3% [n = 8]). No difference was found regarding use of preoperative embolization when stratified by CBR vs. CBR-VASC (CBR 38% vs. CBR-VASC 50%, p = 0.287) as shown in Table IV.

Table IV. Relationship of preoperative embolization to type of resection in patients with CBTs undergoing complete resection alone (CBR) and with vascular reconstruction (CBR-VASC)
Preop embolizationYes (n = 55)No (n = 16)paWilson CI
CBR38%62%0.287CBR 0.27–0.52
CBR-VASC50%50%0.287CBR-VASC 0.28–0.72

a Fisher's exact method and Wilson's CI.

The overall complication rate was 55% (n = 71) for at least one perioperative problem (CBR 47.2% vs. CBR-VASC 82%, p = 0.015). Table V summarizes neurological complications of CBR. Cranial nerve dysfunction was higher in patients requiring vascular repair (CBR 27% vs. CBR-VASC 63%, p = 0.012), but there was no difference in barore-flex failure (CBR 7.27% vs. CBR-VASC 0%, p = 0.351), Homer's syndrome (CBR 5.5% vs. CBR-VASC 6.25%, p = 0.783), or first bite syndrome (CBR 7.27% vs. CBR-VASC 12.5%, p = 0.877). There were no perioperative strokes in either group, and one perioperative death in the CBR group was unrelated to operation (invasive pulmonary aspergillosis).

Table V. Complications of patients with CBTs undergoing complete resection alone (CBR) and with vascular reconstruction (CBR-VASC)
ComplicationCBR (n = 55)CBR-VASC (n = 16)paWilson CI
Cranial nerve injury27%63%0.012CBR 0.17–0.40
CBR-VASC 0.39–0.82
Baroreflex syndrome7.27%0%0.351CBR 0.03–0.17
CBR-VASC 0–0.19
Horner's syndrome5.5%6.25%0.783CBR 0.02–0.15
CBR-VASC 0.003–0.28
First bite syndrome7.27%12.5%0.877CBR 0.03–0.17
CBR-VASC 0.03–0.36

a Fisher's exact method.

Back to Article Outline

DISCUSSION 

Much has been published about the operative management of CBTs. The usual resection techniques involve dissection of the carotid body, staying within the periadventitial plane to avoid damage to the surrounding arterial structures.16 Vascular reconstruction eventually becomes necessary as the tumor becomes more involved with the vessels, with an incidence ranging 4–52% in the literature (Table VI) and being 23% in this study. Limited information exists on the specifics of vascular repair during CBR and the resultant postoperative complications. Although some authors have advocated vascular sacrifice,1 our practice has been to reconstruct the carotid artery, most commonly using interposition bypass vein grafts.

Table VI. Review of the literature on CBR in series with more than 10 patients
Number of patientsVascular repair% Neurological complications
Luna Ortiz et al. (2005)266 (69 CBTs)3 (4.4%)49%
Davidovic et al. (2005)1212 (12 CBTs)5 (41.7%)25%
Dardik et al. (2002)325 (27 CBTs)11 (41%)33%
Patetsios et al. (2002)429 (34 CBTs)8 (28%)46%
Plukker et al. (2001)1139 (45 CBTs)6 (13.3%)18%
Wang et al. (2000)729 (36 CBTs)-41%
Muhm et al. (1997)824 (28 CBTs)10 (35.7%)31.6%
Litle et al. (1996)921 (21 CBTs)11 (52.4%)45%
Netterville et al. (1995)1030 (46 CBTs)12 (26%)13%
Hallett et al. (1988)6139 (153 CBTs)∼50 (33%)40–46%

Reviewing the data published on vascular repair involved in GET operations, comparison of neurological complications with resection alone versus resection plus vascular repair is lacking. The results of the current study are unique regarding GET repair in the context of those resections that require vascular repair. Although there were no perioperative strokes, cranial nerve injury occurred at a higher frequency in patients who had undergone CBR with vascular repair. A 63% incidence of cranial nerve injury was noted with vascular repair versus a 27% incidence with CBR alone (p = 0.012). As a retrospective study, minimal follow-up data were available for review to determine the final outcome of the perioperative cranial nerve injuries, i.e., whether they were permanent or transient.

The current literature does not specifically address the need for vascular repair as it relates to postoperative nerve injury, the relationship of resection type with the common complications following repair, or the specific neurological or vascular complications related to tumor type. Luna-Ortiz et al.1 comment on an increase in the need for vascular sacrifice with a modified Sham-blin's classification that incorporates surgical bleeding and vascular/neurological involvement. Litle et al.9 noted a postoperative morbidity trend similar to ours but did not directly relate vascular reconstruction with postoperative neurological injury. Hallett et al.,6 in one of the largest series to date, showed no changes in postoperative cranial nerve dysfunction (40–46%) over a period of 50 years. Interestingly, their mortality improved with better vascular techniques and monitoring, but the incidence of postoperative cranial nerve dysfunction remained essentially unchanged. Maxwell et al.,17 in examining the Nationwide Inpatient Sample, found that mortality rates were higher during CBR when carotid endarterectomy was required. While perioperative cranial nerve injury reported in our study with resection alone is very consistent with, if not less than, that previously reported, it is difficult to determine if cranial nerve injury with vascular repair is higher than what other studies might have found if there had been a focus on the relationship of cranial nerve injury and vascular repair.

In this study, when vascular repair was required, the tumor was more intimately involved with the arteries, as represented by the differences in Shamblin's classifications. In this study, vascular repair was required in 75% (p = 0.045) of cases classified as Shamblin's type II/III vs. 25% (p = 0.045) of those classified as Shamblin's I. Although there was no difference in mean tumor size (CBR 29.1 ± 11.9 mm vs. CBR-VASC 32.5 ± 9.9 mm, p = 0.133), there was a significant difference in malignant tumor histology when vascular reconstruction was required (CBR 4.4% vs. CBR-VASC 25%, p = 0.034). It seems that the higher postoperative neurological sequelae observed in this study are likely due to involvement of the tumor with the surrounding structures and not the vascular reconstruction itself. This dysfunction seems to be more likely related to locally advanced disease and malignant tumor pathology rather than to the complications of vascular repair.

Although preoperative embolization has been touted as a tool to decrease the morbidity associated with CBR, this recommendation is mostly from case series.18, 19 Litle et al.9 showed no difference in outcomes when comparing preoperative embolization versus no embolization. In this study, no difference was found regarding the use of preoperative embolization when stratified by CBR versus CBR-VASC. Due to the retrospective nature of this study, no specific data were extracted to determine if preoperative embolization was useful in resection of the tumor or in relation to tumor type.

In conclusion, cranial nerve dysfunction is more common during CBR when vascular reconstruction is required. This dysfunction is more likely related to locally advanced disease and malignant tumor histology rather than to complications of the vascular repair. When required, carotid artery reconstruction at the time of CBR can be performed safely. Any additional morbidity or mortality seems to reflect advanced tumor pathology rather than operative technique.

Statistical analysis was performed in collaboration with Rafe Donahue, PhD, Department of Biostatistics and Section of Surgical Sciences, Van-derbilt University Medical Center.

Back to Article Outline

REFERENCES 

  1. Luna-Ortiz K , Rascon-Ortiz M , Villavicencio-Valencia V , Herra-Gomez A . Does Shamblin's classification predict postoperative morbidity in carotid body tumors? A proposal to modify Shamblin's classification . Eur Arch Otorhinolaryngol . 2006;263(2):171–175
  2. Luna-Ortiz K , Rascon-Ortiz M , Villavicencio-Valencia V , Granados-Garcia M , Herrera-Gomez A . Carotid body tumors: review of a 20-year experience . Oral Oncol . 2005;41:56–61
  3. Dardik A , Eisele DW , Williams GM , Perler BA . A contemporary assessment of carotid body tumor surgery . Vasc Endovascular Surg . 2002;36:277–283
  4. Patetsios P , Gable DR , Garrett WV , et al.   Management of carotid body paragangliomas and review of a 30-year experience . Ann Vase Surg . 2002;16:331–338
  5. Westerband A , Hunter GC , Cintora I , et al.   Current trends in the detection and management of carotid body tumors . J Vase Surg . 1998;28:84–92
  6. Hallett JW , Nora JD , Hollier LH , Cherry KJ , Pairolero PC . Trends in neurovascular complications of surgical management for carotid body and cervical paragangliomas: a fifty-year experience with 153 tumors . J Vase Surg . 1988;7:284–291
  7. Wang SJ , Wang MB , Barauskas TM , Calcaterra TC . Surgical management of carotid body tumors . Otolaryngol. Head Neck Surg . 2000;123:202–206
  8. Muhm M , Polterauer P , Gstottner W , et al.   Diagnostic and therapeutic approaches to carotid body tumors. Review of 24 patients . Arch Surg . 1997;132:279–284
  9. Litle VR , Reilly LM , Ramos TK . Preoperative embolization of carotid body tumors: when is it appropriate? . Ann Vase Surg . 1996;10:464–468
  10. Netterville JL , Reilly KM , Robertson D , Reiber ME , Armstrong WB , Childs P . Carotid body tumors: a review of 30 patients with 46 tumors . Laryngoscope . 1995;105:115–126
  11. Plukker JT , Brongers EP , Vermey A , Krikke A , van den Dungen JJ . Outcome of surgical treatment for carotid body paraganglioma . Br J Surg . 2001;88:1382–1386
  12. Davidovic LB , Djukic VB , Vasic DM , Sindjelic RP , Duvnjak SN . Diagnosis and treatment of carotid body paraganglioma: 21 years of experience at a clinical center of Serbia . World J Surg Oncol . 2005;3:10
  13. Sniezek JC , Sabri AN , Netterville JL . Paraganglioma surgery: complications and treatment . Otolaryngol Clin North Am . 2001;34:993–1006 vii
  14. Shamblin WR , ReMine WH , Sheps SG , Harrison EG . Carotid body tumor (chemodectoma). Clinicopathologic analysis of ninety cases . Am J Surg . 1971;122:732–739
  15. Pellitteri PK , Rinaldo A , Myssiorek D , et al.   Paragangliomas of the head and neck . Oral Oncol . 2004;40:563–575
  16. Gordon-Taylor G . On carotid body tumours . Br J Surg . 1940;28:163–172
  17. Maxwell JG , Jones SW , Wilson E , et al.   Carotid body tumor excisions: adverse outcomes of adding carotid endarterectomy . J Am Coll Surg . 2004;198:36–41
  18. Kafie FE , Freischlag JA . Carotid body tumors: the role of preoperative embolization . Ann Vase Surg . 2001;15:237–242
  19. Tripp HF , Fail PS , Beyer MG , Chaisson GA . New approach to preoperative vascular exclusion for carotid body tumor . J Vase Surg . 2003;38:389–391

 Presented at Sixteenth Annual Winter Meeting of the Peripheral Vascular Surgery Society, Park City, UT, January 26–29, 2006.

PII: S0890-5096(06)61458-1

doi:10.1007/s10016-006-9093-0

Annals of Vascular Surgery
Volume 20, Issue 4 , Pages 435-439, July 2006

Article Tools