The surgical management of advanced cervical metastases with carotid artery involvement in patients with primary squamous cell carcinoma of the head and neck can be difficult. The authors retrospectively reviewed 73 patients over a 15-year period comparing the outcomes of carotid artery resection versus peeling the tumour off the carotid artery. Based on these findings, the authors suggest that in the absence of carotid wall involvement, nodal metastatic tumour should be peeled off the carotid artery where possible. This practice appears to have a lower morbidity than that associated with arterial reconstruction. Steps to minimize cerebrovascular injury are discussed.
Cervical metastasis involving the carotid artery is particularly challenging. The treatment modalities include supportive care alone, chemotherapy, external radiation therapy, interstitial radiotherapy (brachytherapy) and surgery. Surgical options include palliative or curative peel, carotid resection without revascularization and bypass grafting. If surgery is performed, salvage of the carotid artery is advised whenever the tumour can be easily separated from the artery. When the tumour is adherent to the artery, a carotid artery peel technique can be used.
The carotid sheath and the arterial wall usually act as a resistant barrier to invasion by squamous cell carcinoma, but advanced lesions of the neck may present with direct invasion or encasement of the artery, especially after primary treatment with radiotherapy. Surgical excision of neck metastases, including the carotid artery, gives the best chance of local-regional control. Historically, the survival of patients with recurrent or residual extranodal disease involving the deep cervical tissue, carotid artery, or skin is poor. Such fixed neck masses carry a poor prognosis due to the presence of extracapsular spread and the invasion into vital regional structures. It can be argued that these patients should be offered surgical resection provided the risk of complications or death is considered acceptable by the patient.
In this retrospective study the authors reviewed the surgical treatment of 73 patients treated in two units: The Chinese Academy of Medical Sciences in Beijing, China, and Poole General Hospital, Dorset, UK, in the 5 year period between January 1990 and 2005. The findings were compared with current practice reported in the literature.
Patients and methods
Seventy-three patients with squamous cell carcinoma of the head and neck and advanced cervical metastases involving the carotid artery were included in the study. Carotid artery involvement was demonstrated by computed tomography/magnetic resonance imaging preoperatively and this was subsequently confirmed in all patients intra-operatively. All cases of carotid invasion were due to involvement by cervical metastases as opposed to direct extension from the primary tumour. All 73 patients were treated either with radical neck dissection combined with primary tumour resection or with radical neck dissection alone. The extent of the tumour and its exact relationship to the carotid artery and surrounding tissues was assessed intra-operatively.
All the patients were followed up by mail, telephone and at regular review whenever possible for 60–120 months.
The relation between the tumour and the carotid artery was classified as follows: (1) tumour encircling <50% of the circumference of the arterial wall; (2) tumour encircling 50–100% of the circumference of the arterial wall; (3) tumour encircling 100% of the circumference of the arterial wall. Further classification was included if the tumour involved the adventitia or adventitia and the media after assessment of frozen sections.
The tumour and adventitia were excised from the carotid artery using a scalpel. Following the ‘ en bloc oncological resection principle’, the tumour was removed with a clear margin and was carefully peeled off the carotid artery. Metastatic disease encasing the carotid was successfully peeled off ( Figs. 1 and 2 ). The success of the resection was termed ‘gross negative’ when the tumour was completely removed from the adjacent tissue and carotid artery adventitia. When the tumour was not completely removed, the resection was termed ‘gross residue’. Frozen sections were sent to determine whether the tumour was invading the adventitia.
The carotid artery was not resected if there was deemed to be potential space between the adventitia and the tumour or if the tumour peeled off smoothly. The carotid artery was resected if the tumour had invaded the media of the arterial wall or if the tumour could not be peeled off. Carotid resection was only performed if the patients were otherwise healthy and capable of tolerating the postoperative neurological complications ( Figs. 3 and 4 ).
Back pressure of the ipsilateral internal carotid artery (ICA) was measured in this patient group at the time of operation. If >70 mmHg the ICA was resected and ligated. A vascularized local muscle flap from the nearest muscle group was wrapped around the arterial wall after resection. If the back pressure was <50 mmHg the carotid artery was not resected and the tumour was removed palliatively. A metal marker was placed in the tumour bed as an indicator for postoperative radiotherapy. Postoperative radiotherapy was given to all the patients with gross tumour residue and those who were never previously irradiated in that area.
The clinic data were analyzed using the SPSS11.5 software package. The extent of the tumour invasion in the carotid artery was calculated. Carotid artery resection was recorded and the presence of gross tumour residue was documented to determine whether there was an impact on tumour recurrence and the 2 year survival rate. The survival rate was calculated using the Kaplan–Meier method.
Of the 73 patients there were 62 male and 11 female patients. Ages ranged from 32 to 75 years (median 58.4 years). Forty-three had laryngeal cancer, 14 had hypopharyngeal cancer, five had nasopharyngeal cancer, four had cervical metastases with unknown primary sites and seven had oropharyngeal cancer. Sixty-five patients had cervical recurrence after previous treatment (surgery alone, surgery combined with radical radiation or radiation). Eight patients were treated with primary surgery, five of these had nasopharygeal carcinoma with cervical metastasis after radical radiotherapy and three had cervical metastasis with an unknown primary site.
The tumour was peeled off the carotid artery in 57 of 73 patients in this series and resected in the remaining 16 patients. 14 of 16 from this ‘resection group’ were free of any postoperative neurological complication. All these had demonstrated an ipsilateral ICA back pressure >60–70 mmHg. Of the remaining two patients, one developed an ipsilateral haemiplegia and the other a contralateral haemiplegia immediately after surgery from carotid artery rupture. This patient died from multisystem failure on the fourth postoperative day. The ipsilateral ICA back pressure was only 50 mmHg in this patient.
8 of 73 patients survived long term. The patients were followed up 48, 56, 66, 74, 96, and 120 months postoperatively. Sixty-five patients succumbed to their disease. The median survival time was 12 months. The 1, 2, 3, and 5 year survival rates were 50.7%, 20.9%, 13.4% and 11.5%, respectively. The causes of death are summarized in Table 1 .
|Causes of death||Numbers||Percentage|
|Carotid artery rupture||10||7.3%|
|Unrelated to tumour||2||1.46%|
Most patients die of local or regional recurrences. In the patients who developed distant metastases, four cases affected bone, three the liver and the rest lung. The extent of tumour involvement in the carotid artery significantly affected the recurrence rate ( Table 2 ). There were no significant differences in the recurrence rates and length of survival between patients in whom the tumour was peeled off the artery and those in whom the carotid artery was resected and ligated ( Tables 2 and 3 ). 4 of the 10 carotid arteries ruptured as the result of tumour recurrences.