DISCUSSION
First described in 1962 by Borsany(14), CBS still is an important complication in head and neck cancer, with a mortality of 40% and neurologic morbidity of 60%(1)(15). Prior radiotherapy, prior radical neck dissection, mucocutaneous fistula, flap necrosis, wound infection, poor nutrition and tumor recurrence can be counted as risk factors for CBS(5)(7)(8). Tumor encasement and infiltration of carotid artery causing inflammation and weakening of carotid artery wall, also plays an important role(5)(9). Overall incidence of CBS in head and neck cancer ranges from 2.5% to 4.5%(2)(5)(10). After examination of medical records of 348 patients treated for head and neck cancer in our clinic, 8 patients with CBS treated by open approach carotid artery ligation, were determined, having a slightly lower rate compared to the literature(2.3%).
CBS was classified as type I(threatened) CBS occurring when carotid artery was exposed through soft tissue breakdown, type II(impending) with limited, sentinel bleeding and type III(rupture) with active bleeding(1)(12). Type III is the most common type of CBS(over 60%) followed by type II and I, respectively(12)(16). Chaloupka et al., reported their findings on endovascular treatment of CBS. Among 4 patients with type I, 10 patients with type II and 5 patients with type III CBS; most common pathologic finding was pseudoaneurysms, followed by tumor encasement(1). However, in another study, extravasation was the most common pathologic finding(in 8 out of 10 patients), while pseudoaneurysm was seen in only 2 patients, which could be explained as all these 8 patients were type III CBS(17). All of our patients presented with acute rupture(type III CBS), thus most common finding was extravasation, followed by pseudoaneurysm.
CCA emerged as the most common site of origin of rupture in our study, which was consistent with the literature reporting a rate of over 60% (5)(12)(17)(18). Bleeding was found to be from CCA near bifurcation in 5(62.5%), ICA in 2(25%) and from main trunk of external carotid artery just above bifurcation in one(12.5%) patient. However, in a study consisting of 6 patients with CBS, ICA was found to be the most common site(in 4 out of 6 patients), followed by CCA(2). In addition, in a larger scale study, again ICA was reported to be the most common site with a rate of greater than 50%, while CCA was being the least common(1).
Previous radiotherapy is considered to be the most important risk factor, by increasing the risk up to seven fold(6)(19), even to 14-fold in case of a radiation dose of >70 Gy(8). After exclusion of patients who were previously irradiated, CBS rate was found to be only 0%–2.4%; however, when patients received radiotherapy analyzed separately, CBS rate increases to 4.5%–21.1% (5). Free radical production triggered by radiotherapy is considered to be causative agent by inducing thrombosis of vaso vasorum, fibrosis of adventitia, and premature atherosclerosis; eventually resulting in weakening of arterial wall(19). In Estomba et al.’s study, except for one out of six patients with CBS had a previous history of radiotherapy and cause of CBS was proved to be radiation induced necrosis in 3 patients, pharyngocutaneous fistula in 2 patients, and tumor recurrence in 1 patient(2). Durmaz et al., stated that history of radiotherapy was the common future in all of their 10 patients with CBS, 7 with recurred cancer and 5 with pharyngocutaneous fistula(17). In a large review by Powitzky, among 140 patients and 161 CBS cases, previous history of radiotherapy was found to be the most common shared feature(89%), followed by nodal metastasis(69%), neck dissection(63%), necrosis(55%), mucocutaneous fistula (40%)(12). Being another risk factor, neck dissection - especially radical neck dissection, with stripping of the carotid – was reported to increase the risk of CBS to eight fold(8). Among 8 patients included in this study, all had a previous history of radiotherapy, with 3 as primary curative treatment and 5 as adjuvant treatment. 3 of the latter also had a history of radical neck dissection. Pharyngocutaneos fistula in 3, wound infection in 5 and necrosis at bleeding site in 6 of the patients were seen, apart from tumor infiltration which was present in all.
Most common tumor primary sites were larynx(23%), hypopharynx(22%) and nasopharynx (21%)(12). In a study consisting of 10 patients with CBS, in a six-year period, 3 had had primary tumor in larynx, 2 each in oral cavity, hypopharynx, nasopharynx, and 1 in salivary gland(17). In our study, primary site was oral cavity in 3(37.5%), oropharynx-hypopharynx in 3(37.5%) and larynx in 2(25%) of the patients.
Surgical ligation has been traditionally the treatment of choice for CBS(18)(20). However, after the introduction of the detachable balloons specifically for treatment of impending CBS in 1984 by Osguthorpe and Hungerford(21), there has been an increasing tendency towards endovascular approach. Over 90% of patients reported in the past fifteen years were treated by endovascular approach, either by stenting or embolization; while surgical ligation was the preferred method in less than 10%(12). Drawbacks of surgical ligation; such as hemodynamic instability complicating perioperative management, hypotension deepened with general anesthesia resulting in cerebral ischemia, consumption coagulopathy due to blood loss resulting in uncontrollable re-hemorrhage, challenging surgical dissection caused by previous operations and radiotherapy; played role in this change of treatment preferences(1). In a paper reporting 6 cases with CBS, among 3 patients managed by endovascular approach, 1 had neurologic sequel; while 1 patient managed by surgical ligation also suffered neurologic complication(2).
Regardless of the treatment choice; angiography is accepted as gold standard for diagnosis(22), by also having the advantage of evaluating the patency of contralateral carotid system with balloon occlusion test, thus leading the treatment option(23)(24). Hence, every treatment should be preceded by angiographic evaluation; yet it may not be possible in every case due to need of emergent intervention.
Average time to CBS from initial time of diagnosis was 2.7 years, mostly originating from proximal part of the carotid bifurcation(12). Among patients who had a second regime of radiotherapy, median time for CBS to occur, from beginning of re-irradiation, was 7.5 months, ranging up to four to five years, in certain cases(10). In our patient population, time to recurrence following primary treatment was between 6-19 months with a mean of 8.25±5.11 months. Time from detection of recurrence or soft tissue exposure to CBS, was from 2 to 12 months with a mean of 5.37±3.11 months, with being earlier in patients having fistula and necrosis (6 patients; 2-4 months with a mean of 4±1.2 months). In this regard, following massive neck or local recurrence with soft tissue exposure, necrosis and fistula, carotid blow-out is to be expected within 4 months and preventive measures must be undertaken during this period.