Introduction

Majority of the patients with head and neck cancer present at an advanced stage. Many patients may not be amenable for definitive treatment due to the advanced presentation and others may develop locoregional recurrences and/or metastasis despite definitive multimodality treatment [1]. Few patients with locoregional recurrence can be candidates for surgery and/or re-irradiation and others would only qualify to receive palliative intent treatment. Palliative intent treatment options included mostly supportive care and chemotherapeutic agents (most often cisplatin) either alone or in combinations with other chemotherapeutic agents including targeted therapy [1]. These palliative patients can have sudden and massive hemorrhages due to damage to blood vessels in the area of the ulcerated tumor bed which can present as a life-threatening complication [2]. Patients who are at high risk of bleeding include those who have tumor-infiltrating into blood vessels, thrombocytopenia due to the myelosuppressive effect of chemotherapy, systemic complications of cancer, drugs like anticoagulants and NSAIDS and concurrent infection [3, 4]. The source of bleeding can be from large-caliber vessels like the internal carotid artery (ICA) or common carotid artery (CCA) or it can be from the branches of external carotid arteries (ECA). Surgical intervention or access to these damaged vessels may be limited or impossible in such cases. Diagnostic imaging, i.e. angiography helps in localization of the injured blood vessel with simultaneous embolization [2].

The Yorkshire Palliative group proposed a general guideline for the management of bleeding in palliative patients, which has embolization as one of the options to control bleeding. Embolization is a technique that was first introduced by Duggan in 1970 for controlling post-traumatic epistaxis. Nowadays, it is used for a variety of indications including in palliative cancer patients to control bleeding from tumors [5]. This procedure involves obliterating the blood vessel supplying the tumor with an embolic agent whose selection depends on the character of embolization, predicted artery closing time, location, and vessel diameter [6]. Palliative embolization is not an often-used modality in palliative patients with sudden bleeding from the tumor. It necessitates the availability of the infrastructure, expertise, and availability round the clock. It also involves costs that may not be affordable by all. Hence we conducted this clinical audit of patients who underwent palliative (selective) embolization at our institute with the primary objective to assess the benefits of the procedure in terms of the bleeding-free interval achieved after selective embolization. The secondary objective was to assess the common clinical presentations and the morbidity following the procedure, if any.

Methodology

In this retrospective study, we included patients with unresectable, unsalvageable recurrent and/or metastatic head and neck cancer who presented with sudden and massive haemorrhage to the emergency department (causality) for which palliative (selective) embolization was performed, from Jan 2015 to Nov 2019. Patients whose bleeding could be effectively controlled with conservative measures and/or surgical intervention were excluded. The study was conducted in accordance with the Declaration of Helsinki as revised in 2013. These patients were managed as per our institute protocol (Fig. 1). Surgical options were not considered by the treating team as it was not considered feasible to have control over the major vessels due to the anatomical location of the tumor (primary and/or neck node). Patients with a reasonable performance status of ≤ 3 were considered for embolization after thorough clinical evaluation. All embolization procedures were done under local anaesthesia once airway access (tracheostomy) was achieved. The femoral artery access was taken in all the cases using 4Fr vascular sheath. Based on the location of the tumor, ipsilateral digital subtracted angiography (DSA) of the carotid vessels was performed to identify the site of the bleeding. If no bleeder was found, contralateral carotid angiography was performed. In the presence of tumor hyperemia with no pseudoaneurysm (PSA), the tumor was embolised using particles [polyvinyl alcohol (PVA) or gel foam]. In case of malignancy involving the tongue, with the risk of non-target embolization of normal tongue parenchyma, gel foam was used as an embolising agent which would enable revascularisation of the normal tongue parenchyma in few days but enable immediate cessation of tumor bleed. When the tumor was located elsewhere in the head and neck region, wherein the risk of non-target embolization was minimal, PVA which is a permanent embolising agent was used. If a pseudoaneurysm (PSA) was identified due to blow out of one of the branches of the external carotid artery and selective cannulation of the PSA was possible, coils were preferred to block the bleeder. However, if selective cannulation of the PSA was technically challenging, liquid embolising agent glue (N-butyl cyanoacrylate) was used for embolization. In case of external carotid artery blowout, the artery was sacrificed by coil embolization and in case of a blowout involving the carotid bulb requiring blocking of internal carotid artery also, a balloon test occlusion (BTO) was done to document adequate cross-circulation across Circle of Willis before coil embolization of ICA. [5] Bleeding following the procedure was documented as an event and the bleeding free interval was calculated from the date of embolization to the last date of follow up available without another bleeding episode or the date of another episode of bleeding. Statistical analysis was done with SPSS version 22, only descriptive analysis was done.

Fig. 1
figure 1

Protocol for managing patients presenting to our causality/emergency department with massive bleeding

Full size image

Results

Out of 124 embolization procedures done for various head and neck neoplasms during the study period, 26 patients with unresectable, unsalvageable recurrent and/or metastatic head and neck cancer underwent selective (palliative) embolization for sudden and massive haemorrhage and were included in the study.

Patients clinical details (Table 1)

Table 1 Demographic and clinical details of patients who underwent palliative embolization in our series
Full size table

The majority of them were men (n = 24, 88.8%) with a median age of 54.5 years. The oral cavity was the most common subsite in our cohort followed by oropharynx. Thirteen patients had unresectable recurrences (50%), followed by unresectable primary at presentation (n = 11). All patients had their primary in the upper aerodigestive tract (UADT), except for two patients with thyroid carcinoma (with lung metastasis). Squamous cell carcinoma was the most common histopathology in the UADT tumors. The disease-free interval in patients with unresectable recurrences was anywhere between 2 months and 5 years. All the patients presented to the casualty/emergency department with oral or nasal bleeding. The patients were evaluated as per our institute protocol for such patients (Fig. 1) and it was decided to perform embolization by the treating team for them after having airway control and stabilizing the patients hemodynamically. All patients had ECOG performance status (PS) of 2 (n = 15, 57.6%) or 3 (n = 11, 52.4%).

Of the 26 cases who underwent embolization to control bleeding, 6 (22.22%) cases were first referred for hemostatic radiotherapy (RT) for the minor ooze or bleed before the major bleeding episode. Out of the six cases, only one case (16%) received hemostatic RT. The others did not receive due to reasons such as previous treatment with RT and/or short interval from the last date of RT. One patient (3.7%) underwent ligation of ECA under GA for control of bleeding, but presented with recurrent episodes of bleed and was subjected to angioembolization.

Selective (palliative) embolization details (Table 1)

The source of bleeding was commonly one or a combination of the branches of the external carotid artery (ECA). In one patient, there was complete occlusion of the internal carotid artery (ICA) with a blow out of the common carotid artery (CCA) in the part of the tumor. Coiling was done on the ECA and the CCA distal and proximal to the blow out. The details regarding the various embolising agents used are given in Table 1. The patients tolerated the procedure well without the development of any neurological deficits. This patient had a bleeding free interval of 1 month and 2 weeks before he had another episode of bleeding and succumbed to it. Two patients of thyroid carcinoma presented with severe hemoptysis from their metastatic lung lesion. The bronchial artery was identified as the bleeder and embolized. These patients were subsequently on tyrosine kinase inhibitors and had good bleeding free intervals. Polyvinyl alcohol and gel foam the two commonly used embolizing agents either alone or in combination. All agents used for embolization were found to be effective in controlling bleeding.

Complications(morbidity) and bleeding-free interval (Table 1)

The most common complications of the procedure includes prolonged bleeding at the puncture site, soft tissue hematomas, intima dissection of the involved artery, allergic reaction to the contrast medium. The most serious complication involves unintentional closure of a different vessel as a result of technical error or reflux of embolization material [7]. All 26 patients in our series tolerated the procedure well without developing any complications following the procedure. The patients had a bleeding free interval ranging from 2 weeks to 21 months. The cost of the embolization at our institute was between 400 and 2100 US Dollars. Rebleed occurred in 3 out of the 26 patients, 1 case of tonsillar, laryngeal, and nasopharyngeal malignancy each. All 26 patients received best supportive care, some of them received palliative chemotherapy or oral metronomic chemotherapy also, after undergoing embolization.

Discussion

The present study shows that palliative embolization can be effectively done in select patients with effective outcomes at a reasonable cost. Angioembolization is effectively used, especially after the introduction of new embolization agents, for selective embolization of vessels. One of the highlights of angioembolization is the possibility of simultaneous angiography which localizes the pathology and selective obliteration of the damaged vessel [8]. It is highly useful when the bleeding is extensive, when there is necrotic tumor mass or when there is a lack of consent for a surgical procedure like ligation of the vessels [9]. Some authors believe that the direct application of embolic agents (onyx or tissue glue) into the tumor gives better results than the transarterial method [10].

History of embolization in the head and neck area goes as far back to the early twentieth century. Brooks was the first to describe a case of carotid-cavernous fistula treated by injecting a muscle fragment connected with a silver clip into the internal carotid artery [11]. This was followed by the use of various embolization agents for pathological conditions like aneurysms, haemangiomas, and other vascular malformations of the head and neck area. The extent and effectiveness of the procedure depend on many factors like vascular network, blood flow rate, type of pathology (tumor, hemorrhage), and type of embolic agent. Over the years, the different agents which have been developed include tissue glue (isobutyl 2-cyanoacrylate), spongostan, coil, and polyvinyl alcohol (PVA). PVA was introduced in 1975 and remains the most widely used embolization agent [12]. Embolic agents can be of 2 types—permanent and temporary. Permanent embolic agents include coils, latex balloons, polyvinyl alcohol, tissue glue, and ethylene–vinyl alcohol; temporary agents include fibrin sponge or spongostan [5]. Embolization can also be classified as preoperative, therapeutic, and palliative. A pre-procedure assessment of the vascular structures is important. This includes a precise assessment of topography, the angle of artery branching as well as vessel diameter allows for proper selection of a catheter and embolic agent. A thorough knowledge of vascular anatomy of the head and neck region is pertinent as complex variations can often be encountered in these patients.

Whenever surgical ligation is not feasible/possible or if the patient does not give consent for surgical ligation, angioembolization can be considered as an option whenever the facility for the same is available, especially within the institute. Rzewnicki, et al. did a retrospective study of 76 patients with advanced head and neck tumors who were subjected to palliative embolization. The therapeutic efficacy for the procedure was found to be 86% [5]. In the available literature, the therapeutic efficacy ranges from 80 to 95% [13]. Our study recorded a 100% therapeutic efficacy. All our patients were either cases who were previously treated with surgery, chemotherapy, chemoradiotherapy, or radiation therapy. These were cases of advanced cancer at presentation or recurrent and/or inoperable cancers and/or with distant metastasis. Previous radiation therapy, recurrent carcinoma, and surgical complications such as infection and fistula formation may all contribute to the onset of such bleeding and may make controlling it more difficult [14].

None of the patients in our study had any major complications, which is comparable to other literature [5]. There are reports in the literature regarding certain rare and dangerous complications like ischemic stroke, cranial nerve paresis, blindness, or sometimes even death [15]. With due precautions and considerable operator experience such complications of embolization are minimized (0.03% mortality and 1.73% morbidity) [16].

The limitation of our study is its retrospective nature and small numbers of patients. However, there is limited literature available regarding palliative (selective) embolization for patients with unresectable, unsalvageable recurrent and/or metastatic head and neck cancer presenting with sudden and massive haemorrhage. The present series will help fill this void to a limited extent and direct future research. Future research should focus on the quality of life of these patients following embolization along with the cost aspects (health economic) associated with the procedure. The protocol followed at our institute to manage these patients is replicable. However, reasonable caution needs to be exercised in choosing the patients for embolization after considering patients performance status, ruling out the possibility of surgical control, availability of expertise and infrastructure for embolization.

Conclusion

Palliative embolization is a minimally invasive technique that can be used effectively, but with caution in select patients with reasonable PS (≤ 3), to control bleeding in advanced head and neck cancer patients in whom the feasibility of surgical control is less likely. Palliative embolization can provide reasonable bleeding-free intervals at an affordable cost. The technique could be preferred especially if the facility and the expertise are available within the institute. The complication rate of this procedure can be minimized by proper case selection, pre-procedure assessment of the anatomy of the vessels, and the use of suitable embolization agents.