a b s t r a c t

Background: Acute head and neck cancer (HNC) bleeding is a life-threatening situation that frequently presents to the emergency department (ED). The purpose of the present study was to analyze the risk factors for the 30-day mortality in patients with HNC bleeding.

Methods: We included patients who presented to the ED with HNC bleeding (n = 241). Patients were divided into the survivor and nonsurvivor groups. Variables were compared, and the associated factors were examined with Cox’s proportional hazard model.

Results: Of the 241 patients enrolled, the most common bleeding site was the oral cavity (n = 101, 41.9%). More than half of the patients had advanced HNC stage while 41.5% had local recurrence. The proportion of active bleeding was significantly higher in the nonsurvivor group (70.5% vs. 53.3%, p = 0.038). 42.3% received blood transfusion and 5.0% required inotropic support. In total, 21.2% of the patients experienced rebleeding, and 18.3% died within 30 days. Multivariate analyses indicated that a heart rate > 100 (beats/min) (HR = 2.42; Cl 1.15-5.06; p = 0.019) and inotropic support (HR = 3.00; Cl 1.14-7.89; p = 0.026) were statistically significant independent risk factors for 30-day mortality.

Conclusions: The results of this study may aid physicians in the evaluation of Short-term survival in HNC bleeding patients and provide critical information for risk stratification and medical decisions.

(C) 2022

1. Introduction

Head and neck cancer (HNC) encompasses a heterogeneous group of malignancies involving the oral cavity, pharynx, and larynx. In Taiwan, HNC is a highly prevalent malignancy with an incidence rate of 33.1 per 10,000 persons and a mortality rate of 12.5 per 10,000 patients among the general population in 2016 [1]. The high incidence rate of HNC in Taiwan is linked to the common habit of betel nut chewing ac- companied with smoking and alcohol drinking [2,3]. Nearly half of the HNC patients were diagnosed with advanced cancer stages (i.e., clinical cancer stages III to IVB) with 5-year survival rates of 54.7% in locally advanced HNC patients and 32.2% in HNC patients with distant organ or lymph node metastasis [1].

* Corresponding author.

E-mail address: [email protected] (H. Yeh).

¹ Chieh-Ching Yen and Heng Yeh contributed equally to this work.

Standard therapies of HNC consist of surgery, chemotherapy, radio- therapy, and concomitant therapy (e.g., any combination of surgery, che- motherapy, and radiotherapy) all having a chance of causing acute bleeding from the head and neck area due to necrosis of arterial walls [4-10]. Acute HNC bleeding is a Life-threatening complication with a var- ied incidence ranging from 0.5 to 10% [11]. Patients with massive bleed- ing may suffer from hemorrhagic shock and airway obstruction due to Blood aspiration. HNC bleeding is usually managed with conservative treatments such as local compression, dressing, packing, hemostatic agents (e.g., Tranexamic acid and epinephrine-soaked gauze), and blood transfusion, as appropriate [12,13]. endovascular treatment is primarily used for Bleeding control when conservative treatment fails [5,14,15]. Al- though major improvements have been made in managing patients with HNC bleeding, a poor overall prognosis still follows because of the ad- vanced cancer stage [16]. Identifying patients at risk of poor short-term survival is of crucial importance because it allows for the advance

https://doi.org/10.1016/j.ajem.2022.05.008

0735-6757/(C) 2022

planning of the most appropriate course of management when massive HNC bleeding occurs. However, there is little research investigating Short-term outcomes of patients with HNC bleeding. The purpose of this study was to analyze the risk factors for 30-day mortality in patients presenting to the emergency department (ED) with HNC bleeding.

1. Materials and methods
   1. Study design and setting

This retrospective, multiple-center observational study was con- ducted with the data from the emergency departments at five hospitals sharing the same electronic medical record (EMR) system in Taiwan in- cluding two Tertiary medical centers and three Regional hospitals. The study sites’ total capacity was over 9000 beds and an annual ED volume of 500,000 patients. The current study was approved by the Chang Gung Medical Foundation Institutional Review Board (IRB no. 202102021B0) and is in accordance with the Declaration of Helsinki. All adult patients who presented to the ED from January 1, 2015 to December 31, 2016 and met the inclusion criteria of the study were enrolled. The study’s endpoint was taken at 30 days post-ED presentation.

• 1. Patient selection and data collection

All adult patients with head and neck cancer (HNC) who were treated in the ED were identified with the EMRs by International Classi- fication of Diseases (ICD)-10 codes “C00-C14″ and” C30-C32″. Next, pa- tients whose record included the keywords of ‘bleeding’, ‘hemorrhage’, or ‘carotid blowout’ were selected. Patients were excluded if they had incomplete medical records, recurrent ED visits due to HNC bleeding, nonSquamous cell carcinoma, or bleeding events originated from other sites. The record of the patients selected from the EMRs were fur- ther reviewed by two physicians for the verification of their inclusion el- igibility (H.Y. and Y.-C.C.).

The patients’ baseline characteristics of sex, age, lifestyle factors (e.g., betel nuts chewer, smoking history, and alcohol use), initial ED vital signs, and comorbidities (e.g., hypertension, diabetes mellitus, cor- onary artery disease, chronic kidney disease, other malignancy, prior stroke, and liver cirrhosis) were retrieved. The collected laboratory data at the initial presentation included white blood cell count, hemo- globin, platelet count, prothrombin time, creatinine, and alanine amino- transferase. Clinicopathological parameters at the initial presentation of primary cancer were obtained. This data included cancer site, tumor- node-metastasis (TNM) stage (based on the TNM staging system by the American Joint Committee on Cancer, 7th edition), cancer Treatment modality, and local recurrence.

Computed tomography (CT) angiography was performed in patients with active life-threatening HNC bleeding when conservative treat- ments failed. The causes of HNC bleeding consisted of tumor-related causes, pseudoaneurysm, fistula formation, and postoperative compli- cations. Tumor-related causes were identified by contrast extravasation, hypervascular tumor staining, great vessel involvement on CT angiogra- phy, or bleeding from the necrotic wound of the tumor confirmed by otolaryngologists with or without fiberoptic endoscopy. Pseudoaneu- rysm was confirmed by CT angiography.

Patients with HNC bleeding received one of the following treat- ments: supportive care, Endovascular therapy, and surgical interven- tion. Supportive care was defined as oral or intravenous tranexamic acid, epinephrine-soaked gauze compression and packing, or observa- tion. Endovascular therapy consisted of Transarterial embolization or covered stent graft placement. Surgical intervention consisted of surgi- cal ligation or Primary repair. Patients who received blood transfusion, which was defined as at least 2 units of Packed red blood cells with or without other blood products, were recorded. Patients who required inotropic support were also noted. The primary outcome was 30-day mortality after the initial bleeding presentation to the ED.

• 1. Statistical analysis

Patient characteristics, previous medical history, laboratory findings, and presentations of cancer and bleeding were presented as numbers and percentages for the categorical variables, while the continuous data were presented as mean +- standard deviation (SD). Comparisons between survivors and nonsurvivors were examined with Chi-square test or Fisher’s exact test, as appropriate, for the categorical variables and with independent Student’s t-tests or Mann-Whitney U tests for the normally distributed and the skewed continuous variables, respec- tively. Kaplan-Meier analyses were performed to assess the cumulative overall survival rate. To identify the independent risk factors for 30-day mortality, we used a stepwise approach by selecting the variables that were significant in the univariate analysis for the multivariate Cox pro- portional hazards model. All analyses were performed using SPSS soft- ware v26 (SPSS Inc., Chicago, IL). A two-sided p value of <0.05 was considered statistically significant.

1. Results
   1. Patient characteristics of survivors and nonsurvivors

A total of 241 patients satisfied the inclusion criteria of this study. Among the patients, 139 (57.7%) were discharged from the ED, 95 (39.4%) were hospitalized, and 7 (2.9%) died in the ED. 36 (14.9%) did not survive during the hospital admission. The mean age was 56.9+-

10.8 years, with 223 patients being male (92.5%) (Table 1). Age and sex distribution did not differ across the survivors and the nonsurvivors. At the initial ED presentation, the heart rate (beats/min) was signifi- cantly higher in the nonsurvivor group (115.6 vs. 99.5, p < 0.001). The most common primary index cancer site was the oral cavity (n = 101, 41.9%), followed by hypopharynx (n = 48, 19.9%), oropharynx (n = 42, 17.4%), nasopharynx (n = 34, 14.1%), and larynx (n = 16, 6.6%). More than half of the patients had advanced HNC stage while 100 (41.5%) had local recurrence. As for the cancer treatment, 124 (51.5%)

Table 1

Characteristics of patients with head and neck cancer bleeding according to survival status. Variable Survivor Nonsurvivor P value

	N = 197		N = 44
Age (year)	57.3+-10.2		55.1+-13.1	0.229
Male	181(91.9)		42(95.5)	0.540
SBP (mmHg)	141.8+-34.1		130.4+-38.6	0.055
DBP (mmHg)	85.4+-17.8		82.5+-19.0	0.353
SBP < 90 (mmHg)	10(5.1)		5(11.4)	0.156
Heart rate (beats/min)	99.5+-21.2		115.6+-22.6	<0.001?
Heart rate > 100 (beats/min)	92(46.7)		31(72.1)	0.003?
Smoking history	157(79.7)		37(84.1)	0.506
Betel nut chewer	135(68.5)		28(63.6)	0.531
Alcohol use	95(48.2)		25(56.8)	0.303
Hypertension	74(37.6)		13(29.5)	0.317
Diabetes mellitus	44(22.3)		9(20.5)	0.785
Coronary artery disease	18(9.1)		2(4.5)	0.544
Chronic kidney disease	17(8.6)		3(6.8)	1.000
Other malignancy	17(8.6)		3(6.8)	1.000
Prior stroke	17(8.6)		2(4.5)	0.540
Liver cirrhosis	21(10.7)		4(9.1)	1.000
Antiplatelet agenta	30(15.2)		5(11.4)	0.511
Laboratory exam
WBC (103/ul)	11.0+-8.1	13.5+-8.1		0.059
Hb (g/dl)	10.9+-2.3	9.9+-2.0		0.003?
PLT (103/ul)	254+-99	300+-149		0.058
PT (sec)	12.1+-1.49	13.5+-3.45		0.016?
Creatinine (mg/dL)	1.09+-1.21	1.19+-1.40		0.631
ALT (U/L)	38.1+-115.0	53.9+-99.8		0.442

Count data are expressed as number (percentage) and continuous values are expressed as mean +- SD.

SBP: systolic blood pressure; DBP: diastolic blood pressure; WBC: white blood cell; Hb: hemoglobin; PLT: platelet count; PT: prothrombin time; ALT: alanine aminotransferase.

^a Antiplatelet agents are defined as aspirin or clopidogrel.

patients underwent surgical resection, 106 (44.0%) underwent concom- itant neck dissection, and 87 (36.1%) underwent concurrent flap recon- struction. Additionally, 185 (76.8%) patients received adjuvant concurrent chemoradiotherapy and 22 (9.1%) had no cancer-related treatment due to either patient choice or newly diagnosed cancer. The characteristics of patients are summarized in Table 1.

• 1. Bleeding and various treatment modalities

Of the patients with HNC bleeding, 136 (56.4%) were found to have active bleeding whereas 105 (43.6%) reported the spontaneous cessa- tion of bleeding after the presentation to the ED. The proportion of ac- tive bleeding was significantly higher in the nonsurvivor group (70.5% vs. 53.3%, p = 0.038). The causes of HNC bleeding were as follows: tumor-related (n = 192, 79.7%), pseudoaneurysm (n = 25, 10.4%),

Table 2

Features of cancer in patients with head and neck cancer according to survival status.

Variable	Survivor		Nonsurvivor	P value
	N = 197		N = 44
Cancer site				0.374
Oral cavity	85(43.1)	16(36.4)
Nasopharynx	25(12.7)	9(20.5)
Oropharynx	35(17.8)	7(15.9)
Hypopharynx	37(18.8)	11(25.0)
Larynx	15(7.6)	1(2.3)

T stage 0.297

T1 26(13.2) 2(4.5)

T2 32(16.2) 9(20.5)

T3 26(13.2) 3(6.8)

T4 108(54.8) 28(63.6)

Unknown 5(2.5) 2(4.5)

N stage 0.332

postoperative complications (n = 23, 9.5%), and fistula formation	N0	75(38.1)	12(27.3)
(n = 1, 0.4%). CT angiography was immediately preformed in 79	N+	115(58.4)	31(70.5)

0.308

(32.8%) patients to determine the location of the bleeding, and signifi- Unknown		7(3.6)	1(2.3)
cantly more patients in the nonsurvivor group received CT angiography M stage M0 177(89.8) 36(81.8)
(50% vs. 28.9%, p = 0.007). Of these patients, 8 (10.1%) showed contrast	M1	14(7.1)	6(13.6)
extravasation on imaging (all from the external carotid artery), 19	Unknown	6(3.0)	2(4.5)

(24.1%) had a pseudoaneurysm (12 from the external carotid artery, 5 from the Internal carotid artery, and 2 from the Common carotid artery), 6 (7.6%) had a pseudoaneurysm combined with contrast extravasation (5 from the external carotid artery and 1 from the internal carotid ar- tery), and 46 (58.2%) had no identified vessel involvement. Various treatments were performed for the management of HNC bleeding, in- cluding only supportive care (n = 200, 82.3%), transarterial emboliza-

0.015

Initial cancer treatment Surgical resection	109(55.3)	15(34.9)
Chemoradiation	146(74.1)	39(88.6)	0.039
Neck dissection	92(47.2)	14(32.6)	0.081
Flap reconstruction	74(37.9)	13(30.2)	0.342
Local recurrence Bleeding cause	79(41.4)	21(47.7)	0.441 0.064
Tumor-related	151(76.6)	41(93.2)
Pseudoaneurysm	22(11.2)	3(6.8)

*

tropic support. The nonsurvivor group showed a significantly higher rate of inotropic support (13.6% vs. 3.0%, p = 0.010) and a higher, yet non-significant, rate of blood transfusion (54.5% vs. 39.6%, p = 0.07) (Table 2).

• 1. Univariate and multivariate cox regression analyses of 30-day mortal- ity in patients with HNC bleeding

In total, 51 (21.2%) patients experienced rebleeding and 44 (18.3%) died within 30 days (Fig. 2). The causes of death were as follows: 27 (61.4%) tumor rebleeding, 11 (25%) Aspiration pneumonia, 3 (6.8%) sep- sis, 1 (2.3%) cancer cachexia (cancer patients who suffer from skeletal

0.038?

Self-limited	92(46.7)	13(29.5)
Active bleeding	105(53.3)	31(70.5)

Emergent CTA 57(28.9) 22(50.0) 0.007?

tion (n = 32, 13.3%), covered stent placement (n = 5, 2.1%), surgical	Postoperative complication	23(11.7)	0(0)
ligation (n = 3, 1.2%), and primary repair (n = 1, 0.4%) (Fig. 1). In the ED, 102 (42.3%) received blood transfusion, and 12 (5.0%) required ino-	Fistula formation Bleeding type	1(0.5)	0(0)

Bleeding treatment 0.559

Supportive care 164(83.2) 36(81.8)

Embolization 24(12.2) 8(18.2)

Covered stent	5(2.5)	0(0)
Surgical ligation	3(1.5)	0(0)
Primary repair	1(0.5)	0(0)
Blood transfusion	78(39.6)	24(54.5)	0.070
Inotropic support	6(3.0)	6(13.6)	0.010?

Count data are expressed as number (percentage) and continuous values are expressed as mean +- SD.

CTA: computed tomography angiography.

Fig. 1. A 51-year-old male was diagnosed with left hypopharyngeal cancer status post total pharyngolaryngectomy with flap reconstruction. In (A), the angiography result revealed acute extravasation from the right facial artery (black arrow). In (B), embolization with gelform emulsion (white arrow) was completed and the active bleeding ceased in seconds.

Fig. 2. Kaplan-Meier survival curves of patients with head and neck cancer bleeding.

muscle loss that cannot be treated by nutritional support and results in progressive functional impairment) [17], 1 (2.3%) tumor-related airway obstruction, and 1 (2.3%) intracranial hemorrhage. Univariate and mul- tivariate Cox regression analyses were used to investigate the risk fac- tors for 30-day mortality. The univariate predictors included heart rate

> 100 (beats/min) (HR = 2.81; Cl 1.44-5.48; p = 0.002), surgical resec- tion (HR = 0.46; Cl 0.25-0.87; p < 0.016), active bleeding (HR = 1.95; Cl 1.02-3.73; p = 0.043), inotropic support (HR = 4.26; Cl 1.08-10.01; p = 0.001), WBC > 11.0 (10 [3]/ul) (HR = 1.93; Cl 1.07-3.49; p =

0.030), Hb < 11.0 (g/dl) (HR = 2.93; Cl 1.45-5.94; p = 0.003), and PT

> 13 (sec) (HR = 2.03; Cl 1.10-3.75; p = 0.024). The multivariate anal- yses indicated that heart rate > 100 (beats/min) (HR = 2.42; Cl 1.15-5.06; p = 0.019) and inotropic support (HR = 3.00; Cl

1.14-7.89; p = 0.026) were statistically significant independent risk factors for 30-day mortality (Table 3) (See Fig. 3).

1. Discussion

To the best of our knowledge, this is the largest retrospective cohort study evaluating the short-term outcomes of patients with HNC bleed- ing and the first study on HNC bleeding in the emergency medicine lit- erature. We found that the 30-day mortality rate of HNC bleeding was as high as 18.3%, and the associated independent risk factors included a heart rate > 100 (beats/min) and inotropic support in these patients.

Previous studies have identified several Predisposing factors for HNC bleeding, including chemoradiation, oncologic head and neck surgery

Table 3

univariate and multivariate analyses of predictors for 30-day mortality with Cox proportional hazards model.

	Univariate			Multivariate
	HR(95%CI)	P value		HR(95%CI)	P value
Age	0.98(0.96,1.01)	0.248
Male	1.71(0.41,7.06)	0.460
Heart rate > 100 (beats/min)	2.81(1.44,5.48)	0.002	2.42(1.15,5.06)		0.019?
SBP < 90 (mmHg)	2.05(0.81,5.22)	0.131
Hypertension	0.71(0.37,1.35)	0.291
Diabetes mellitus	0.91(0.44,1.89)	0.793
Antiplatelet agent	0.77(0.30,1.96)	0.583
Surgical resection	0.46(0.25,0.87)	0.016	0.73(0.37,1.41)		0.347
Chemoradiation	2.45(0.97,6.22)	0.059
Neck dissection	0.56(0.30,1.06)	0.074
Flap reconstruction	0.73(0.38,1.40)	0.349
T stage	1.22(0.90,1.65)	0.211
N stage	1.22(0.92,1.61)	0.167
M stage	1.88(0.79,4.45)	0.154
Local recurrence	1.24(0.69,2.24)	0.478
Active bleeding	1.95(1.02,3.73)	0.043	1.46(0.71,3.03)		0.307
Blood transfusion	1.72(0.95,3.12)	0.072
Inotropic support	4.26(1.80,10.1)	0.001	3.00(1.14,7.89)		0.026?
WBC > 11.0 (103/ul)	1.93(1.07,3.49)	0.030	1.58(0.83,3.00)		0.164
Hb < 11.0 (g/dl)	2.93(1.45,5.94)	0.003	1.99(0.87,4.51)		0.101
PLT < 150 (103/ul)	1.18(0.55,2.55)	0.666
PT > 13 (sec)	2.03(1.10,3.75)	0.024	1.45(0.75,2.81)		0.265

HR: hazard ratio; 95% CI: 95% confidence interval.

* P value < 0.05.

Fig. 3. Kaplan-Meier survival curves of patients with head and neck cancer bleeding (A) Stratified by patients with or without heart rate > 100 (beats/min) (B) Stratified by patients with or without inotropic support.

with or without radical neck dissection, postoperative nonhealing wound, surgical site infection, pharyngocutaneous fistula, local recur- rence, as well as carotid artery invasion by fungating tumors [4]. Specif- ically, surgery has been found to potentially result in carotid blowout syndrome (CBS) in patients with HNC [15]. Combined with radical neck dissection, the carotid artery is more likely to rupture due to the poor Nutritional status of the arterial wall and pseudoaneurysm forma- tion [18]. Nevertheless, the proportion of surgical resection was signifi- cantly higher in the survivor group of the present study. The reason for this pattern is speculated to be that the patients with early stage cancer were more likely to receive surgical treatment than the patients with late stage cancer. Chemoradiation has been identified as another risk factor for CBS development and the overall survival in patients with HNC [16]. Radiation produces free radicals that weaken the adventitial layer, which supplies 80% of the blood flow to the carotid artery [19- 21]. Chemotherapy induces thrombocytopenia and vascular toxicity, which may lead to the development of thrombosis and atherogenesis [22]. The carotid artery subsequently ruptures due to the arterial wall damage caused by ischemia. In accordance with the previous findings, the current study demonstrated that significantly more patients re- ceived chemoradiation in the nonsurvivor group. Therefore, in addition to being a risk factor for long-term mortality, chemoradiation is a non- negligible risk factor for short-term mortality as well.

The patients in the current study had a slightly lower 30-day mortal- ity rate (18.3%) than the reported 30-day mortality rates in the previous studies. One recent retrospective cohort study conducted by Liang et al. reported a 30-day mortality rate of 33% in patients with CBS undergoing intervention [18]. Another study by Chen et al. noted a mortality rate within 30 days after emergency management for CBS of 21.8% [23]. The reason for this discrepancy may be because the present study en- rolled patients with HNC bleeding originating from a variety of causes. For example, the patients with postoperative wound bleeding were all alive within the 30-day follow-up period. As for the overall survival rate, prior studies have suggested cancer stage and the disease progres- sion as risk factors rather than bleeding-related complications [18]. For instance, Jacobi et al. identified 36 patients suffering from CBS and dem- onstrated that the overall survival rate of this group was comparable to that of the patients without CBS after the adjustment for Union Internationale Contre le Cancer (UICC) stage [16]. Nonetheless, we found that a heart rate > 100 (beats/min) and inotropic support were significantly associated with a higher 30-day mortality rate. Hemody- namic status, apart from cancer stage-related parameters, may serve as a vital prognostic factor for short-term survival for several reasons.

First, patients with HNC bleeding can experience ongoing sanguinous oozing interspersed with life-threatening major hemorrhage. They may expire abruptly if being asphyxiated or developing hemorrhagic shock. Additionally, the depletion of the intravascular volume to the point of cardiovascular compromise can result in reduced macro- and microcirculation followed by inadequate tissue perfusion and multior- gan failure [24-28]. As for blood pressure, the blood pressure response in patients with HNC bleeding was greatly variable and difficult to use as an early indicator of Hemodynamic compromise because of the body’s various compensation mechanisms, the amount of blood transfu- sion, and the use of inotropic agents [29]. This may result in the fact that blood pressure, whether treated as a continuous or dichotomous vari- able, did not significantly differ across the survivor or nonsurvivor groups in our cohort.

Prior studies predominantly reported the risk factors and the out- comes of patients with CBS, which accounts for only 4% of the HNC pa- tients [16,18,23,30-33]. The definition of CBS varies across studies and is mostly defined as a rupture of the carotid artery or its extracranial branches, which typically comprise the common carotid artery, the in- ternal carotid artery, and the proximal external carotid artery [4]. How- ever, such studies on CBS did not involve the majority of the HNC patients presenting to the ED with acute hemorrhage. In the present study, the positive yield rate of a target bleed on CT angiography was low, and 58.2% of the patients who received CT angiography had no pseudoaneurysm or contrast extravasation on imaging. Similarly, Cannavale et al. reported 24 patients with suspicion for CBS, and CT an- giography revealed no direct signs of bleeding in 21 (87.5%) patients [34]. Okcesiz et al. examined 13 patients with HNC bleeding who re- ceived pre-procedural CT angiography, and there were no significant findings on imaging in 7 (53.8%) patients [15]. The most likely explana- tion for this pattern is that HNC bleeding usually originates from the tumor itself with findings of hypervascular tumor and mucosal staining on imaging [7,11]. Another interesting finding in our study is that in line with the recent studies, most bleeding is located in the external carotid artery and its distal branches, which is noticeably different from the ear- lier literature on CBS [5,14,15,18,23,32].

When Life-threatening bleeding occurs, Initial resuscitation and timely standardized treatment by Multidisciplinary teams are impor- tant for the quick identification and control of the source of bleeding. Traditionally, bleeding is managed with surgical intervention, primary repair, or ligation of the artery [6]. In recent years, the endovascular ap- proaches, which include deconstruction using endovascular emboliza- tion or reconstruction with covered stents, have become the first-line

treatments [9-11]. Despite the development of multidisciplinary treatments, the prognosis of patients with HNC bleeding remains poor given the advanced cancer stage. HNC bleeding confers a signif- icant source of psychological burden for patients and their families, and acute catastrophic bleeding can be challenging for the frontline physicians [12,13]. When using inotropic agents, an indicator of short-term mortality, to support patients’ hemodynamic status, phy- sicians are recommended to inform patients and their family of the increased mortality and implement appropriate treatment plans (e.g., whether to pursue aggressive attempts to prolong life or sup- portive care) upon discussion.

1. Limitations

This study had several limitations. First, the current study was retro- spective in nature and intrinsically does not provide the robust evidence that a prospective study provides. Second, the decision to perform an open surgery, endovascular treatment, or Conservative therapy was made with the interventionalist’s clinical judgement in conjunction with the patient’s preference. Therefore, selection bias might have af- fected our results since some patients with advanced cancer might have refused intensive life-sustaining therapies. Third, there were some missing laboratory data (e.g., aspartate transaminase, albumin, and total bilirubin) in our database, so these items could not be included and analyzed in this study. Finally, this is a single-country study with predominantly Asian male population. Further large-scale international collaborative studies are needed to verify if our findings extend to other populations.

1. Conclusion

Early planning of the appropriate management of patients with HNC bleeding requires the prediction of short-term survival at the ED. A heart rate > 100 (beats/min) and inotropic support are the most impor- tant risk factors for the 30-day mortality in the HNC patients in our study. Our results may aid physicians in the evaluation of short-term survival in HNC patients with life-threatening bleeding and provide critical information for risk stratification and medical decisions.

Ethics approval and consent to participate

The study was conducted according to the guidelines of the Declara- tion of Helsinki and approved by the Institutional Review Board of Chang Gung Medical Foundation (IRB no. 202102021B0, Date of Approval: 2021/12/03).

Consent for publication

This study qualified for a waiver of informed consent.

Fundings

This research received no external funding.

Credit authorship contribution statement

Chieh-Ching Yen: Formal analysis, Investigation, Writing – original draft, Conceptualization. Heng Yeh: Supervision, Data curation, Concep- tualization. Che-Fang Ho: Conceptualization. Chien-Han Hsiao: Formal analysis, Data curation. Kuang-Yu Niu: Resources. Chung-Cheng Yeh: Resources. Jian-Xun Lu: Supervision. Chia-Chien Wu: Investigation. Yun-Chen Chang: Data curation. Chip-Jin Ng: Resources.

Declaration of Competing Interest

The authors declare no competing interests.

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

The article processing charge was supported by the Chang Gung Memorial Hospital in Taiwan.

The authors have no commercial associations or sources of support that might pose a conflict of interest.

References

1. Chou CP, Lai WA, Pan BL, Yang YH, Huang KS. The effects of hospice care for terminal head and neck cancer patients: a nationwide population-based matched cohort study. J Palliat Med. Jan 11 2021. https://doi.org/10.1089/jpm.2020.0375.
2. Hsu W-L, Yu KJ, Chiang C-J, Chen T-C, Wang C-P. Head and neck cancer incidence trends in Taiwan, 1980~2014. Int J Head Neck Sci. 2017;1(3):180-9.
3. Du E, Mazul AL, Farquhar D, et al. Long-term survival in head and neck cancer: im- pact of site, stage, smoking, and human papillomavirus status. Laryngoscope. Nov 2019;129(11):2506-13. https://doi.org/10.1002/lary.27807.
4. Suarez C, Fernandez-Alvarez V, Hamoir M, et al. Carotid blowout syndrome: modern trends in management. Cancer Manag Res. 2018;10:5617-28. https://doi.org/10. 2147/cmar.s180164.
5. Chou CT, Rath TJ, Johnson JT, Goyal LK. Catastrophic hemorrhage after chemoradia- tion for advanced stage oropharyngeal carcinoma: a case series. Laryngoscope. May 2021;131(5):1049-52. https://doi.org/10.1002/lary.29167.
6. Bergamini C, Ferris RL, Xie J, et al. Bleeding complications in patients with squamous cell carcinoma of the head and neck. Head Neck. Jun 12 2021. https://doi.org/10. 1002/hed.26772.
7. Zhao LB, Shi HB, Park S, et al. Acute bleeding in the head and neck: angiographic findings and endovascular management. AJNR Am J Neuroradiol. 2014;35(2): 360-6. https://doi.org/10.3174/ajnr.a3761.
8. Rzewnicki I, Kordecki K, Lukasiewicz A, et al. Palliative embolization of hemorrhages in extensive head and neck tumors. Pol J Radiol. 2012;77(4):17-21. https://doi.org/ 10.12659/pjr.883624.
9. Miller T, Burns J, Farinhas J, et al. Covered stents safely utilized to prevent cata- strophic hemorrhage in patients with advanced head and neck malignancy. J Neurointerv Surg. Nov 2012;4(6):426-34. https://doi.org/10.1136/neurintsurg- 2011-010083.
10. Shah H, Gemmete JJ, Chaudhary N, Pandey AS, Ansari SA. Acute Life-threatening hemorrhage in patients with head and neck cancer presenting with carotid blowout syndrome: follow-up results after initial hemostasis with covered-stent placement. AJNR Am J Neuroradiol. Apr 2011;32(4):743-7. https://doi.org/10.3174/ajnr.A2379.
11. Vilas Boas PP, de Castro-Afonso LH, Monsignore LM, Nakiri GS, de Mello-Filho FV, Abud DG. Endovascular management of patients with head and neck cancers pre- senting with acute hemorrhage: a single-center retrospective study. Cardiovasc Intervent Radiol. Apr 2017;40(4):510-9. https://doi.org/10.1007/s00270-016- 1550-8.
12. Recka K, Montagnini M, Vitale CA. Management of bleeding associated with malig- nant wounds. J Palliat Med. Aug 2012;15(8):952-4. https://doi.org/10.1089/jpm. 2011.0286.
13. Harris DG, Noble SI. Management of terminal hemorrhage in patients with advanced cancer: a systematic literature review. J Pain Symptom Manage. Dec 2009;38(6): 913-27. https://doi.org/10.1016/j.jpainsymman.2009.04.027.
14. Chatani S, Sato Y, Murata S, et al. transarterial embolization for bleeding in patients with head and neck cancer: who benefits? Laryngoscope. 2021;131(11):E2777-83. https://doi.org/10.1002/lary.29611.
15. Okcesiz I, Donmez H, Vural A, Karabiyik O, Yuce I, Cagli S. The role of CT angiography and endovascular treatment in acute-massive head and neck bleeding. Eur Arch Otorhinolaryngol. May 14 2021. https://doi.org/10.1007/s00405-021-06813-6.
16. Jacobi C, Gahleitner C, Bier H, Knopf A. Chemoradiation and local recurrence of head and neck squamous cell carcinoma and the risk of carotid artery blowout. Head Neck. Sep 2019;41(9):3073-9. https://doi.org/10.1002/hed.25796.
17. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol. May 2011;12(5):489-95. https://doi.org/ 10.1016/s1470-2045(10)70218-7.
18. Liang NL, Guedes BD, Duvvuri U, et al. Outcomes of interventions for carotid blowout syndrome in patients with head and neck cancer. J Vasc Surg. 2016;63(6):1525-30. https://doi.org/10.1016/j.jvs.2015.12.047.
19. Venkatesulu BP, Mahadevan LS, Aliru ML, et al. Radiation-induced endothelial vascu- lar injury: a review of possible mechanisms. JACC Basic Transl Sci. Aug 2018;3(4): 563-72. https://doi.org/10.1016/j.jacbts.2018.01.014.
20. Yeh S-A. Radiotherapy for head and neck cancer. Semin Plast Surg. 2010;24(02): 127-36. https://doi.org/10.1055/s-0030-1255330.
21. Weintraub NL, Jones WK, Manka D. Understanding radiation-induced vascular dis- ease. J Am Coll Cardiol. Mar 23 2010;55(12):1237-9. https://doi.org/10.1016/j.jacc. 2009.11.053.
22. Cameron AC, Touyz RM, Lang NN. Vascular complications of cancer chemotherapy. Can J Cardiol. 2016;32(7):852-62. https://doi.org/10.1016/j.cjca.2015.12.023.
23. Chen YJ, Wang CP, Wang CC, Jiang RS, Lin JC, Liu SA. Carotid blowout in patients with head and neck cancer: associated factors and Treatment outcomes. Head Neck. Feb 2015;37(2):265-72. https://doi.org/10.1002/hed.23590.
24. Rowe K, Fletcher SJ. The emergency management of the patient presenting with shock. Acute Med. 2011;10(2):103-5.
25. Standl T, Annecke T, Cascorbi I, Heller AR, Sabashnikov A, Teske W. The nomencla- ture, definition and distinction of types of shock. Dtsch Arztebl Int. 2018. https:// doi.org/10.3238/arztebl.2018.0757.
26. Cannon JW. Hemorrhagic shock. N Engl J Med. Jan 25 2018;378(4):370-9. https:// doi.org/10.1056/NEJMra1705649.
27. Aoki M, Abe T, Saitoh D, Hagiwara S, Oshima K. Use of vasopressor increases the risk of mortality in traumatic hemorrhagic shock: a nationwide cohort study in Japan. Crit Care Med. Dec 2018;46(12):e1145-51. https://doi.org/10.1097/ccm. 0000000000003428.
28. Yo CH, Lai CC, Hsu TC, et al. National trends of organ dysfunctions in sepsis:an 11- year longitudinal population-based cohort study. J Acute Med. Dec 1 2019;9(4): 178-88. https://doi.org/10.6705/j.jacme.201912_9(4).0004.
29. Little RA, Kirkman E, Driscoll P, Hanson J, Mackway-Jones K. Preventable deaths after injury: why are the traditional ‘vital’ signs poor indicators of blood loss? J Accid Emerg Med. Mar 1995;12(1):1-14. https://doi.org/10.1136/emj.12.1.1.
30. Yamazaki H, Ogita M, Kodani N, et al. Frequency, outcome and prognostic factors of carotid blowout syndrome after hypofractionated re-irradiation of head and neck cancer using CyberKnife: a multi-institutional study. Radiother Oncol. Jun 2013; 107(3):305-9. https://doi.org/10.1016/j.radonc.2013.05.005.
31. Lu HJ, Chen KW, Chen MH, et al. serum albumin is an important prognostic factor for carotid blowout syndrome. Jpn J Clin Oncol. May 2013;43(5):532-9. https://doi.org/ 10.1093/jjco/hyt043.
32. Lu H-J, Chen K-W, Chen M-H, et al. Predisposing factors, management, and prognos- tic evaluation of acute carotid blowout syndrome. J Vasc Surg. 2013;58(5):1226-35. https://doi.org/10.1016/j.jvs.2013.04.056.
33. Chang F-C, Lirng J-F, Luo C-B, et al. Patients with head and neck cancers and associ- ated postirradiated carotid blowout syndrome: endovascular therapeutic methods and outcomes. J Vasc Surg. 2008;47(5):936-45. https://doi.org/10.1016/j.jvs.2007. 12.030.
34. Cannavale A, Corona M, Nardis P, et al. Computed tomography angiography findings can predict Massive bleeding in head and neck tumours. Eur J Radiol. Apr 2020; 125:108910. https://doi.org/10.1016/j.ejrad.2020.108910.