Role of age and anticoagulants in recurrent idiopathic epistaxis

Abstract

Objective

Epistaxis is one of the commonest causes of attendance of Otolaryngology emergency rooms. Given its incidence, potential severity and high recurrence rate, a systematic and careful management is mandatory. This work aims to define prognostic factors of epistaxis recurrence.

Material and methods

Retrospective review of medical records of patients with epistaxis admitted to our emergency department from January 2012 to December 2016. Data of 1005 patients with idiopathic epistaxis were analysed and independent risk factors for recurrence were determined by multiple logistic regression analysis.

Results

Recurrence of epistaxis was found in 303 (30.1%) patients. Patients with recurrent epistaxis were older (p < .001) and more commonly had a history of hypertension (p = .001) and antiplatelet (p = .048) and anticoagulant (p = .001) use than those with episodic epistaxis. Age (adjusted OR 1.21, 95%CI 1.08–1.32, p = .003) and anticoagulant use (adjusted OR 2.68, 95%CI 1.94–3.70, p = .009) were predictors of increased risk of recurrent epistaxis. Gender, alcohol abuse, medical history, active bleeding at admission, unidentified bleeding point or treatment modalities were not associated with recurrence.

Conclusion

Age and use of anticoagulation drugs were risk factors for recurrence of epistaxis. None of the previously described risk factors for episodic epistaxis were found to increase the risk of recurrence. Knowledge of factors involved in recurrence might provide important information for assessment and management of increased risk patients.

Resumen

Objetivo

La epistaxis es una causa frecuente de asistencia a los servicios de urgencias de otorrinolaringología. Dada su incidencia, posible gravedad y alta tasa de recurrencia, es obligatorio un manejo sistemático y cuidadoso. Este trabajo tiene como objetivo definir los factores pronósticos de recurrencia de la epistaxis.

Material y métodos

Revisión retrospectiva de los registros médicos de pacientes con epistaxis admitidos en nuestro servicio de urgencias desde enero de 2012 a diciembre de 2016. Se analizaron los datos de 1.005 pacientes con epistaxis idiopática, y se determinaron los factores de riesgo independientes de recurrencia mediante análisis de regresión logística múltiple.

Resultados

La recurrencia de la epistaxis se encontró en 303 (30,1%) pacientes. Los pacientes con epistaxis recurrente eran mayores (p < 0,001), tenían con más frecuencia antecedentes de hipertensión (p = 0,001) y de tratamiento con antiplaquetarios (p = 0,048) y anticoagulantes (p = 0,001) que aquellos con epistaxis episódica. La edad (OR ajustada: 1,21; IC 95%: 1,08-1,32; p = 0,003) y el uso de anticoagulantes (OR ajustada: 2,68; IC 95%: 1,94-3,7; p = 0,009) fueron factores predictivos de mayor riesgo de epistaxis recurrente. El sexo, el abuso del alcohol, el historial médico, el sangrado activo al ingreso, el punto de sangrado no identificado o las modalidades de tratamiento, no se asociaron a la recurrencia.

Conclusión

La edad y el uso de medicamentos anticoagulantes fueron factores de riesgo para la recurrencia de la epistaxis. Ninguno de los factores de riesgo descritos anteriormente para la epistaxis episódica aumentó el riesgo de recurrencia. El conocimiento de los factores involucrados en la recurrencia podría proporcionar información importante para la evaluación y el manejo de los pacientes de mayor riesgo.

Introduction

Epistaxis is a common cause of attendance at emergency department. It is estimated to occur in 60% of adults over their lifetimes, requiring medical attention in 10% of the cases.¹ Despite being, in the majority of cases, self-limited,² massive haemorrhage can occur, demanding resuscitation and bleeding arrest. There is a bimodal age distribution with peak incidence in children and adults from 45 to 65 years-old.³ Epistaxis may occur following trauma, surgery or nasossinusal tumours, but it is most commonly idiopathic.⁴ Traditional risk factors for this condition include trauma, rhinitis, sinusitis, coagulopathies and atmospheric conditions (temperature and humidity).⁵ The role of hypertension and antiplatelet and antithrombotic drugs as potential factors predisposing to epistaxis remains under profound debate.^6,7 Epistaxis control is often achieved with first-line interventions as thermal or chemical cautery, haemostatic agents and nasal packing⁴; When these conservative measures fail, surgical (vessel ligation) or radiological (embolisation) interventions are required.

Several studies have been performed regarding factors associated with epistaxis, but few have addressed the specific factors underlying recurrence of this condition.^6,8

This study, carried out with 1005 patients presenting with idiopathic epistaxis, aimed to identify prognostic factors for recurrent epistaxis.

Methods

This was a retrospective study of medical files of patients diagnosed with epistaxis in our Department between January 1, 2012 and December 31, 2016.

Data collected from medical files included age, gender, antiplatelet and anticoagulation drugs use and past history (hypertension, hematologic diseases, nasal pathology, trauma). Examination findings recorded was the presence/absence of active bleeding and the bleeding point. Whenever a bleeding point was not promptly identified by anterior rhinoscopy, a thorough nasal endoscopy was performed, after having topical anaesthesia delivered with cotton strips imbibed with a solution of lidocaine 1% and adrenaline 1:100 000. In severe epistaxis, no attempt was made to assess the nose endoscopically. Regarding treatments, these were classified in local measures, cautery and nasal packing. Local measures included the use of cotton strips with vasoconstrictors, local pressure and ice. Electrocautery with curved bipolar forceps was performed whenever there was an identifiable and assessable bleeding point. Nasal packing was the chosen treatment when an epistaxis was refractory to local measures and/or cautery or in those whose bleeding point was not identifiable. In severe epistaxis, given the difficulty to assess the nose and the need for fast control of bleeding, nasal packing was routinely performed and effective with no need for surgical revision of hemostasis.

Recurrence was defined as the presence of a new episode of epistaxis within 30 days of the first. The studied variables were compared between patients with and without recurrent epistaxis.

Children under 16 years old as well as older patients with traumatic or post-operative epistaxis, nasal or paranasal sinus tumours and hereditary haemorrhagic telangiectasia (HHT) were excluded from this study.

Statistical analysis

Data were analysed using IBM Statistic Package for Social Sciences v 23. Results of descriptive statistics analysis were reported as mean, median, standard deviation and frequency. Baseline characteristics of patients, disease and treatment were stratified according to the presence of recurrence of epistaxis. Comparison between these two groups was performed with Chi-Square test and Student's T-test in order to define risk factors for recurrence of epistaxis. These risk factors were, then, analysed using the stepwise method in the multivariate logistic regression analysis. Statistic significance was assumed for p < 0.05.

Results

Population characteristics

From 2012 to 2016, 12 848 patients were admitted in our Emergency Department and 1428 (11.0%) had epistaxis. After excluding 41 patients with traumatic epistaxis, 13 with nasossinusal tumours, 21 with recent nasal surgery, 7 with HTT, 184 with missing data in clinical files and 157 children, our study population was 1005 patients. Of these, 58.8% were males. The median age was 61.0 years – Fig. 1.

Figure 1.

Distribution of epistaxis by age group.

Baseline characteristics of the patients, stratified by recurrence, are summarised in Table 1. Recurrence of epistaxis was found in 303 (30.1%) patients. Patients with recurrent epistaxis were significantly older than those who had a single episode of epistaxis (p < 0.001). The majority of patients (92.9%) enrolled in this study had no past diseases and this was true for both groups. History of hypertension was present in 541 (53.8%) patients with epistaxis and this finding was more frequent in cases of recurrence (p = 0.001). Other traditional risk factors, namely alcohol and tobacco use, presence of deviated nasal septum, rhinitis and sinusitis, history of nasal surgery and presence of haematological disorders, were studied and there were no significant differences in these factors between patients with and without epistaxis. Antiplatelet drugs were used in 21.8% patients with and 16.5% patients without recurrent epistaxis, respectively (p = 0.048). Similarly, anticoagulation was more commonly found in cases of recurrent epistaxis (p = 0.001).

Table 1.

Baseline characteristics stratified by recurrence of epistaxis.

Variable	Cases (n = 1005)		p	Odds ratio
	Recurrent (n = 303)	Not recurrent (n = 702)
Age	60.2 ± 22.7	50.3 ± 26.9	0.000**	1.016
Gender
Male	180	411	0.834*	1.036
Female	123	291	0.834*	1.036
Past histories
Hypertension	187 (61.7%)	355 (50.4%)	0.001*	1.585
Nasal surgery	3 (1.0%)	18 (2.6%)	0.149*	0.380
Alcohol abuse	5 (1.7%)	6 (0.9%)	0.322*	1.946
Hematologic disorder	10 (3.3%)	15 (2.1%)	0.276*	1.563
Medication
Antiplatelet	66 (21.8%)	116 (16.5%)	0.038*	1.434
Anticoagulant	80 (26.4%)	118 (16.8%)	0.001*	1.775
Active bleeding	207 (68.3%)	431 (61.4%)	0.039*	0.737
Bleeding point
Kiesselbach's plexus	255 (84.2%)	585 (83.3%)	0.781*	1.063
Posterior	18 (5.9%)	24 (3.4%)	0.085*	1.784
Not identified	30 (9.9%)	93 (13.2%)	0.143*	0.720
Inpatient treatment	4 (1.3%)	14 (2.0%)	0.607*	0.657
Treatment
Local measures	37 (12.2%)	114(16.2%)	0.103*	0.717
Cautery	118 (38.9%)	327(46.6%)	0.028*	0.736
Nasal packing	148 (48.8%)	261 (37.2%)	0.001*	0.613

*

x² test.

**

T-student test.

Epistaxis characteristics

Epistaxis occurred more commonly from December to March – Fig. 2. Active bleeding was reported in 207 (68.3%) patients with recurrent epistaxis and 431 (61.4%) patients without recurrent epistaxis (p = 0.039). Bleeding point was identifiable in 840(87.8%) epistaxis and there were not statistical differences in location of bleeding points between patients with or without recurrent epistaxis – Table 1.

Figure 2.

Distribution of epistaxis by month of admission to Hospital.

Treatment characteristics

Treatments included local measures (12.2%), electrocautery (38.9%) and nasal packing (48.8%). The latter was performed in 148(48.8%) patients with and 261 (37.2%) patients without recurrent epistaxis (p = 0.001). Absorbable (Spongostan®, Surgicel®) or/and non-absorbable materials (Merocel®) were regularly used for packing. Two-balloon catheters were used for posterior epistaxis control in 11(1.1%) patients. Electrocautery was more commonly performed in patients without recurrent epistaxis (p = 0.009). 18 (1.8%) patients received inpatient treatment; the median length of stay was 4.0 days.

Risk factors for recurrent epistaxis

On univariate analysis, age (unadjusted OR 1.02, 95%CI 1.01–1.04, p = 0.000), active bleeding (unadjusted OR 0.74, 95%CI 0.55–0.98, p = 0.037), history of hypertension (unadjusted OR 1.58, 95%CI 1.20–2.07, p = 0.001), antiplatelet drugs (OR 1.43, 95%CI 1.02–2.00, p = 0.035), anticoagulants (OR 1.78, 95%CI 1.18–2.33, p = 0.001) and nasal packing (unadjusted OR 1.61, 95%CI 1.22–2.10, p = 0.001) were predictive of recurrent epistaxis, whereas cautery (unadjusted OR 0.74, 95%CI, p = 0.028) were predictive of decreased risk of recurrence. Results of multivariate analysis are presented in Table 2. After adjusting for possible confounders, age (adjusted OR 1.21, 95%CI 1.08–1.32, p = 0.003) and anticoagulation (adjusted OR 2.68, 95%CI 1.94–3.70, p = 0.009) were positive predictors of recurrent epistaxis. No other studied factors, namely gender, alcohol abuse, bleeding point or inpatient treatment were associated with recurrence of epistaxis.

Table 2.

Multivariate analysis of predictors of epistaxis recurrence.

Variable	p	Odds ratio	95% IC OR
			Low	High
Age	0.003	1.212	1.008	1.324
Active bleeding	0.144	0.799	0.592	1.080
Hypertension	0.705	1.074	0.740	1.560
Anticoagulant	0.009	2.682	1.944	3.702
Antiplatelet	0.165	1.312	0.894	1.924
Electrocautery	0.110	0.364	0.105	1.256
Packing	0.065	0.717	0.535	0.961

Discussion

Epistaxis is one of the commonest otolaryngologic emergencies and it is estimated to occur in 60% of adults over their lifetime, requiring medical attention in 10% of those cases.¹ During the studied period, epistaxis were responsible for 11% of all emergencies. This condition was more commonly found in males (58.8%), patients older than 50 years old (64.8%) and the bleeding point was mainly the Kiesselbach's plexus (49.5%), findings that are globally supported by the literature.^6,9–12 Epistaxis were more frequently observed during the cold months of winter, similarly to what has been previously described.^12–14

Several studies attempted to describe the risk factors for idiopathic epistaxis. Age above 50 and male gender are consensually more common among patients with epistaxis.³ The use of anticoagulants is believed to be associated with superior risk of nosebleed, requiring longer inpatient stays and more aggressive haemostatic measures.^6,11,15,16 Impact of other conditions, namely hypertension and antiplatelet therapies is controversial. Positive correlation between hypertension and epistaxis has been described,^2,17–19 although subsequent studies have proven it otherwise.^20–23 Gifford and Orlandi²⁴ observed that patients with hypertension presented with epistaxis of more difficult control. Regarding the use of aspirin, an increased risk of epistaxis,^25–27 and need of hospitalization²⁵ have been described, although other authors have advocated that frequency of aspirin use was not superior among patients with this condition.^15,28

Recurrence of epistaxis is a major concern in ENT practice and it should be investigated for the underlying cause. In the present study presence of active bleeding at hospital admission, past history of hypertension, use of anticoagulant and antiplatelet drugs and treatment with electrocautery were more frequently associated with recurrence. Patients with recurrent epistaxis were older than patients with episodic epistaxis. Nevertheless, after adjustment for confounding factors, only age and use of anticoagulation drugs were significant risk factors for epistaxis recurrence. Several authors have studied risk factors for single epistaxis, but few articles addressed factors associated with recurrence and, although some authors have attempted to describe factors related to refractory epistaxis, namely hypertension, aspirin, alcoholism, antiplatelet and antithrombotic drugs,^17,25,29 these studies lacked a multivariate analysis. Accordingly, recent publications have provided more accurate information of the role of some factors on recurrence of epistaxis. Ando et al.,⁶ in their series of 299 patients have reported that unidentified bleeding point increased the risk of recurrent epistaxis and suggested that electrocautery should be the first choice of treatment as it was associated with inferior number of recurrences, when compared with gauze packing. Other authors⁸ have identified hypertension, diabetes mellitus and congestive heart failure as well as the use of warfarin, regardless of INR, as factors that increase the risk of recurrence of epistaxis. Laffont et al.¹² also reported anticoagulation as a risk factor and introduced two variables, which had not been previously studied, active bleeding at hospital admission and smoking, both associated with increased risk of recurrence. Additionally, inpatient treatment was a negative predictor of recurrence, according to these authors. More recently, Kobayashi et al.³⁰ have studied the influence of antiplatelet and antithrombotic drugs on recurrence of epistaxis and observed that these were not risk factors.

The present study corroborates with the previous^8,12,31 regarding the role of anticoagulation. Additionally, it defined age as a potential factor involved in epistaxis recurrence. Future studies, namely prospective might add important information regarding this subject. Limitations of this study include its retrospective and descriptive design, size of the sample and the lack of use of some treatment modalities, namely arterial embolisation and ligation. Additionally, some recurrences might have been missed if patients had sought medical attention at other institutions.

Conclusion

In the present study, age and use of anticoagulation drugs were risk factors for recurrence of epistaxis. None of the other previously described risk factors for episodic epistaxis were found to increase the risk of recurrence. Knowledge of factors involved in recurrence might provide important information for assessment and management of this increased risk patients.

Conflict of interests

Authors have no conflict of interests to declare.