1. Description of the problem
Epistaxis is the most common otolaryngologic emergency, with an overall incidence of 108 per 100,000 per year. Sixty percent of the populace at large will have at least 1 episode of epistaxis in their lifetime. Though epistaxis in the ICU may have a larger subset of the difficult or unusual causes, the majority are due to common, well-understood, and easily managed causes.
As with all bleeding, in an emergency, adequate pressure applied to the correct location will stop the bleeding. Most bleeds can be stopped with exterior nasal pressure or coagulated using rhinoscopy or nasal endoscopy and various coagulation methods. In a patient requiring Intensive Care Unit or Critical Care management, however, the rare causes of epistaxis and severe bleeds are seen more frequently than in the general populace, and therefore when early measures are unsuccessful expert consultation should be sought.
If basic measures outlined below are unable to control the bleeding, seek expert consultation with someone with experience in nasal endoscopy and epistaxis management.
In most cases of epistaxis the initial presenting symptom will be either blood seen emanating from the nares or a recurrent expectoration of blood from the mouth. Posterior oropharyngeal drainage will confirm that the expectoration of blood is from a nasal source and is not hematemesis or hemoptysis. 90-95% of cases of bleeding will be from the anterior nasal cavity, with 80% from the anterior septum. Bleeding from the anterior septum can be quickly seen with a light source often not requiring a nasal speculum.
Kiesselbach’s plexus (a.k.a. Little’s area) is the most frequent site of bleeding, representing the large majority of bleeds in an area approximately 1.5 cm from the anterior mucocutaneous junction where the terminal branches of the anterior ethmoid artery, posterior ethmoid artery, posterior nasal artery, and nasolabial artery anastomose. Other anterior bleeding can be seen on the inferior turbinate’s anterior surface. Bleeding that cannot be quickly seen with anterior rhinoscopy using a nasal speculum should prompt the use of an endoscope to evaluate for deeper or more superior sources of bleeding.
Posterior bleeding is routinely discussed but is a rare cause of bleeding, representing only 5-10% of bleeding from the nasal cavity. Suspicion of a posterior bleed is elevated when the anterior septum and inferior turbinate are evaluated and no source of bleeding is seen. Posterior bleeds are frequently more severe than anterior bleeds, but do not be fooled: the nasal cavity is a small area and can “overflow” with very small volumes of blood. Posterior bleeding is most frequently from branches of the sphenopalantine artery or from the artery itself and will thus emanate from the lateral nasal wall posterior to the middle turbinate.
Other common clinical features include declining hemoglobin/hematocrit, regurgitation of swallowed blood, pallor, hypovolemic shock, and eventual hemodynamic compromise.
Key management points
Theoretical management of spontaneous epistaxis is a simple process, but in practice it can be significantly more difficult, especially in a critically ill patient.
1) Pressure – External pressure occluding the nares bilaterally and completely for 2 rounds of 15 minutes without release of pressure to “look and see if bleeding has stopped” will control the majority of anterior bleeding. Oxymetazoline or phenylephrine can be sprayed into the nares before either or after both attempts at external pressure. Alternatively, pledgets moistened with oxymetazoline or phenylephrine can be placed just inside the floor of the nose and pressure held over these for the hemostatic effect of the medication as well as pressure.
2) Evaluation of the source of bleeding – Using a headlamp and nasal speculum, anterior rhinoscopy can demonstrate the location of approximately 90% of bleeds, with Kiesselbach’s plexus the source of 80%. Using a Frazier-tip suction, attempt to evaluate the bleeding, suctioning pooled blood as much as possible as you evaluate anterior to posteriorly.
If a source of bleeding is identified or a large vessel that appears to be a culprit is seen, coagulation either via silver nitrate or with suction diathermy is the next step. Do not coagulate bilaterally on the septum in a similar location as this has been shown to cause septal perforation. If this does not demonstrate an obvious source of bleeding nasal endoscopy may be necessary, and thus expert consultation is warranted.
3) Nasal packing – Due to the acuity of care or the cause of bleeding, coagulation may not be possible and thus the nasal cavity will need to be packed. This can be performed either using commercially available products such as Rapid Rhino® or Pope Merocel®. These products should be covered in antibiotic ointment and inserted along the floor of the nasal cavity with firm steady pressure.
Traditional nasal packing is highly effective and undertaken using antibiotic ointment-covered gauze strips packed in a layered form starting posterior inferiorly and continuing anteriorly then superiorly. The decision between traditional packing and the commercially available products is one of personal preference and timing than definitive superiority. Traditional packing can be time-consuming and require expert placement, whereas commercially available products can be placed in seconds with minimal instruction.
Packing of any type should remain in place for at least 24-72 hours to prevent rebleeding when the packing is removed. Evidence for or against systemic antibiotics with nasal packing is poor; however, reports of toxic shock syndrome associated with nasal packing are numerous, and thus gram-positive antibiosis should be considered for the duration of the packing. Though systemic antibiotics have poor evidence, local antibiotics have substantially more potent recommendations and should be applied to any packing source before placement.
4) Evaluation of the cause of bleeding – If a traumatic cause is not immediately evident, evaluation of the patient’s coagulation status, as well as thrombocyte quantification and functional assays, should be undertaken. Most causes of epistaxis are idiopathic; however, all efforts should be done to evaluate for correctable causes, including hypertension, coagulopathy, dry air, and foreign bodies.
Foreign bodies should obviously be removed and this may be as simple as switching from a nasogastric tube to an orogastric tube. Humidified air should be placed on all patients with epistaxis and hypertension should be managed appropriately. Coagulopathy has been directly studied and should be corrected if the INR ≥ 4 with fresh frozen plasma. Thrombocytopenia with platelets < 50 should be corrected to > 50 with appropriately crossmatched platelets. If platelets are found to be non-functioning, correction with crossmatched platelets is appropriate.
2. Emergency Management
As above, the key steps to management are to first control the bleeding. Within patients in the critical care setting this may prove to be more difficult than an emergency department, office, or hospital ward setting, as more frequently the bleeding may be due to coagulopathy, continuous positive airway pressure, posterior bleeding, long-term indwelling foreign bodies (nasogastric tubes or nasal feeding tubes), and facial trauma, which are more prevalent in the critical care setting than the general populace.
Where possible, attempt control/evaluation of bleeding as above while paying close attention to the already tenuous hemodynamic stability of the critically ill patient. For all suspected facial trauma bleeding causes, seek expert consultation immediately with a trained otolaryngologist or craniomaxillary facial trauma surgeon.
At the presentation of the bleeding all attempts should be made to quantify the bleeding and the period of time the bleeding has been occurring. Quite obviously, 250 mL of bleeding within minutes is significant, but if it has occurred over the last 2 hours, time can be taken to find bleeding vessels. If possible (considerations made for the need for CPAP), attempt to follow the above process: hold pressure with or without the use of vasoconstricting medications; evaluate the source of bleeding with coagulation; pack the nose if necessary; evaluate the cause.
In emergent situations this may not be possible and moving straight to nasal packing may be necessary. If this is the case the commercially available products (Rapid Rhino® inflatable epistaxis packing or Pope Merocel® sponges) should be the choice due to the alacrity with which they can be placed. Again, these are placed with firm pressure along the floor of the nose.
Recently 2 small blinded studies have shown decreased patient discomfort and rebleeding with Rapid Rhino® products, but further investigation needs to be performed to firmly recommend these over other products, as the studies involved nasal surgery, not strictly epistaxis. If immediate nasal packing is required, urgent otolaryngologic consultation is highly recommended. Continued slow bleeding after packing is routinely seen for the first 24 hours and sometimes longer in patients with coagulopathy.
This packing may not be sufficient with posterior bleeds, especially proximal bleeds of the sphenopalantine artery or its branches. If bleeding continues, posterior packing can be undertaken but should be performed by someone with experience guiding the therapy.
The key is to occlude the choanae bilaterally. This can be undertaken with commercially available inflatable posterior packing systems, using a double-pass method with gauze packing or an 18 French Foley catheter with a 30-mL balloon. Inflatable posterior packing is significantly faster and better tolerated by patients; however, rebleeds are more common with inflatable packing than traditional gauze packing.
If the commercially available inflatable posterior packing is chosen, the pack is passed along the floor of the nasal cavity until completely in the nose. The posterior balloon is inflated, then the anterior balloon as well. The gauze pack is placed by first placing a catheter in the nasal cavity until it is seen in the oropharynx. Then, using forceps, the catheter is pulled out of the mouth. A gauze pack with 2 strings attached (one to remain in the nose holding the pack in place and one to remain in the mouth for retrieval of the pack) is then tied to the catheter and then pulled until it wedges in the choanae.
Using the nasal string, the pack is secured to the nose. Be wary of pressure necrosis of the columella or alar rim. If the Foley catheter is chosen, the catheter is placed in the nose and passed until it is seen in the oropharynx. The balloon is then inflated to 15 mL and retracted against the choanae. Anterior nasal packing is then performed with either the commercially available products or gauze strips as per above. The catheter is secured at the nose with care to protect the columella.
These steps are repeated in the second naris if necessary. Once the bleeding has been controlled, evaluation by a trained endoscopist can proceed and find bleeding sources. In these patients intubation may be necessary to protect the airway.
If these steps do not control the bleeding, further evaluation must be undertaken by a trained nasal endoscopist immediately. Bleeding that cannot be controlled by packing, including posterior packing, is routinely due to bleeding from sources that require surgical ligation. Endoscopic sphenopalantine artery ligation is highly successful, as is anterior ethmoid artery ligation, without the morbidity of more drastic maxillary artery, external carotid, or internal carotid ligation, though these methods may be necessary.
If emergent evaluation by an otolaryngologist/rhinologist is not possible, embolization or superselective embolization can be performed with similar overall outcomes but significantly greater risks.
See flow chart in Figure 1.
Drugs and dosages
The medications involved in the treatment of epistaxis are limited:
— Oxymetazoline HCl 0.05% 1 or 2 sprays nasal solution in each nostril every 12 hours
— Phenylephrine 1% 1 or 2 sprays in each nostril every 4 hours
The diagnostic process for epistaxis is quite simple, with a complex differential including local causes – idiopathic/ multifactorial (most common), trauma (postoperative, nose picking, foreign body, facial fractures, nasal oxygen, CPAP), inflammatory/ Infectious (allergic, viral or bacterial rhinosinusitis, granulomatous diseases, environmental irritants), neoplastic (hemangioma, hemangiopericytoma, nasal papilloma, pyogenic granuloma, angiofibroma, carcinoma), structural (septal deformity, perforation, or spur), drugs (topical steroids, cocaine, occupational substances), and systemic (hypertension, arteriosclerosis, coagulopathy, platelet deficiencies or dysfunction, leukemia, von Willebrand’s disease, hereditary hemorrhagic telangiectasia, liver failure).
The initial diagnostic workup comprises lab studies to evaluate for systemic causes and nasal endoscopy to evaluate for traumatic, neoplastic, infectious, or structural causes. As above, the primary process is not usually discovered. To further that point, in a prospective study of routine coagulation profiles in patients presenting with epistaxis to the ED, 5% of those patients on warfarin with epistaxis had a supratherapeutic INR, 2.5% of those patients admitted for epistaxis and only 2% overall had an elevated INR. Further studies need to be obtained to determine if this holds true for larger patient populations.
Nasal endoscopy should be performed on all patients who are not found to have a single source of anterior septal bleeding on anterior rhinoscopy. Further imaging should be performed only if there is clinical suspicion of a traumatic, neoplastic, infectious, or deep vessel process.
Complete blood count, coagulation tests, platelet function assays, liver function tests, CT sinus, sinus MRI, arteriogram, nasal endoscopy
See nasal endoscopy photo of hemagioma of the nasal cavity (Figure 2) and axial CT of head with juvenile nasopharyngeal angiofibroma (Figure 3).
Understanding the pathophysiology of epistaxis requires a thorough understanding of the vascular anatomy. The blood supply for the nasal cavity is from both the internal and external carotid. As stated previously, the majority of cases of epistaxis are due to anterior bleeds from the Kiesselbach’s plexus, an area approximately 1.5 cm posterior to the mucocutaneous junction.
The vessels of this region are anastomoses from the anterior ethmoid artery, posterior ethmoid artery, posterior nasal artery, and nasolabial artery. The ethmoidal arteries arise as terminal branches of the ophthalmic artery as the first branch of the internal carotid. Severe bleeding due to trauma is frequently from naso-orbito-ethmoid fractures that transect the anterior ethmoid artery. The posterior nasal artery is a terminal branch of the sphenopalantine artery, which is the final branch of the external carotid. The nasolabial artery is a terminal branch of the facial artery, the fourth branch of the external carotid.
Posterior bleeds arise from 2 different regions, the sphenopalantine artery and Woodruff’s region. Woodruff’s region is an anastomotic region posterior and inferior to the middle turbinate where the sphenopalantine artery, arising as the final branch of the external carotid, meets the pharyngeal branches of the internal maxillary artery, the penultimate branch of the external carotid.
The sphenopalantine artery itself is subject to bleeding as it exits the sphenopalantine foramen, posterior to the middle turbinate. This is the classic location for posterior bleeding without a traumatic cause. The vessels of this region are described as having little muscular wall and little intervening submucosal tissue.
The pathophysiology of bleeding in the nasal cavity is similar to the pathophysiology of bleeding anywhere in the body. There must be full-thickness damage to vessels due to trauma, barotrauma, or destruction from neoplastic or infectious sources. Although some form of “trauma” must occur to the vessel, spontaneous epistaxis is the most common etiology, which is thought to be multifactorial. Most patients deny a history of nose picking as a source of trauma and should be merely encouraged that it is often an unrecognized habit and will result in rebleeding, so all efforts to curtail nose picking should be undertaken.
Hypertension is often discussed as an etiology for barotraumatic rupture, but many argue hypertension is not a significant cause and only an exacerbating factor. Due to the nature of the nasal mucosa as respiratory epithelium, excessive drying is poorly tolerated and a common etiology for cracking of the skin barrier and damage to surrounding vessels. Coagulopathy, though the second most commonly cited systemic cause of epistaxis, does require a primary cause for the vascular damage, and both should be treated as underlying pathology.
Frank facial trauma is often found in intractable anterior bleeding with damage to the anterior ethmoid artery via a naso-orbito-ethmoid (NOE) fracture. This is a common fracture from anterior blows to the nasal bridge and should be evaluated and treated by a CMF specialist or an otolaryngologist. Neoplastic causes of epistaxis can be from direct vasoinvasion, solid tumors outstripping their own blood supply, or apoptosis.
Hereditary hemorrhagic telangiectasias are an exotic and well-described cause for recurrent epistaxis in those without other factors. HHT is an autosomal dominant disorder characterized by arteriovascular malformations/ telangiectasias affecting the cutaneous and mucosal surfaces. 90% of these patients will have epistaxis with varying severity. This is no doubt a “zebra” among causes of epistaxis, and prolonged discussion is beyond the scope of this text except to state that recurrent episodes of epistaxis are common in these patients, and they may require nasal closure at the choanae to prevent further bleeding.
As above, epistaxis is the most common otolaryngologic emergency, with at least 60% of the populace at large experiencing at least 1 episode of nose bleeding and 6% requiring hospitalization for it. Approximately 1 in 200 ED visits are epistaxis-related, with bimodal peaks in those less than 10 years of age and those greater than 70 years of age.
December to February shows a relative increase of 47% from the other 9 months of the year, lending further support to the theory that cool dry air is associated with epistaxis. True idiopathic epistaxis represents about 10% of the causes when dry air, microtrauma, and hypertension are included as possible etiologies. Only in the rare instance of a neoplastic or coagulopathic process is the exact cause of epistaxis determined.
What's the evidence?
Description of the problem
Simmen, DB, Jones, NS, Cummings, CW, Flint, PW, Harker, LA. “Epistaxis”. Otolaryngology, Head and Neck Surgery. vol. 5. 2005. pp. 682-693. (Excellent and thorough review of epistaxis.)
Derkay, CS, Hirsch, BE, Johnson, JT. “Posterior nasal packing. Are intravenous antibiotics really necessary?”. Arch Otolaryngol Head Neck Surg. vol. 115. 1989. pp. 439-41. (Cultured removed antibiotic-impregnated packing only demonstrated heavy colonization with polymicrobial, including gram-negative organisms. Systemic and impregnated antibiotic group demonstrated light colonization of removed packing with gram-positive organisms.)
Ha, JF, Hodge, JC, Lewis, R. “Comparison of nasendoscopic-assisted cautery versus packing for the treatment of epistaxis”. ANZ J Surg. vol. 81. 2011 May. pp. 336-9. (Endoscopic cautery is more effective than packing and significantly more cost-effective.)
García Callejo, FJ, Muñoz Fernández, N, Achiques Martínez, MT. “[Nasal packing in posterior epistaxis. Comparison of two methods]”. Acta Otorrinolaringol Esp. vol. 61. 2010 May-Jun. pp. 196-201. (Posterior packing with inflatable material is faster and better tolerated, but traditional gauze packing has fewer rebleeding complications, 17% vs 26%.)
Biswas, D, Mal, RK. “Are systemic prophylactic antibiotics indicated with anterior nasal packing for spontaneous epistaxis?”. Acta Otolaryngol. vol. 129. 2009 Feb. pp. 179-81. (Systemic antibiotics are not necessarily indicated in anterior packs left in place for less than 24 hours.)
Burton, MJ, Doree, CJ. “Interventions for recurrent idiopathic epistaxis (nosebleeds) in children”. Cochrane Database Syst Rev. vol. 4. 2004. pp. pCD 004461(Excellent overall review of treatment methods of epistaxis for children.)
Choudhury, N, Sharp, HR, Mir, N. “Epistaxis and oral anticoagulant therapy”. Rhinology. vol. 42. 2004. pp. 92-97. (Patients with INR < 4 with associated epistaxis should be treated with absorbable packing such as oxidized cellulose. Those with INR > 4 with associated epistaxis should be immediately treated with FFP.)
Lanier, B, Kai, G, Marple, B. “Pathophysiology and progression of nasal septal perforation”. Ann Allergy Asthma Immunol. vol. 99. 2007 Dec. pp. 473-9. (Bilateral cautery of the nasal septum is a known cause of septal perforation.)
Márquez Moyano, JA, Jiménez Luque, JM, Sánchez Gutiérrez, R. “[Toxic shock syndrome associated with nasal packing]”. Acta Otorrinolaringol Esp. vol. 56. 2005 Oct. pp. 376-8. (Most recent case report of toxic shock syndrome due to nasal packing.)
Kucik, CJ, Clenney, T. “Management of epistaxis”. Am Fam Physician. vol. 71. 2005 Jan 15. pp. 305-11. (Primary care guide to management of epistaxis with diagrams for traditional packing methods.)
Pope, LE, Hobbs, CG. “Epistaxis: an update on current management”. Postgrad Med J. vol. 81. 2005. pp. 309-314. (Junior physician guide to general epistaxis management.)
Ozcan, C, Vayisoglu, Y, Kiliç, S. “Comparison of rapid rhino and merocel nasal packs in endonasal septal surgery”. J Otolaryngol Head Neck Surg. vol. 37. 2008 Dec. pp. 826-31. (Bleeding after removal of Merocel packing was noted in 15 patients. Rapid Rhino demonstrated no bleeding. Pain was decreased while packs were in place with Rapid Rhino and after removal.)
Arya, AK, Butt, O, Nigam, A. “Double-blind randomised controlled trial comparing Merocel with Rapid Rhino nasal packs after routine nasal surgery”. Rhinology. vol. 41. 2003 Dec. pp. 241-3. (Rapid Rhino causes less pain on removal and less bleeding. Pain while in place is similar.)
hakeel, M, Trinidade, A, Iddamalgoda, T. “Routine clotting screen has no role in the management of epistaxis: reiterating the point”. Eur Arch Otorhinolaryngol. vol. 267. 2010. pp. 1641-4. (INR elevated in only 2% of overall populace with epistaxis and supratherapeutic in 5% of warfarin users with epistaxis.)
Wormald, PJ, Bailey, BJ, Johnson, JT, Newlands, SD. “Epistaxis”. Head & Neck Surgery: Otolaryngology. 2006. pp. 505-514. (Overall workup and treatment of epistaxis.)
Chiu, T, Dunn, JS. “An anatomical study of the arteries of the anterior nasal septum”. Otolaryngol Head Neck Surg. vol. 134. 2006. pp. 33-36. (Kiesselbach’s plexus is the most common bleeding site, and specific anatomy of the anterior nasal septum is reviewed.)
Schwartzbauer, HR, Shete, M, Tami, TA. “Endoscopic anatomy of the sphenopalatine and posterior nasal arteries: implications for the endoscopic management of epistaxis”. Am J Rhinol. vol. 17. 2003. pp. 63-66. (Review of the posterior arterial anatomy)
Small, M, Murray, JAM, Maran, AGD. “A study of patients with epistaxis requiring admission to hospital”. Health Bull (Edinb). vol. 40. 1982. pp. 20-29. (Epidemiologic data for epistaxis.)
Pallin, DJ, Chng, YM, McKay, MP. “Epidemiology of epistaxis in U.S. emergency departments, 1992 to 2001”. Ann Emerg Med. vol. 46. 2005. pp. 77-81. (Epidemiologic data for epistaxis.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.