Abstract
OBJECTIVES: Posterior epistaxis is a common otolaryngologic emergency. Management is controversial because of the many treatment options available. These options vary in efficacy, rates of complications, and cost. Posterior nasal packing is the medical management most frequently used to control posterior epistaxis. It is associated with major complications, including stroke, myocardial infarction, arrhythmias, and death. Because of these potential complications, many otolaryngologists monitor patients with posterior nasal packing in the intensive care unit (ICU). However, the level of care used to monitor these patients is variable, and standards have not been established.
METHODS: From 1991 to 1997, 46 patients had posterior nasal packing placed to control epistaxis. Management, complications, and hospital charges were analyzed.
RESULTS: Six patients (13%) were admitted to the ICU, 2 (4%) were admitted for telemetry monitoring, and 38 (83%) were sent to the ENT ward for continuous pulse oximetry. Four major complications occurred (1 episode of syncope [emergency department], 2 arrhythmias [ICU], and 1 death [hospice]). Twenty-six patients were treated with posterior packing in the ENT ward, at a mean cost of $2988. Fourteen patients underwent intervention (5 ligations, 6 endoscopic cauterizations, and 3 angiograms), with a mean cost of $5482. Six patients spent time in the ICU, with a mean cost of $8242. Patients treated in the ENT ward had significantly lower costs than those undergoing intervention (P = 0.017) or those admitted to the ICU (P = 0.020).
CONCLUSION: We propose that most patients with posterior epistaxis can be treated in specialized ENT wards. This can be done without increasing complications and with significantly decreased costs.
Posterior epistaxis is a frequently encountered emergency. The objectives of treatment are to control hemorrhage while minimizing complications. Posterior bleeding sites are difficult to identify and control outside of the operating room. Packing is considered the most conservative way to control posterior bleeding. It may be accomplished by use of gauze, a Foley catheter, or other catheters designed specifically for the nasal cavity (Fig 1). Many complications of posterior packing have been described. Major complications, including death, cerebral ischemia, and myocardial infarction, have been documented. These complications have been attributed to hypotension, oversedation, or the nasopulmonary reflex. 1 The existence of a nasopulmonary reflex with posterior nasal packing is controversial. 1 – 3 Surgery also has been used as an alternative to nasal packing. 4 – 8 Shorter hospital stays with good control rates after early surgical interventions have been reported. 6 , 9 – 12 Surgery may also be used in patients in whom conservative medical management has failed.
The risk of major complications associated with a posterior nasal pack has led to some controversy over how these patients should be monitored. The treatment of patients with posterior nasal packing varies significantly. Although many authors state that hospital admission is advisable, the level of care is rarely addressed and may vary from routine care on a nonspecialized ward to admission to a surgical intensive care unit (ICU).
At our institution we have selectively treated most patients with posterior nasal packing and placement in the ENT ward. Patients are admitted to the surgical ward where the nursing staff is experienced in the care of head and neck surgical patients. There they are monitored with continuous pulse oximetry. Another subset with significant medical problems or those requiring more sedation have been treated in the ICU. We wanted to examine whether ENT ward stay was safe, effective, and more cost effective than ICU care. The length of stay, methods of patient monitoring, complications, and costs were examined.
70134-8-fig1.png)
Methods used for posterior nasal packing. Top, Epistat catheter. Middle, Classic gauze packing. Bottom, Foley catheter.
Costs associated with hospital admission and care
70134-8-table1.png)
Costs associated with intervention for epistaxis control
70134-8-table2.png)
METHODS AND MATERIAL
Between 1992 and 1997 the records of patients admitted to Buffalo General Hospital with a diagnosis of epistaxis were reviewed. Forty-six patients were admitted and required posterior nasal packing to control epistaxis. Patients treated surgically and not requiring posterior packing and inpatients in whom epistaxis developed were excluded. Records were evaluated for age, sex, length of stay, number of days with a posterior pack in place, type of pack used, monitoring methods, diet, oxygen administration, transfusions, analgesia, sedation, surgical procedures, complications, and costs. Monitoring methods included frequency of vital signs, pulse oximetry, telemetry, and ICU stays. Charts were examined for concomitant medical conditions including hypertension, coronary artery disease, renal disease, pulmonary disease, morbid obesity, arrhythmias, and smoking and alcohol abuse. Hospital charges were analyzed by obtaining itemized billing sheets for each hospitalization. We reviewed total costs, cost per day, as well as costs for procedures and monitoring. Specific hospital charges were also reviewed and averaged. Costs for the hospital stay and monitoring can be seen in Table 1. Costs for associated interventions can be seen in Table 2.
RESULTS
Subjects were 46 patients undergoing posterior nasal packing to control epistaxis. The mean age was 63 years (range 29 to 82 years) with a male/female ratio of 1.7:1. The mean length of hospital stay was 6.0 days. Eleven patients (24%) were treated surgically, 5 with a ligation procedure and 6 with endoscopic cauterization during the hospital stay. These patients were hospitalized for an average of 5.55 days (range 1 to 13 days), whereas the 35 patients treated with packing alone were hospitalized for an average of 6.14 days (range 2 to 19 days) (P < 0.05). The patients who later underwent surgical procedures had posterior nasal packs in place a mean of 2.72 days, and those treated with packing only had their packs in place for a mean of 3.66 days. Fifteen patients (33%) received transfusions: 6 of 11 (55%) from the surgical group and 9 of 35 (26%) from the nonsurgical group (P < 0.05).
Six patients (13%) were admitted to the ICU, 2 (4%) to a telemetry bed, and the remaining 38 (83%) to the ENT ward. No step-down or intermediate care ward existed during the study period. Of those admitted to the ENT ward, 6 (16%) were given continuous pulse oximetry with alarms set for heart rate (<60 or >120 beats/minute) and saturation (<90%). Seven patients (18%) had spot checks ranging from every 2 to 12 hours; the other 25 (66%) had no oxygen monitoring. Twenty-six (68%) of those patients admitted to the ENT ward were given humidified oxygen by mask. Seven (15%) had their vital signs taken every 2 hours, 30 (65%) every 4 hours, and 9 (20%) every 8 hours.
Thirty patients’ (65%) noses were packed with gauze, 15 (33%) with an Epistat nasal catheter, and 1 (2%) with a Foley catheter. Forty-one (89%) of the patients were given prophylactic antibiotics. Nine (20%) were given nothing by mouth while the posterior pack was in place, 22 (48%) were given only liquids, and 15 (33%) were fed regular diets. Analgesics actually received by the patients varied from 0 to 75 mg of Demerol every 3 hours.
70134-8-fig2.png)
Mean hospitalization costs per patient. Difference between all 3 groups is significant (P < 0.05).
Concomitant medical conditions were present in many patients. These consisted of hypertension (50%), coronary artery disease (one third of those with previous myocardial infarctions) (50%), chronic obstructive pulmonary disease (17%), diabetes (15%), history of an arrhythmia (13%), and morbid obesity (7%). Thirty-three percent of patients were smokers, and 15% abused alcohol. One third were on aspirin or Coumadin.
Two (4%) patients had sinusitis during their hospital stay and were treated with antibiotics and decongestants. Significant facial swelling ipsilateral to an Epistat catheter developed in 1 patient; the swelling resolved after the catheter was removed. One patient had a syncopal episode on insertion of an Epistat catheter. A 4-mm area of alar necrosis developed in 1 patient with an Epistat nasal pack. Two patients were noted to have had arrhythmias. While on the telemetry ward, atrial fibrillation developed in 1 patient with chronic obstructive pulmonary disease, right heart failure, and end-stage lymphoma. With no intervention it resolved in 12 hours. The other patient with hypertension, coronary artery disease, and alcohol abuse had a 30-second run of Mobitz type II block while in the medical ICU. This patient was asymptomatic, and the arrhythmia resolved spontaneously. This patient remained in the ICU for 2 days and then was transferred to a telemetry bed until the pack was removed on day 10. There were no further events. The patient with end-stage lymphoma, a coagulopathy, and progressive right heart failure died 6 days after admission to the hospice unit. The death was not related to the nasal packing. Oxygen saturation did not drop below 90% in any patients.
For cost analysis, the patients were divided into 3 groups. Group I (26 patients [57%]) had only posterior packing with no time spent in the ICU. Their mean hospital stay was 5.9 days (range 2 to 19 days), with a mean hospital bill of $2988 (range $1199 to $6998), or $506/day. Group II (14 patients [30%]) underwent an angiogram and/or a surgical procedure but spent no time in the ICU. Their mean hospital stay was 5.8 days (range 1 to 13 days), with a mean hospital bill of $5485 (range $1937 to $9883), or $946/day. Group III (6 patients [13%]) spent part of their hospital stay in the ICU (2 to 5 days) and had no angiography or surgical treatment. Their mean hospital stay was 7.6 days (range 4 to 14 days), with a mean hospital bill of $8241 (range $4564 to $11,531), or $1084/day. There were no patients with both an ICU stay and a surgical procedure. The mean hospital stay for all patients was 6.0 days, with a mean cost of $4433, or $739/day. The difference between all 3 groups was significant (P < 0.05) (Fig 2).
DISCUSSION
Our study population is representative of patients with posterior epistaxis. The average age of 63 years with a 1.7:1 male to female ratio is seen in most reports. The incidences of hypertension, cardiac disease, diabetes, pulmonary disease, smoking, alcohol abuse, and anticoagulant use are similar to those of previous studies. The length of stay for the surgical group (5.8 days) and the 2 nonsurgical groups combined (6.2 days) is similar to that of previous studies. 4 – 8 Some studies have shown surgical treatment to reduce hospital stay. In this study the groups had almost equal lengths of stay, and conservative treatment had failed in most but not all of the patients who underwent procedures. Patients who went directly to surgery were not included in the study.
Minor complications were encountered in only 9% of our patients. These consisted of sinusitis, eustachian tube dysfunction, otitis media, and alar necrosis. There was no difference in minor complications between the Epistat and gauze packing groups. This rate of minor complications may be lower than the actual rate because of some limitations of the study. It was retrospective, and follow-up data were not available. Major complications occurred at a rate of 6%. Two arrhythmias, a syncopal episode, and 1 death occurred in 3 patients. An 81-year-old man (group III) with a history of hypertension, coronary artery disease, and alcohol abuse had a short episode of Mobitz II heart block. He was monitored with continuous pulse oximetry but had no associated desaturation. A 71-year-old woman (group II) with a history of hypertension, alcohol abuse, and smoking had a syncopal episode on insertion of an Epistat catheter while in the emergency department. She later underwent endoscopy with electrocautery. A 57-year-old man (group III) in right heart failure with pulmonary disease and end-stage lymphoma had a 12-hour episode of atrial fibrillation that resolved. He later died in the hospice unit.
Cost comparison between groups was not surprising. The average hospital cost for group I patients (packing only) of $2697 is less than that of group II patients (surgery and/or angiography), $5678. Group II represents patients in whom conservative treatment failed who went on to procedures later in their hospital course than would patients selected initially for surgical management. Group III was made up of patients with a portion of their stay in the ICU. Their cost was higher, $8244, and their stay was longer. There are 2 obvious reasons for this. First, daily charges for the ICU are much higher than those for the ENT ward, and our patients were not randomly selected for intensive monitoring. They were patients with more medical problems who were believed to be at higher risk for complications.
Most patients in group I were treated with posterior packing and admission to the ENT ward with pulse oximetry monitoring. This group did not have an increase in complications, and a significant savings was shown.
The total hospital charges for all patients in our study was $203,918, or $4433/patient. With selective management a total of 126 ICU days were avoided, saving an estimated $161,198, or $3506/patient.
CONCLUSION
This study evaluated management paradigms of patients requiring posterior nasal packs to control their epistaxis. Most patients (87%) were treated in the ENT ward. No increase in complications was seen in this group. In this study 17% of patients went on to require intervention. About 17% of patients were admitted to either an ICU or telemetry bed on the basis of the admitting physician's judgment.
We propose that by admitting most patients with posterior nasal packing to the ENT ward with continuous pulse oximetry and selectively admitting some patients to higher levels of care, resources are better allocated with decreased cost without increased complications.