You are on page 1of 7

Original Research—Sinonasal Disorders

Head and Neck Surgery

Successful Implementation of a Clinical 2016, Vol. 155(5) 879–885

Ó American Academy of
Otolaryngology—Head and Neck
Care Pathway for Management of Surgery Foundation 2016
Reprints and permission:
Epistaxis at a Tertiary Care Center
DOI: 10.1177/0194599816657045

Peter S. Vosler, MD, PhD1, Jason I. Kass, MD, PhD2,

Eric W. Wang, MD1, and Carl H. Snyderman, MD, MBA1

No sponsorships or competing interests have been disclosed for this article. pistaxis is a common problem encountered by otolar-
yngologists. The majority is anterior septal bleeding
easily and conservatively managed by the patient with-
out seeking medical treatment.1,2 Management issues arise
Objective. We compare the management of patients with when epistaxis emanates from the sphenopalatine artery (SPA)
severe epistaxis before and after the implementation a clini- posteriorly or when patients are anticoagulated. These patients
cal care pathway (CCP) to standardize care, minimize hospi- require greater intervention and have longer hospital stays.3
tal stay, and decrease cost. Early SPA ligation (ESPAL) has the highest control rate
Study Design. Single prospective analysis with historical control. for severe epistaxis (87%-97%)4-6; patients have improved
comfort versus packing7; it has a lower side-effect profile
Setting. Tertiary academic hospital. than that of embolization8; and it results in shorter hospital
Subjects and Methods. Patients treated for epistaxis between stays and lower cost.9-11 Despite the evidence in support of
October 2012 to December 2013 were compared with a ESPAL, care of patients with severe epistaxis is not standar-
prospective analysis of patients treated for severe epistaxis dized, and there is variability in practice between and within
after implementation of a CCP from June 2014 to February institutions. Implementation of an ESPAL protocol for severe
2015. Severe epistaxis was defined as nasal bleeding not able epistaxis therefore requires a multidisciplinary approach by
to be controlled with local pressure, topical vasoconstric- implementation of a clinical care pathway (CCP).
tors, or simple anterior packing. CCPs are evidence-based protocols used to implement treat-
ment guidelines and minimize variability in patient care to
Results. Severe epistaxis was similar in the pre- and post- optimize outcome and reduce cost.12 They require the coopera-
CCP cohorts: 24.7% (n = 42) vs 18.9% (n = 22), respec- tion of clinical providers and hospital management, identifica-
tively. Implementation of early sphenopalatine artery ligation tion of barriers to implementation, and clear communication
resulted in decreased number of days packed (3.2 6 1.6 to among those involved to effect quality improvement.13
1.4 6 1.6; P = .001), decreased hospital stay (5.2 6 3.9 to The purpose of this study is to measure the effect of a CCP
2.1 6 1.3 days; P \ .001), an increased percentage of sphe- for the treatment of severe epistaxis. For the purposes of this
nopalatine artery ligations (31.0% vs 54.5%; P = .035), admis- study, severe epistaxis is defined as (1) nasal bleeding not able
sion to an appropriate hospital location with access to key to be controlled with local pressure, simple anterior packing,
resources (41.7% vs 83.3%; P = .007), and decreased overall chemical cautery, or topical vasoconstrictors or (2) hemorrhage
cost of hospitalization by 66% ($9435 saved). No patients that compromises the airway. A retrospective review of epis-
received embolization after the CCP was implemented. taxis management and outcomes was undertaken and com-
Conclusions. Implementation of a CCP decreased hospital pared with prospective outcomes after establishment of a CCP.
stay and days of packing, facilitated definitive care in patients
with severe epistaxis, improved patient safety, and
decreased cost. The results of this study can serve as a Department of Otolaryngology–Head and Neck Surgery, University of
model for the management of severe epistaxis and for Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
Department of Otolaryngology–Head and Neck Surgery, Boston
future quality improvement measures.
University, Boston, Massachusetts, USA

This article was presented at the 2015 AAO-HNSF Annual Meeting and
Keywords OTO EXPO; September 27-30, 2015; Dallas, Texas.

epistaxis, clinical care pathway, nasal packing, sphenopalatine Corresponding Author:

Peter S. Vosler, MD, PhD, Department of Otolaryngology–Head and Neck
artery ligation, embolization, costs, surgery
Surgery, University of Pittsburgh Medical Center, 200 Lothrop Street, Suite
500, Pittsburgh, PA 15213, USA.
Received March 7, 2016; revised May 19, 2016; accepted June 8, 2016. Email:
880 Otolaryngology–Head and Neck Surgery 155(5)

This study was approved by the University of Pittsburgh’s insti-
tutional review board under the protocol ENT/Otolaryngology
Clinical Patient Review and Predictive Assessment Project
(PRO13030372) and the Quality Improvement Committee

Retrospective Data
Retrospective review of epistaxis patients presenting to the
emergency department (ED) or inpatients at UPMC
Presbyterian and Montefiore hospitals was conducted between
October 2012 to December 2013. Two databases identified epis-
taxis patients: Crimson and Mars. Crimson is proprietary soft-
ware that collects patient information regarding diagnosis,
length of stay, and hospital cost, among many other data points.
Mars is a search engine that allows for keyword searches within
patient care records. ‘‘Epistaxis’’ was searched in patient charts
from October 2012 to December 2013, and the data were cross-
referenced with the patients found via the Crimson software.
Data collected included severe versus routine epistaxis, otolaryn-
gology consult, presence and type of packing, length of packing,
length of hospital stay for ED patients, location of hospital
admission, epistaxis as an inpatient or prompting presentation to
the ED, anticoagulation, ESPAL, embolization, and cost.
Exclusion criteria were prior skull base or nasal surgery and
epistaxis not managed by an otolaryngologist.

CCP Implementation
A multidisciplinary approach to epistaxis treatment was
designed involving otolaryngologists, emergency medicine
Figure 1. Severe epistaxis treatment algorithm. d/c, discharge;
physicians, and hospital administration. A treatment algo- ENT, ear nose throat/otolaryngology; OSH, outside hospital; SPA,
rithm was developed for severe epistaxis (Figure 1). sphenopalatine artery.
Emergency medicine physicians consulted the otolaryngol-
ogy resident on call for all severe epistaxis patients. The emer-
gency medicine physicians and hospital administrators involved
with inpatient placement admitted patients to the neurosurgery electrocautery. If the source was posterior, SPA ligation was
units at Presbyterian Hospital because it is the location of the performed. Anticoagulation was held in all patients taken to
operating rooms (ORs) and interventional radiology (IR) suites; the OR, and patients had reversal of their anticoagulation only
furthermore, the nurses on the neurosurgical floors where the if they were supratherapeutic.
combined otolaryngology-neurosurgery skull base patients are
admitted are more experienced with management of patients Prospective Data
with nasal packing. If there were no beds on the neurosurgical Data were prospectively recorded for all patients presenting to
floors, patients were placed on other inpatient floors in the ED at Presbyterian Hospital who were treated for epistaxis
Presbyterian Hospital. by an otolaryngologist from June 1, 2014, until May 31, 2015.
Otolaryngology residents were provided with an epistaxis The epistaxis CCP was implemented as described above with
management protocol to standardize epistaxis treatment otolaryngology consultation for all severe epistaxis patients,
(Figure 2). All patients were initially treated with topical admission to an appropriate epistaxis-management floor, and
agents and then examined with nasal endoscopy to evaluate ESPAL for appropriate patients. Data were cross-referenced
for bleeding sites. Otolaryngology attendings were notified with the 2 databases used in the retrospective series to ensure
about the algorithm to treat all severe epistaxis patients with capture of all epistaxis cases. The same data were collected in
ESPAL and about the proper admission location. Patients with the prospective and retrospective groups, and the same exclu-
severe epistaxis failing conservative management by residents sion criteria were used.
were evaluated by the attending and then taken to the OR at
the earliest availability. The packing was removed, and the Cost
nose was endoscopically examined. If bleeding sources other Per-patient cost data were available via a database through
than the SPA were identified, they were controlled with the Department of Finance managing UPMC Presbyterian-
Vosler et al 881

Figure 2. Protocol for epistaxis evaluation and treatment. AC, anticoagulation.

Montefiore. This database was started in April 2013, and it examine the established practice for epistaxis management
recorded individual costs of patient care. It did not evaluate prior to implementation of the severe epistaxis CCP. Two
reimbursement. For this study, cost data for individual databases yielded 170 patients treated for epistaxis, where 42
patients were populated into bins—including total supplies (24.7%) were found to have severe epistaxis. Most patients
(intervention), total drugs, blood, salaries and benefits, with severe epistaxis presented to the ED (61.9%), and 50%
depreciation, unit operating, and unit supporting cost—and were anticoagulated at time of presentation (Table 1).
expressed as an average based on treatment group. Treatment for severe epistaxis consisted of packing only
(57.1%), ESPAL (31.3%), and embolization (11.9%).
Statistics Average length of hospital stay in the retrospective group
Comparison of the pre-CCP-treated epistaxis patients in was 5.2 6 3.9 days, and the average number of days packed
regard to length of stay and cost of treatment was performed was 3.23 6 1.64 (Table 1). Eight patients were intubated
via the nonparametric Mann-Whitney U test. Comparison of for epistaxis in the pre-CCP group: 5 received packing-only,
hospital location was analyzed with a Pearson chi-square 2 had ESPAL, and 1 went to IR. All of the packing-only
test. Multivariate analysis of the groups based on treatment patients intubated were inpatients, and the ESPAL and IR
intervention and pre- and post-CCP was performed with the patients presented to the ED.
Kruskal-Wallis test. Significance was set as a \ 0.05.
Results A CCP for management of severe epistaxis with ESPAL and
limited duration of packing was initiated on June 1, 2014,
Pre-CCP through coordination with the otolaryngology residents on
Retrospective review of severe epistaxis management was call, the ED, and hospital administration. Prospective data
performed from September 2012 to November 2013 to yielded 118 patients treated for epistaxis, where 22 (18.6%)
882 Otolaryngology–Head and Neck Surgery 155(5)

Table 1. Clinical Characteristics and Outcome Data for Pre- and Post-CCP Implementation.a
Variable Pre-CCP Post-CCP Pb

Male sex 57.1 59.1 NS

Age, y, mean (range) 63.2 (38-94) 62.0 (23-85) NS
Total epistaxis, n 170 118 NS
Severe epistaxis 24.7 (42 of 170) 18.5 (22 of 118) NS
AC 50 (21 of 42) 40.9 (9 of 22) NS
ED presentation (severe) 61.9 (26 of 42) 81.8 (18 of 22) NS
Intubation for epistaxis 19 (8 of 42) 4.5 (1 of 22) NS
Failed embolization 20 (1 of 5) 0 NS
Failed OR 15.4 (2 of 13) 0 NS
Packing only 57.1 (24 of 42) 36.4 (8 of 22) .035
SPA ligation 31.0 (13 of 42) 54.5 (12 of 22) .009
IR 9.5 (4 of 42) 0 (0 of 22) NS
Days packed, mean 6 SD 3.2 6 1.6 1.4 6 1.6 .001
Hospital stay, d, mean 6 SD 5.2 6 3.9 2.1 6 1.3 \.001
Packing only and AC 54.2 (13 of 24) 50 (4 of 8) NS
SPA and AC 30.76 (4 of 13) 41.7 (5 of 12) NS
IR and AC 80 (4 of 5) 0 NS
Appropriate hospital location admission 42.3 (11 of 26) 83.3 (15 of 22) .007
Abbreviations: AC, anticoagulation; CCP, clinical care pathway; ED, emergency department; IR, interventional radiology; NS, not significant; OR, operating
room; SPA, sphenopalatine artery.
Values presented as % (n) unless noted otherwise.
a \ 0.05.

had severe epistaxis. The demographic information did not hospital floor with immediate access to essential epistaxis
differ between the retrospective and prospective epistaxis management resources (ie, IR suite and the OR). Distance
patients (Table 1). As in the retrospective data, most patients from critical resources and inappropriate hospital admission
with severe epistaxis presented to the ED (81.8%), and nearly location varied from 0.2 to 0.3 miles (distance measured
half (40.9%) were anticoagulated. with a pedometer). Following implementation of the CCP,
Implementation of the CCP resulted in decreased treat- admission of patients with severe epistaxis to an appropriate
ment with packing only (from 57.1% to 36.4%; P = .035), hospital floor increased to 83.3% (P = .007; Table 1).
increased ESPAL (from 31% to 54.5%; P = .009), and no
patients receiving embolization (P = .254; Table 1). Pre- versus Post-CCP Cost Data
Hospital stay decreased from 5.2 6 3.9 to 2.1 6 1.3 days Data for per-patient cost was initiated 8 months after initia-
(P = .001), and duration of packing decreased from 3.2 6 tion of the retrospective study, which limited the number of
1.6 to 1.4 6 1.6 days (P \ .001; Table 1). patients with severe epistaxis for analysis; patients with
Analysis of the data by CCP and intervention revealed no missing data were excluded. Given these limitations, 15
significant difference in hospital stay in the packing-only inter- patients in the pre-CCP cohort and 16 patients in the post-
vention; however, average days packed decreased following CCP were analyzed.
CCP implementation (from 3.7 6 1.3 to 2.3 6 2.0 days; P \ Patient cost data revealed a 66% decrease in total cost
.05) in the packing-only treatment group (Figure 3). There was after CCP implementation ($9435 saved; Table 2).
a decrease in days packed (2.2 6 1.6 vs 0.9 6 1.0 days; P = Comparison of packing only and ESPAL between the pre-
.004) and length of hospital stay (4.8 6 2.8 vs 1.8 6 1.3 days; and post-CCP cohorts demonstrated a 37% decrease in cost
P = .006) after CCP implementation in SPA-treated patients. ($2871 saved). Notably, the category that included the inter-
When stratified by treatment, there was no difference vention, total supplies, showed an 84% cost reduction
between the pre- and post-CCP groups with regard to the ($3748 saved). Further analysis revealed that all patients
percentage of anticoagulated patients (Table 1). There was with high total supply costs received embolization.
a higher percentage of patients who were anticoagulated in Decreased unit operating cost was the only measure that
the packing-only group versus the ESPAL intervention. reached significance (P = .047). All other costs were not
significant between the pre- and post-CCP cohorts.
Admission Location Stratification of the cost data by intervention and CCP
Retrospective data revealed that only 42.3% of patients implementation revealed no difference in cost between the
admitted from the ED for severe epistaxis were placed on a packing-only and ESPAL treatments.
Vosler et al 883

epistaxis that presents within a given time frame. It is

important to note that there was a decrease in cost when the
packing-only and ESPAL cohorts were examined, as these
patients likely had similar severity of epistaxis.

SPA Ligation
Our treatment algorithm is based on multiple reviews showing
that ESPAL has the highest control rate for severe epistaxis
management.1,4-6,15 It is a rapidly implemented intervention
resulting in decreased duration of packing, and we have shown
a decrease in cost using a modeling paradigm.9,10 Risk analysis
demonstrates implementing ESPAL as a low-risk intervention
versus embolization.16 Implementation of the CCP decreased
length of packing and hospital stay even when ESPAL-treated
Figure 3. Effect of clinical care pathway and treatment group on patients treated in the pre- and post-CCP groups were com-
number of days packed and duration of hospital stay. *P \.05. pared. It is recommended that all surgical candidates who have
CCP, clinical care pathway; IR, interventional radiology; SPA, sphe- posterior epistaxis, even those who are controlled with poster-
nopalatine artery. ior packing, undergo ESPAL.

Nasal Packing
Discussion Packing as the sole method of control is an acceptable inter-
vention, especially in patients with multiple medical comorbid-
Management of epistaxis is variable and dependent on a prac- ities where undergoing general anesthesia is contraindicated.
titioner’s individual training and preference.14 Implementing Unfortunately, the control rate with packing is only 52% to
a systematic approach to ensure airway protection and effi- 62%, and the level of discomfort is increased with it.6,7,17
cient control to minimize blood loss facilitates expeditious Patients are also packed for anywhere from 3 to 5 days, which
control of severe epistaxis. Establishing a CCP for treatment likely adds to the discomfort of the intervention and the
of severe epistaxis (1) provides a standardized and systematic increased cost with prolonged hospitalization.
approach to epistaxis management; (2) results in decreased The number of patients packed and packing duration
length of packing, hospital stay, and cost; and (3) facilitates decreased with implementation of the CCP. This effect was
admission to an appropriate hospital location. realized even within the packing-only treatment group, as
Resident experience is a potential confounding factor in patients who received packing had decreased packing dura-
the study. The majority of the residents who treated patients tion only after the CCP was established. This may be due to
with severe epistaxis were second-year residents with poten- the improved ability of residents to manage epistaxis, the
tially limited endoscopic skills. However, the resident expe- changing management paradigm, or the management by
rience was the same between the pre- and post-CCP fewer attendings.
cohorts, and patients receiving posterior packing all bene-
fited from epistaxis control in the OR. Intubation
A major limitation of this study is the use of historical
Multiple factors contributed to the discrepancy between the
controls, as factors other than our CCP could have influ-
pre- and post-CCP cohorts needing intubation, including the
enced the outcomes. The cost data are particularly suscepti-
stochastic nature of spontaneous epistaxis, the environmen-
ble to this bias, since factors such as inflation and changes
tal factors, and the treating physicians in the ED and inpati-
in vendors could influence the total cost and supply cost. As
ent units. Only 1 patient receiving embolization required
the retrospective cohort data were continuous with the pro-
intubation, so severity of bleed was not likely a contributing
spective cohort, it is unlikely that inflation substantially
factor to the decision for embolization.
influenced the results. Supply cost was different only when
the IR group was compared, as there was little difference Embolization
between the pre- and post-CCP groups when IR was
excluded (Table 2). Embolization has improved efficacy as compared with pack-
Difference in severity of epistaxis between groups could ing only, with 80% to 90% efficacy.18 Only 4 patients
also influence the results. There are gradations of epistaxis received embolization for severe epistaxis in this study, and
within the severe epistaxis cohort that could influence the they all were in the retrospective cohort. This is not to say
results. More patients were intubated in the retrospective that there is no role for embolization in the management of
cohort (19% vs 4.5%), which likely influenced treatment severe epistaxis. Rather, our results demonstrate that from a
decision and cost. Our study was designed to examine con- morbidity and cost perspective, embolization should be con-
secutive cohorts of patients before and after implementation sidered after packing and ESPAL if the proper resources for
of a CCP, and it is impossible to control the severity of surgical intervention are not available.
884 Otolaryngology–Head and Neck Surgery 155(5)

Table 2. Per-Patient Cost Data for Severe Epistaxis Management.a

Pre-CCP, $ Post-CCP, $ Decrease to Post-CCP, %

Packing SPA IR Total PO 1 SPA Packing SPA Total Pre-CCP Pre-CCP (PO 1 SPA)

Patients, n 6 5 4 15 11 3 13 16
Total supplies 249 1457 14,629 4486 798 263 848 738 –84 –7
Total drugs 519 496 1332 728 509 88 230 203 –72 –60
Blood 1062 486 934 836 800 95 74 78 –91 –90
Salaries and benefits 2504 3609 8064 4355 3006 2281 2088 2124 –51 –29
Depreciation 231 269 660 358 248 156 217 206 –42 –17
Unit operatingb 322 417 1310 617 365 253 190 202 –67 –45
Unit supportingc 1731 2507 5500 2995 2084 1628 1333 1388 –54 –33
Total cost 6617 9241 32,427 14,374 7810 4764 4980 4939 –66 –37

Abbreviations: CCP, clinical care pathway; IR, interventional radiology; PO, packing only; SPA, sphenopalatine artery ligation.
Values presented as average per-patient cost ($) unless noted otherwise.
Costs contained within the patient facing cost centers but not specifically attributable to patient encounters (ie, unit of admission).
Costs contained in nonpatient facing cost centers that are related to patient care (ie, environmental services, maintenance).

This pathway is not suitable for all institutions. Cost

Implementation of a similar pathway with ESPAL is dependent Implementation of our CCP resulted in a decreased per-
on available surgeons who can expeditiously perform the surgery patient cost average by 66%. Cost was similar in the
to ensure that there is no recurrent epistaxis and to decrease hos- packing-only and ESPAL in the pre-CCP group, and the
pital stay. If embolization is the only viable intervention at a decrease in cost was realized in both the packing-only and
given institution, then it should be utilized for patients refractory ESPAL treatments in the post-CCP group, with a 37%
to packing only. The presence of an interventional radiologist or decrease (Table 2).
an endovascular-trained surgeon is essential to minimize potential Embolization resulted in the highest per-patient cost
complications.19,20 A recent retrospective review of nearly 70,000 average; however, this was not significantly different from
patients revealed that 92.6% received surgical intervention for the other interventions (likely because the study was under-
epistaxis management, with only 6.4% receiving embolization.21 powered). The majority of the cost was in the supply cate-
This suggests that surgical capability is more widespread and is gory, followed by the salaries-and-benefits category—thus
another reason to encourage implementation of our CCP at other revealing that the major drivers of the increased cost were
institutions. the embolization materials used, the IR suite and equipment,
and the interventional radiologists, who are predominately
Admission Location neurosurgeons at our institution.
Patients admitted to the otolaryngology service were tradi- Analysis could not be made between pre- and post-CCP
tionally admitted to the otolaryngology inpatient floor or implementation, as no patients with severe epistaxis under-
another available floor located in Montefiore Hospital went embolization. Decreased cost with ESPAL versus
regardless of admission diagnosis. This a potential logistical embolization is consistent with a recent modeling paper
problem for management of epistaxis patients, as the dis- demonstrating that ESPAL is a more cost-effective interven-
tance to travel to critical resources varied between 0.2 and tion.11 Decreased cost with the CCP is likely from shorter
0.3 miles and included the use of 1 or 2 sets of elevators as hospital stay and earlier definitive intervention.
well as travel across bridges connecting hospitals that
spanned 3 city blocks. Although the nursing staff on the oto- Clinical Care Pathway
laryngology floor is familiar with the house staff and the Use of CCPs is improving patient care in multiple areas of
head and neck patients, they are not as familiar with the otolaryngology, including postoperative pathways for head
management of patients with nasal packing. and neck surgery with microvascular reconstruction,22 length
This situation is unique for our institution, but it demon- of stay,23 and postdischarge health care utilization among
strates how a CCP with coordination among the ED, hospital patients with head and neck cancer resection.24 Implementing
administrators, and the otolaryngology department can work a CCP for severe epistaxis management improved outcomes
together to improve patient care. Implementation of the CCP of hospital stay, cost, and packing duration in this study. It is
increased appropriate patient placement from less than half likely that other realms of otolaryngology would benefit from
(42.3%) to 83.3%. These patients were not randomly similar protocols.
placed—they were preferentially sent to neurosurgery floors, Barriers to implementation of our pathway included
where the nursing staff was familiar with nasal packing from proper dissemination of the protocol to the multidisciplinary
our skull base surgery service. teams involved (ED and hospital administrators). Resident
Vosler et al 885

turnover with the completion of training rotations required vig- 7. Nikolaou G, Holzmann D, Soyka MB. Discomfort and costs in
ilant surveillance and reminders of the protocol. Surgeon avail- epistaxis treatment. Eur Arch of Otorhinolaryngol. 2013;270:
ability was a barrier, as only 2 attending surgeons (E.W.W. 2239-2244.
and C.H.S.) at the hospital perform ESPAL. Availability of the 8. Douglas R, Wormald PJ. Update on epistaxis. Curr Opin in
ORs was also a barrier, as patients were placed on standby Otolaryngol Head Neck Surg. 2007;15:180-183.
and taken to the OR upon the earliest availability; however, 9. Dedhia RC, Desai SS, Smith KJ, et al. Cost-effectiveness of
this rarely resulted in delays .24 hours. endoscopic sphenopalatine artery ligation versus nasal packing
as first-line treatment for posterior epistaxis. Int Forum
Conclusion Allergy Rhinol. 2013;3:563-566.
Implementation of a CCP for management of severe epis- 10. Moshaver A, Harris JR, Liu R, et al. Early operative interven-
taxis using a multidisciplinary approach and development of tion versus conventional treatment in epistaxis: randomized
an ESPAL algorithm resulted in decreased hospital stay, prospective trial. J Otolaryngol. 2004;33:185-188.
days of packing, cost, and improved patient care with appro- 11. Rudmik L, Leung R. Cost-effectiveness analysis of endoscopic
priate hospital admission location. This pathway can be gen- sphenopalatine artery ligation vs arterial embolization for
eralized to improve epistaxis management at hospitals with intractable epistaxis. JAMA Otolaryngol Head Neck Surg.
adequate otolaryngology presence. 2014;140:802-808.
12. Evans-Lacko S, Jarrett M, McCrone P, et al. Facilitators and
Acknowledgments barriers to implementing clinical care pathways. BMC Health
We acknowledge the contribution of the statistical analysis provided Serv Res. 2010;10:182.
by Daniel G. Winger, MS, and Li Wang, MS, of the Clinical and 13. Jabbour M, Curran J, Scott SD, et al. Best strategies to implement
Translational Science Institute. We also thank Jon Petrie, the direc- clinical pathways in an emergency department setting: study protocol
tor of finance at UPMC Presbyterian Shadyside, and Carly Kyper,
for a cluster randomized controlled trial. Implement Sci. 2013;8:55.
senior financial analyst in the UPMC Presbyterian Shadyside
14. Fox R, Nash R, Liu ZW, et al. Epistaxis management: current
Finance Department, for providing the per-patient cost data.
understanding amongst junior doctors. J Laryngol Otol. 2016;
Author Contributions 15. Snyderman CH, Goldman SA, Carrau RL, et al. Endoscopic
Peter S. Vosler, design of study, acquisition, analysis, and inter- sphenopalatine artery ligation is an effective method of treat-
pretation of data, manuscript writing and revision; Jason I. Kass, ment for posterior epistaxis. Am J Rhinol. 1999;13:137-140.
conception and design of work, manuscript writing and revision; 16. Leung RM, Smith TL, Rudmik L. Developing a laddered algo-
Eric W. Wang, conception and design of work, manuscript writing rithm for the management of intractable epistaxis: a risk analy-
and revision; Carl H. Snyderman, conception and design of work,
sis. JAMA Otolaryngol Head Neck Surg. 2015;141:405-409.
manuscript writing and revision.
17. Schaitkin B, Strauss M, Houck JR. Epistaxis: medical versus
Disclosures surgical therapy: a comparison of efficacy, complications, and
Competing interests: None. economic considerations. Laryngoscope. 1987;97:1392-1396.
Sponsorships: None. 18. Christensen NP, Smith DS, Barnwell SL, et al. Arterial embo-
lization in the management of posterior epistaxis. Otolaryngol
Funding source: None.
Head Neck Surg. 2005;133:748-753.
19. Husted JW, Stock JR, Green WH, et al. Interventional radiol-
ogy in a community hospital. Radiology. 1982;143:273-274.
1. McClurg SW, Carrau R. Endoscopic management of posterior 20. O’Brien J, Baerlocher MO, Asch MR, et al. Limitations influ-
epistaxis: a review. Acta Otorhinolaryngol Ital. 2014;34:1-8. encing interventional radiology in Canada: results of a national
2. Small M, Murray JA, Maran AG. A study of patients with survey by the Canadian Interventional Radiology Association
epistaxis requiring admission to hospital. Health Bull. 1982; (CIRA). Cardiovasc Intervent Radiol. 2007;30:847-853.
40:20-29. 21. Brinjikji W, Kallmes DF, Cloft HJ. Trends in epistaxis emboli-
3. Supriya M, Shakeel M, Veitch D, et al. Epistaxis: prospective zation in the United States: a study of the Nationwide Inpatient
evaluation of bleeding site and its impact on patient outcome. Sample 2003-2010. J Vasc Interv Radiol. 2013;24:969-973.
J Laryngol Otol. 2010;124:744-749. 22. Dautremont JF, Rudmik LR, Yeung J, et al. Cost-effectiveness
4. Kumar S, Shetty A, Rockey J, et al. Contemporary surgical analysis of a postoperative clinical care pathway in head and
treatment of epistaxis: what is the evidence for sphenopalatine neck surgery with microvascular reconstruction. J Otolaryngol
artery ligation? Clin Otolaryngol Allied Sci. 2003;28:360-363. Head Neck Surg. 2013;42:59.
5. McDermott AM, O’Cathain E, Carey BW, et al. Sphenopalatine 23. Kagan SH, Chalian AA, Goldberg AN, et al. Impact of age on
artery ligation for epistaxis: factors influencing outcome and clinical care pathway length of stay after complex head and
impact of timing of surgery. Otolaryngol Head Neck Surg. neck resection. Head Neck. 2002;24:545-548.
2016; 154:547-552. 24. Dautremont JF, Rudmik LR, Nakoneshny SC, et al. Understanding
6. Soyka MB, Nikolaou G, Rufibach K, et al. On the effective- the impact of a clinical care pathway for major head and neck
ness of treatment options in epistaxis: an analysis of 678 inter- cancer resection on postdischarge healthcare utilization. Head
ventions. Rhinology. 2011;49:474-478. Neck. 2016;38(suppl 1):E1216-E1220.