ORIGINAL CONTRIBUTION

Outcomes of sphenopalatine and internal maxillary artery

ligation inside the pterygopalatine fossa for posterior
epistaxis*
Kristina Piastro1, Robert Scagnelli2, Neil Gildener-Leapman1, Rhinology 56; 2: 144-148, 2018
Carlos D. Pinheiro-Neto1 https://doi.org/10.4193/Rhin17.212

1
Division of Otolaryngology / Head and Neck Surgery, Department of Surgery, Albany Medical Center, Albany, New York, NY, USA *Received for publication:
2
Albany Medical College, Albany, New York , NY, USA October 17, 2017
Accepted: November 29, 2017

Abstract
Objective: Analysis of the efficacy of sphenopalatine artery (SPA) and internal maxillary artery (IMAX) ligation within the pterygo-
palatine fossa to control posterior epistaxis.

Methods: Demographic and clinical data were collected in sixty-two consecutive patients who had SPA/IMAX ligation surgery.
Clinical outcomes such as re-bleed rates and complications were acquired.

Results: A total of 62 patients were studied. Thirty-eight percent of patients had previously undergone silver nitrate nasal cautery
for epistaxis. Nine patients had undergone previous attempt of SPA procedure or embolization in other services. Two patients
returned to the operating room for anterior ethmoid ligation. There was one mortality within 30 days of surgery. Follow up ranged
from 3 months to 56 months (median= 28 months).

Conclusions: Dual SPA and IMAX ligation is effective in the control of difficult epistaxis cases, even in those patients with prior
surgical intervention.

Key words: epistaxis, sphenopalatine artery ligation, nosebleeds, pterygopalatine fossa

Introduction packing has been shown to fail to control posterior epistaxis in

Epistaxis is a common complaint. Approximately 60% of people up to 48% of patients (2). When conservative measures fail more
will experience an episode in their lifetimes. Epistaxis occurs in invasive means to obtain control of bleeding may be required.
a bimodal distribution, affecting predominantly males that are The bleeding site of a posterior epistaxis is located at Wood-
<10 years or >50 years old. Epistaxis has been identified more ruff’s plexus, over the posterior middle turbinate or along the
commonly in patients with other cardiovascular co­morbidities. posterior superior aspect of the septum or lateral nasal wall.
Roughly two-thirds of these cases will be spontaneous and most Branches of the sphenopalatine artery (SPA) are responsible
will self-resolve. However, approximately 6% of patients will for such epistaxis. SPA ligation is the most common surgical
require either medical or surgical interventions. An even smaller procedure performed for posterior epistaxis that is refractory to
percentage, 0.2% (1), will require hospitalization for control of nasal packing (3). Previously a Caldwell­Luc approach was prefer-
hemorrhage. red, however with the advancement of endoscopic techniques,
Location of nasal hemorrhage is divided into anterior and authors have been able to demonstrate decreased complication
posterior sources. The majority of bleeding is anterior in location rate and a shorter length of hospital stay (4).
and can be treated conservatively, either with cautery, anterior The terminal portion of the IMAX is located within the pterygo-
nasal packing or simple application of dorsal nasal pressure. palatine fossa. It gives off several branches in that location,
Posterior bleeds occur in approximately 5­to 10% of cases. Nasal including a terminal branch that passes through the sphe-

144
 Outcomes of SPA-IMAX artery ligation

nopalatine foramen along the lateral nasal wall and it is named

sphenopalatine artery. This artery can arborize into ten terminal
branches, and it is therefore important to either identify and
ligate all branches or have the possibility of persistent bleeding
(5, 6)
. In order to avoid the often-difficult process of identification
of the branching pattern of the SPA, especially in actively he-
morrhaging patients, we propose ligation of the artery laterally
within the pterygopalatine fossa.
Our objective is to analyze the success rates of SPA/IMAX liga-
tion within the pterygopalatine fossa to control severe posterior
epistaxis.

Materials and methods

A retrospective analysis of all consecutive patients undergoing
SPA/IMAX ligation between January 2013 and June 2017 perfor-
med by the senior author (CPN). Our Institutional Review Board
approved this retrospective chart review. Standard demogra-
phic data including: age, sex, anticoagulant use, severity of
bleed, and prior history of SPA ligation, nasal cautery, and/or
embolization procedures.
Selection criteria and indications for procedure:
1. History of severe posterior epistaxis requiring packing. Figure 1. (A) Orbital process of the palatine bone (dashed line). (B) The
2. Failure of standard in office techniques as well as conserva- first hemoclip ligates the sphenopalatine artery. The second and third
tive measures such as humidification. (arrow) hemoclips are placed inside the pterygopalatine fossa to ligate
3. Failure of anterior-posterior nasal packing. the internal maxillary artery.
Clinical efficacy and outcomes data were gleaned from posto-
perative patient chart review. Data was collected with attention
for postoperative complications; including dry eye and V2 careful medialization of the middle turbinate was performed
hypoesthesia, blood transfusion, return to the operating room followed by uncinectomy and wide maxillary antrostomy. Partial
for epistaxis, other management of recurrent epistaxis, and basic anterior and posterior ethmoidectomy were subsequently com-
laboratory data. pleted, opening the inferior portion of the posterior ethmoid
Long-term follow-up data was gathered by chart review as well and leaving the superior portion of the ethmoid intact. This
as phone interview. The total follow up time was defined as the is done to have a better control of the orbital process of the pala-
time from surgery to the date of the last clinic visit or phone tine bone during the opening of the pterygopalatine fossa. This
interview. In total 44 out of 62 patients were reached via telep- partial ethmoidectomy also opens a larger working space for
hone for supplementary interview. The remaining 18 patients’ control of epistaxis. The mucosa over the perpendicular plate of
follow up was based solely on information from the patient me- the palatine bone (posterior to the posterior fontanelle) was ele-
dical record. Six specific questions were assessed in all patients: vated. The dissection was progressed posteriorly until identifica-
whether there was an improvement in frequency and severity tion of the ethmoidal crest and the sphenopalatine foramen. The
of nosebleeds; if any nosebleeds happened since the surgery; SPA was identified. At that point, a 2mm Kerrison rongeur was
if any severe nosebleeds happened requiring urgent medical used to remove the orbital process of the palatine bone through
evaluation; if any nasal packing was placed after surgery; if any the sphenopalatine foramen. That step is performed carefully to
local cauterization was needed since surgery; and if any revision avoid injury to the artery. Part of the posterior wall of the maxil-
SPA ligation or embolization were required. lary sinus was also removed. The periosteum of the pterygo-
palatine fossa was exposed and carefully opened with a Cottle
Surgical technique dissector to expose the most proximal aspect of the artery. Three
After the patient was placed under general anesthesia with titanium-made hemoclips were used for the ligation. Karl Storz®
orotracheal intubation, a 0° endoscope was used to inspect (Tuttlingen, Germany) hemoclip applier was used. One hemoclip
both nasal cavities. In case of urgent surgery, the nasal packing was placed to ligate the SPA between the palatine bone and the
was removed only after all instrumentation was ready. Pledgets mucosal flap and the other two clips were placed to ligate the
soaked in oxymetazoline were placed in the nasal cavity. Then IMAX inside the pterygopalatine fossa. Finally oxidized cellulose

145
Piastro et al.

Table 1. Patient demographics, pre-operative clinical and laboratory mL ± 169.4).

tests. The clinical follow-up ranged from 3 months to 56 months (Me-
dian follow up is 28 months). Seven patients (11.3%) reported at
All Patients least one episode of mild nosebleeds after surgery. All patients
N = 62 who had an episode of anterior epistaxis after surgery had a
history of anticoagulant and/or antiplatelet use. Two of those
Age 18 - 96 yo (median = 60.5)
patients (3.2%) who had mild epistaxis required placement
Sex 65% male
of anterior absorbable packing. One patient reported office
BMI 29.2 ± 5.6 cauterization with silver nitrate 6 months after surgery. Two
Estimated Blood Loss (mL) 161.7± 169.4 patients required anterior ethmoid artery ligation in a separate
setting for recurrent anterior epistaxis. None of the 62 patients
Packings prior to surgery 1.82 ± 2.5
required revision SPA/IMAX ligation. One patient presented to
Previous cauterization 23/62 (37,1%)
the emergency room within 30 days of surgery with bilateral
Previous SPA ligation/Emboli- posterior nasal bleeding after having left side SPA/IMAX ligation.
9/62 (14,5%)
zation
This patient had bilateral anterior-posterior packing placed with
Hypertension 30/62 (48,4%)
a subsequent cardiac arrest and death. There were no reports of
Anticoagulant Use 42/62 (67,7%) V2 hypoesthesia or dry eye in our patient population.
Pre-op Hemoglobin 10.6 ± 2.77
Discussion
Pre-op Hematocrit 32.2 ± 8.04
The management of epistaxis has evolved in recent years. Endo-
Pre-op Platelets 225 ± 63 scopic approaches have greatly added to the armamentarium
BMI = Body mass index; SPA = Sphenopalatine artery; Pre-op = Pre- of control of nosebleeds. Endoscopic SPA ligation has become
operative the mainstay of treatment for posterior epistaxis that is refrac-
tory to packing or medical management (7). The SPA anatomy is
varied and the surgeon must be aware of its branching pat-
was used to cover the pterygopalatine fossa and clips (Figure 1). tern (5). 14.5% of our patients had previously undergone SPA
cautery, ligation or embolization procedures performed in other
Results institutions. The published rate of failure for SPA cauterization or
Between January 2013 and June 2017 a total of 62 patients ligation varies from 2% to 13% (8-11). Anatomic variations of the
underwent SPA/IMAX ligation, with 12 patients undergoing SPA branches and foramen location may contribute to inade-
concomitant septoplasty. The demographic data as follows is quate epistaxis control. We observed overall 95.2% control of
summarized in Table 1. Forty-one patients (66%) underwent the epistaxis with ligation of the SPA associated with ligation of the
procedure acutely as an inpatient and 21 patients (34%) had IMAX inside the pterygopalatine fossa. Importantly there were
planned outpatient surgery after discharge from the hospital. no complications specific to the approach such as dry eye or V2
Forty patients (65%) were male. Patient age varied from 18­to 96 hypoesthesia.
years old, with an average age of 61 years (median=60.5). Anti- Bleeding from the anterior or posterior ethmoidal arteries can
coagulant and/or antiplatelet use was present in 67.7% of the also be a cause of epistaxis after SPA ligation (9, 12). From our 9
patient population, and hypertension was identified on patient patients (14.5%) who had undergone prior surgical procedures
presentation in 48.4%. to address the SPA in other services, all of them presented with
Recorded emergent visits for epistaxis were available for 58 posterior bleeding, which indicates a local failure to control the
patients with an average of 2.55 ± 2.71 visits. The number of SPA. None of them presented with bleeding from the anterior
previous interventions with anterior-posterior nasal packing per ethmoidal artery area. For our SPA-IMAX ligation series two
patient was 1.82 ± 2.53 packings. patients presented with severe anterior bleeding months after
Preoperative lab work was available for 53 patients. Preoperative initial surgery, these patients were revised with anterior ethmoid
hemoglobin was 10.6 ± 2.77, preoperative hematocrit was 32.2 artery ligation. During these anterior ethmoid artery ligation
± 8.04, and preoperative platelets were 225 ± 63. surgeries, no bleeding was present from the SPA area and no
Thirty-seven percent of patients had previously undergone intervention was performed in that region.
silver nitrate nasal cautery for epistaxis. 14.5% of patients had One patient with severe comorbidities required bilateral
undergone previous attempt at SPA procedure or emboliza- anterior-posterior packing within 30 days after having left sided
tion in other services. Intra-operative estimated blood loss was SPA-IMAX ligation. He had a cardiac arrest immediately after the
available for 50 of the patients and was highly variable (161.7 packing was placed and subsequent death. It was not possible

146
 Outcomes of SPA-IMAX artery ligation

to identify the exact location of the bleeding since the patient using conventional ligation of the SPA branches have reported
was having bilateral hemorrhage. the need for blood transfusion ranging from 45-81% of patients
The age range in this population was 18-96 years old. A poste- (9, 14)
. It can be argued that SPA-IMAX ligation limits blood loss
rior bleed in a patient as young as 18 is rare. This patient initially and perioperative complications such as blood transfusion. In
presented with a posterior bleed on post-operative day eight of our patient population, there were no immediate perioperative
a functional endoscopic sinus surgery. complications requiring additional interventions.
All patients had at least two severe nosebleeds prior to surgery We understand that comprehensive dissection and ligation of all
with need for anterior-posterior packing or one severe bleed branches of the sphenopalatine artery may have similar results.
and failure of the anterior-posterior packing. The nasal packing However, with the ligation of the artery inside the pterygopala-
average prior to the surgery in our patient population was 1.8 tine fossa, all branches are treated including branches that do
packs. Pre-operative hemoglobin levels (10.6 ± 2.77) and hema- not pass through the SPA foramen like the pharyngeal branch
tocrit levels (32.2 ± 8.04) were both below normal in our patient that passes through the palatovaginal canal and the descending
population, likely a result of the recurrent posterior bleeding palatine artery decreasing the overall blood flow in the area. In
prior to surgery. some cases, there are small arterial branches that pass to the
Seven patients (11.3%) reported at least one episode of mild nasal cavity through small bone canals other than the SPA fora-
nosebleeds after surgery on clinical follow-up. Anteroseptal men. Apparently the SPA/IMAX technique “devascularizes” more
location was evident on exam and this was treated with simple arterial branches to the nasal cavity compared to the isolated
silver nitrate anterior cautery or absorbable packing placed in SPA ligation and avoids the search for different branches due to
contact to the anterior septum with resolution of symptoms. extensive anatomical variation in this area. Due to limitations
The seven patients who experienced anterior nosebleeds were in this single arm retrospective case series, we are unable to
on an anticoagulant or antiplatelet medicine. It is important to demonstrate superiority to the standard SPA ligation approach.
be vigilant for anterior bleeding as other arteries like the facial And it may be argued that the standard SPA ligation approach
artery and anterior ethmoidal artery have an important role is technically less challenging. However, in our experience, this
in blood supply to the anterior septum (13). Importantly, in our procedure can be taught to trainees after a 3-month rotation in
series, we appear to have very close to absolute control of the rhinology. We believe that for referral centers and experienced
sphenopalatine artery supplied mucosa through an SPA-IMAX surgeons, this combined approach seems to provide a high rate
ligation approach. From the 3 patients who had severe nose- of local control of epistaxis in the sphenopalatine artery sup-
bleeds after surgery and required intervention (surgery and/or plied mucosa, even in cases of prior failure.
anterior-posterior packing), 2 of them had no bleeding from the
SPA area previously ligated. The bleeding was exclusively from Conclusion
the anterior ethmoidal area and controlled with specific ligation. A dual SPA-IMAX approach seems to provide near complete
Since the source of bleeding was not possible to be identified in control over the sphenopalatine supplied region. Especially in
one patient who had unilateral SPA-IMAX ligation, this proce- anticoagulated patients, a concomitant anterior epistaxis source
dure showed at least 98.4% bleeding control at the SPA supplied should be sought.
mucosa. None of the patients required revisional SPA ligation
in our series. These patients should be closely evaluated for Authorship contribution
the potential for anterior sources of epistaxis prior to SPA-IMAX KP: Data collection and interpretation, manuscript writing/edi-
ligation, so that a concurrent anterior ethmoid artery or greater ting; RS: Data collection and interpretation, manuscript writing/
palatine artery ligation may be considered (13). One potential editing; NGL: Manuscript writing/editing. CPN: Principal Inves-
concern with this procedure is that the vessels were only clip- tigator and corresponding author, Primary surgeon for patients
ped, not cauterized or sectioned. There may be concern that this investigated, Manuscript writing/editing.
could lead to revascularization. However, we have not seen any
cases of rebleed from a ligated vessel. Conflict of interest
Estimated blood loss was highly variable (161.7 mL ± 169.4). We have no conflicts of interest to disclose.
However, no patients required blood transfusion. Other studies

References Dec;45(6):1423-49. 3. Sacks R, Sacks PL, Chandra R. Chapter 3:

1. J i n d a l G , G e m m e t e J , G a n d h i D. 2. Christensen NP, Smith DS, Barnwell SL, Wax Epistaxis. Am J Rhinol Allergy. 2013 May-
Interventional neuroradiology applica- MK. Arterial embolization in the manage- Jun;27 Suppl 1:S9-10.
tions in otolaryngology, head and neck ment of posterior epistaxis. Otolaryngol 4. Nouraei SA, Maani T, Hajioff D, Saleh HA,
surgery. Otolaryngol Clin North Am. 2012 Head Neck Surg. 2005 Nov;133(5):748-53. Mackay IS. Outcome of endoscopic sphe-

147
Piastro et al.

nopalatine artery occlusion for intrac- sphenopalatine artery for severe posterior May;126(5):1033-8.
table epistaxis: a 10-year experience. epistaxis. Otolaryngol Head Neck Surg. 2001 14. Barlow DW, Deleyiannis FWB, Pinczower
Laryngoscope. 2007 Aug;117(8):1452-6. Apr;124(4):464-7. EF. Effectiveness of surgical manage-
5. Simmen DB, Raghavan U, Briner HR, 10. Abdelk ader M, Leong SC, White PS. ment of epistaxis at a tertiary care center.
Manestar M, Groscurth P, Jones NS. The Endoscopic control of the sphenopalatine Laryngoscope. 1997 Jan;107:21-4.
anatomy of the sphenopalatine artery for artery for epistaxis: long-term results. J
the endoscopic sinus surgeon. Am J Rhinol. Laryngol Otol. 2007 Aug;121(8):759-62.
2006 Sep-Oct;20(5):502-5. 11. McDermott AM, O'Cathain E, Carey BW,
Carlos D. Pinheiro-Neto MD PhD
6. Budrovich R, Saetti R. Microscopic and O'Sullivan P, Sheahan P. Sphenopalatine
endoscopic ligature of the sphenopalatine Ar ter y Ligation for Epistaxis: Factors Division of Otolaryngology
artery. Laryngoscope. 1992 Dec;102(12 Pt Influencing Outcome and Impact of Timing Head and Neck Surgery
1):1391-4. of Surgery. Otolaryngol Head Neck Surg.
Albany Medical Center
7. McClurg SW, Carrau R. Endoscopic manage- 2016 Mar;154(3):547-52.
ment of posterior epistaxis: a review. Acta 12. Felippu A, Mora R, Guastini L. Endoscopic 47 New Scotland Avenue, MC-41
Otorhinolaryngol Ital. 2014 Feb;34(1):1-8 transnasal cauterization of the anterior Albany, NY, 12208
8. Kumar S, Shetty A, Rockey J, Nilssen E. ethmoidal artery. Acta Otolaryngol. 2011
Contemporar y surgical treatment of Oct;131(10):1074-8.
USA
epistaxis. What is the evidence for spheno- 13. Butrymowicz A, Weisstuch A, Zhao A,
palatine artery ligation? Clin Otolaryngol Agarwal J, Pinheiro-Neto CD. Endoscopic Tel: +1-518-262 5575
Allied Sci. 2003 Aug;28(4):360-3. endonasal greater palatine artery cauteri-
9. Voegels RL, Thomé DC, Iturralde PP, zation at the incisive foramen for control E-mail: pinheic@amc.edu
Butugan O. Endoscopic ligature of the of anterior epistaxis. Laryngoscope. 2016

148