Veterinary Surgery

37:781–785, 2008

Acute Oropharyngeal and Esophageal Stick Injury in

Forty-One Dogs

IVAN P. DORAN, BVSc Cert SAS, CAROL A. WRIGHT, BVSc Cert VR Cert SAS MRCVS, and ALASDAIR HOTSTON
MOORE, MA VetMB Cert SAC Cert VR Cert SAS

Objective—To report clinical findings, treatment, and outcome in dogs with acute (o7 days)
oropharyngeal or esophageal stick injury.
Study Design—Retrospective study.
Animals—Dogs (n ¼ 41) with acute oropharyngeal or esophageal injury.
Methods—Dogs had clinical and radiographic examination, and frequently, cervical surgical ex-
ploration. The decision to operate was based on radiographic findings of cervical emphysema.
Outcome was determined by owner or veterinarian interview.
Results—Of 41 dogs, 27 had oropharyngeal injury and 14 had esophageal injury. Five dogs with
esophageal injury died. All dogs with radiographic evidence of cervical emphysema (n ¼ 34) had
ventral median cervical exploration or necropsy; 11 had wood fragment(s) retrieved. In 7 dogs
without radiographic signs of cervical emphysema, wounds involving the pharynx or soft palate
were treated by local debridement and lavage using an oral approach. Mean follow-up time was
36.4 months. All wounds healed without complication; however, 1 dog that was not surgically
explored had a piece of wood surgically retrieved 3 months later.
Conclusions—Radiographic evidence of cervical emphysema is a frequent finding in dogs with acute
penetrating oropharyngeal or esophageal injury and indicates trauma to the deeper cervical tissues.
Acute penetrating injury of the oropharyngeal region, when treated appropriately, has a better
prognosis than acute esophageal penetration.
Clinical Relevance—Ventral median cervical surgical exploration is recommended in dogs with
acute penetrating injury of the oropharynx or esophagus if there is radiographic evidence of tissue
emphysema.
r Copyright 2008 by The American College of Veterinary Surgeons

INTRODUCTION and discharging sinus tracts of the head and neck.1,2

Reportedly, esophageal penetration has an extremely
guarded prognosis for recovery1,3; and although unclear
O ROPHARYNGEAL INJURY associated with
carrying, chewing, or retrieving sticks occurs in
dogs.1,2 Clinical signs associated with acute (o7 days)
why these dogs are at particular risk, a progressive cell-
ulitis associated with bacterial inoculation of traumatized
and chronic injury differ. When the penetrating injury is periesophageal tissues and subsequent airway compro-
acute, dogs typically have signs associated with oral pain, mise, mediastinitis, pneumomediastinum, pneumothorax,
and drool a mixture of saliva and blood. Without treat- and septic shock has been proposed.1 Wood fragment(s)
ment, depressed behavior, loss of appetite, swelling of the may or may not be identified in the traumatized tissues.
cervical region, dyspnea, and shock may occur.1 Chron- In 2 reports of oropharyngeal injury associated with
ically affected dogs typically manifest abscess formation sticks,1,2 only 21 of 115 dogs were admitted with acute

From the Small Animal Hospital, School of Clinical Veterinary Science, University of Bristol, Langford, Bristol, UK; and Vale Vets
Animal Hospital, Dursley, UK.
Address reprint requests to Ivan P. Doran, BVSc Cert SAS, Small Animal Hospital, School of Clinical Veterinary Science, University
of Bristol, Langford, Bristol BS40 5DU, UK. E-mail: [email protected].
Submitted December 2007; Accepted March 2008
r Copyright 2008 by The American College of Veterinary Surgeons

0161-3499/08
doi:10.1111/j.1532-950X.2008.00448.x
781
782 ACUTE OROPHARYNGEAL AND ESOPHAGEAL STICK INJURY IN DOGS

injury. The incidence of radiographic emphysema in these withdrawn. Insufflation was not used. Right lateral radio-
acute cases and the relationship between radiographic graphs of the pharynx, neck, and thorax were obtained.
emphysema and surgical findings have not been reported. Radiographic signs of cervical emphysema were considered
We report clinical findings and outcome in 41 dogs an indication for surgical exploration of the neck. Emphysema
with acute oropharyngeal or esophageal injury associated was identified by the presence of gas lucencies within the cer-
vical soft tissues, outside of the lumen of the esophagus or
with a history of carrying, chasing, or chewing a stick.
trachea. In the absence of abnormal radiographic signs, iden-
tified oropharyngeal wounds were treated locally through the
MATERIALS AND METHODS mouth.

Inclusion Criteria Surgical Approaches

Medical records (1997–2007) were reviewed for dogs that
Surgical Exploration of the Neck Region. With the anes-
had acute (o7 days) injury of the oropharynx or esophagus
thetized dog positioned in dorsal recumbency, a skin incision
associated with a history of carrying, chasing, or chewing a
was made from the cricoid cartilage extending caudad for two-
stick followed by acute onset apparent distress. Only dogs
thirds of the distance to the manubrium to expose the paired
with complete medical records including a radiographic report
sternohyoideus muscles, which were separated and retracted.
were included. Radiographs were reported by a board-
The trachea and esophagus were retracted as needed and fas-
certified radiologist. Penetrating injury was defined as perfo-
cial tissues bluntly dissected to facilitate examination of the
ration of the pharyngeal or esophageal wall and the presence
cervical soft tissues, including the retropharyngeal region.
of a tract of traumatized cervical soft tissue. This was con-
Foreign material was removed and suction of accumulated
firmed by ventral median cervical surgical exploration or by
fluid was performed. Copious lavage with lactated Ringer’s
necropsy in all dogs with radiographic cervical emphysema.
solution and further suction was then conducted. Tears in the
Injury to the cervical tissues was identified at surgery by cer-
pharyngeal or esophageal wall were minimally debrided with
vical tissue swelling, foreign material, necrotic tissue, and the
Metzenbaum scissors and then closed using 2 metric
local accumulation of fluid.
polydioxanone suture in 1 or 2 layers in a simple interrupted
pattern. An active suction drain (Jackson Pratt 100 mL, Car-
Supportive Treatment dinal Health, Swindon, UK) was routinely placed into the
cervical soft tissues, dorsal to the sternothyroideus muscles
Intravenous (IV) fluids, typically lactated Ringer’s solu- which were apposed using 2 or 3 metric polyglecaprone 25 in a
tion, were administered for circulatory support, beginning simple continuous pattern. The skin incision was closed using
preoperatively, and opiates were administered for analgesia. 2 metric monofilament nylon in a Ford interlocking pattern.
Opiate choice (morphine sulfate, methadone hydrochloride, In all dogs that had cervical exploratory surgery, a 20 Fr
fentanyl citrate or in combination) and dose were variable percutaneous endoscopically placed gastrostomy (PEG) tube
depending on clinician preference. Either meloxicam (0.2 mg/ was placed (Percutaneous endoscopic gastrostomy kit, Mila
kg IV) or carprofen (4 mg/kg IV) was typically administered International Inc., Erlanger, KY) or, in dogs with esophageal
if the dog was not hypotensive and there was no history of trauma, a 20 Fr gastrostomy tube (Veterinary feeding tube,
corticosteroid administration or renal disease. IV amoxicillin Smiths Medical Pm Inc., Waukesha, WI) was placed via left
clavulanate (20 mg/kg every 6 hours) and IV metronidazole flank laparotomy.
(10 mg/kg twice daily) were administered. Oral Approach. In dogs with no abnormal radiographic
signs, pharyngeal wounds were debrided, lavaged, and sutured
through the mouth. Wounds of the soft palate were lavaged
Oropharyngeal and Radiographic Examination with lactated Ringer’s solution but not sutured. Subsequent
After induction of anesthesia with IV propofol (4 mg/kg), supportive care was as for dogs undergoing median cervical
and endotracheal intubation, inspection of the oropharynx exploration.
was performed using 2 laryngoscopes and a flexible gastro-
scope. Both sides of the tongue and the sublingual area were Aftercare
inspected by drawing the tongue laterally and then dorsally.
The laryngoscopes were used to illuminate and, by probing Ongoing analgesia was provided using intramuscular in-
with the laryngoscope blades, to displace tissues and inspect jections of morphine sulfate (0.3 mg/kg every 4 hours) or
the tonsillar, oral palatal surfaces, pharyngeal, and perilaryn- methadone hydrochloride (0.3 mg/kg every 4 hours) in com-
geal mucosa to the esophageal aditus. The gastroscope was bination with meloxicam (0.1 mg/kg IV once daily) or car-
used to complete the examination of the pharyngeal wall ob- profen (4 mg/kg IV once daily). Dogs were fed a prescription
scured by the soft palate. Digital palpation and probing with convalescent diet (Prescription Diet a/d, Hills Pet Nutrition,
the laryngoscope blades were also used to identify puncture Topeka, KS) starting on the 1st postoperative day. Food was
wounds. The esophagus was examined by passing the gastro- given by the gastrostomy tube for surgically explored dogs, or
scope into the esophagus as far as the cardia and then in- orally for dogs that had received local wound treatment
specting the esophageal wall as the gastroscope was through the mouth. Active suction drain chambers were emp-
 DORAN, WRIGHT, AND MOORE 783

tied when full and an ongoing record of the volume of fluid Table 1. Distribution of Sites of Oropharyngeal Penetration
retrieved was made. Active suction drains were removed when Present Griffiths et al
the daily rate of fluid recovery fell below 0.5 mL/kg body- Site of Injury Study (n ¼ 41) (n ¼ 17)
weight. Dogs with gastrostomy tubes were fed orally on the
10th postoperative day and, if no sign of dysphagia or regur- Esophageal 14 (34%) —
gitation ensued, the gastrostomy tubes were removed, by Lateral pharyngeal wall 13 (32%) 4 (24%)
Sublingual 5 (12%) 7 (41%)
traction, the following day. In cases of esophageal trauma,
Rostral pharyngeal wall 3 (7%) 2 (12%)
esophagoscopy was repeated before feeding tube removal.
Soft palate 3 (7%) —
Follow-up radiography was performed on the day after Tonsillar 3 (7%) 3 (18%)
surgery if dogs had pneumomediastinum or pneumothorax or
if there was concern regarding a dog’s clinical progress. This Six cases of esophageal penetration also had dorsal pharyngeal
was to check for radiographic signs of developing mediastini- trauma.
tis, such as mediastinal widening (dorsoventral view), a wors-
ening severity of pneumomediastinum/pneumothorax or the tube). Of these 10 dogs, 3 died (all had esophageal pen-
accumulation of pleural fluid. Dogs were discharged when etrations) and 7 survived (4 esophageal, 3 pharyngeal). Of
able and willing to eat (canned food), when active suction the 34 dogs with radiographic emphysema, the mortality
drains had been removed, and when considered pain free.
rate between dogs with purely cervical emphysema and
Owners were instructed to present their dogs to their own
veterinarian 2 days after discharge, and subsequently for skin dogs that had also pneumothorax or pneumomediasti-
suture removal as needed. num was not significantly different (P4.05, Table 3).

Outcome Surgical Findings

Follow-up data were obtained via telephone interview, Thirty-two dogs with cervical emphysema had surgical
both with dogs’ owners and with the referring veterinarians. exploration via a ventral median cervical approach and,
Information regarding dysphagia, regurgitation, nasal dis- in 11 dogs, a piece of wood was recovered. All dogs that
charge, surgical wound complications, and discharging sinus
had cervical exploration had a bluntly traumatized tract
formation was sought.
of soft tissue evident in the neck. Two further dogs, with
radiographic evidence of emphysema, died before surgery
RESULTS could be performed. These 2 dogs had a necropsy and
stick penetration into the cervical soft tissues was con-
Forty-one dogs (24 females, 17 males; mean age, 4.8
firmed.
years [range, 5 months to 13 years]; mean weight, 23.1 kg
Seven dogs (17%) had no abnormal radiographic
[range, 8–55 kg]) met the inclusion criteria. Breeds were
findings and did not have cervical exploratory surgery.
Border Collie (n ¼ 17), crossbreed (n ¼ 10), English
These 7 dogs included 3 dogs that had a single rostral
Springer spaniel (n ¼ 4), Labrador Retriever (n ¼ 3),
pharyngeal wound, 3 dogs that had a single soft palate
and 1 each of German Shepherd, Lurcher, Mastiff, We-
wound, and 1 dog with a single wound to its left lateral
imaraner, Jack Russell terrier, Rottweiler, and Boxer.
pharyngeal wall. The 3 dogs with rostral pharyngeal
Each dog had a history of carrying, chasing, or chewing a
wounds had surgery to debride, lavage, and suture their
stick followed by an acute onset of vocalization, gagging
wounds via an oral approach. The 3 dogs with soft palate
and pawing at the mouth.
wounds and 1 dog with a left lateral pharyngeal wall
Clinical signs on admission were salivary drooling
wound had only wound lavage with lactated Ringer’s
(n ¼ 35), depression (n ¼ 29), pain on neck flexion (n ¼
solution. Three dogs had substantial hemorrhage from
24), subcutaneous emphysema in the cervical region
the oropharyngeal cavity on admission and 1 dog re-
(n ¼ 22), pain elicited on opening the mouth (n ¼ 16),
quired blood transfusion. All of these dogs had trauma to
blood in the saliva (n ¼ 9), and collapse (n ¼ 3). Lesions
were oropharyngeal (27 dogs) and esophageal (14 dogs;
Table 2. Mortality Rates for Dogs with Oropharyngeal or Esophageal
Table 1). Dogs with esophageal penetration (n ¼ 14) were Penetration
significantly more likely to die than those with pharyn-
Number of Dogs with Number of Dogs with
geal penetration (n ¼ 24) (Po.01, Table 2). Esophageal Penetration Pharyngeal Penetration
Radiographic records were available for all dogs.
Thirty-four dogs (83%) had radiographic evidence of Survived after injury 9 24
Died after injury 5 0
cervical emphysema. Ten of these dogs also had pneumo-
thorax or pneumomediastinum (in the 2 dogs with Fisher’s exact test (Po.01).
pneumothorax, this was unilateral and not considered Three dogs had soft palate injury only and are not included in this
severe enough to warrant placement of a thoracostomy table.
784 ACUTE OROPHARYNGEAL AND ESOPHAGEAL STICK INJURY IN DOGS

Table 3. Mortality Among Dogs with Radiographic Signs of or esophageal penetration. Furthermore, all dogs that
Emphysema had radiographic cervical emphysema were confirmed, by
Number of Dogs with Number of Dogs with surgical exploration or necropsy, as having a tract of
Emphysema Limited to Pneumothorax or bluntly traumatized soft tissue within the cervical soft
the Cervical Region Pneumomediastinum tissues; although a wood fragment(s) was only retrieved
Survived after 22 (5 esophageal, 17 7 (4 esophageal, 3 in 11 dogs. Early cervical surgical exploration has been
injury pharyngeal) pharyngeal) recommended for oropharyngeal stick injury when in-
Died after injury 2 (both esophageal) 3 (all 3 esophageal) volvement of the deeper tissues of the neck is suspected.1,2
We followed that recommendation in dogs with radio-
Fisher’s exact test (P ¼ .138).
All of these dogs also had cervical emphysema. graphic evidence of cervical emphysema. Based on
our experience, dogs with acute penetrating pharyngeal
injuries have a significantly higher likelihood of survival
a tonsillar crypt and all had radiographic signs of cervical than those with acute penetrating esophageal injuries.
emphysema, with traumatized cervical tissue subse- The ventral median cervical surgical approach
quently confirmed during median cervical exploration. provides good access to the retropharyngeal and cervi-
cal tissues. We feel that this approach is appropriate in all
Outcome but the most rostral of pharyngeal stick penetrations,
where an oral approach may need to be adopted because
Five dogs died, 2 before surgical exploration of the of the difficulties in accessing the rostral pharyngeal wall
neck could be performed and 3 after cervical surgery. via a ventral cervical approach. The benefits of surgical
Each of the 5 dogs had signs consistent with mediastinitis. exploration include removal of any foreign material, re-
These signs included rectal temperature above 39.51C pair of pharyngeal or esophageal tears, and lavage of the
(n ¼ 4), dyspnea (n ¼ 5), heart rate 4150 beats/min (n ¼ traumatized tissues to reduce microbial contamination.
5), neutropenia (n ¼ 5) (ref range 4.0–16.0
103 cells/mL), We administered a combination of amoxicillin clavu-
mediastinal widening on dorsoventral thoracic radiogra- lanate and metronidazole for antimicrobial prophylaxis.
phy (n ¼ 5), and radiographic pleural fluid (n ¼ 2). Staphylococcus aureus, Streptococcus spp., and Escheri-
Of 36 surviving dogs, 31 had follow-up (mean, 36.4 chia coli are among the most common commensal
months; range, 4–108 months). All surgical wounds bacteria in the canine oropharyx.4 Clostridium spp. are
healed uneventfully and none of these dogs developed common soil-borne bacteria that can cause wound infec-
discharging sinus tracts or abscesses. The dog that had tions, particularly in the presence of traumatized tissue.5
a single wound to its left lateral pharyngeal wall but no Amoxicillin-clavulanate has a broad spectrum of activity
radiographic evidence of cervical emphysema, and that and is active against gram positive bacterial species such
was treated with simple local wound lavage, had persis- as S. aureus and Streptococcus spp.5 Metronidazole is
tent pain on cervical palpation and had a piece of wood active against anaerobic bacteria such as Clostridium
surgically retrieved from the cervical soft tissues 3 months spp.5 E. coli isolates have considerable variation in
later; no further problem occurred. Soft palate wounds antibiotic susceptibility, due in part to plasmid-mediated
identified in 3 dogs were reported, after subsequent spread of resistance factors.5 Many E. coli isolates are
examination by the dogs’ referring veterinarians, to have susceptible to amoxicillin-clavulanate, however many are
healed uneventfully. One dog, whcih had had a penetrat- not and the addition of a fluoroquinolone to the antibi-
ing wound to its proximal esophagus, experienced inter- otic regime may be prudent.5
mittent regurgitation of food immediately after eating, In the human literature, there are descriptions of a
but this resolved after 3 months in response to dietary condition termed descending cervical mediastinitis
modification (softer food). Two dogs, which had both (DCM) where bacteria gain passage through the pharyn-
had lateral pharyngeal wall penetrating injury and me- geal or esophageal wall and spread through the cervical
dian cervical exploration, gagged during swallowing un- tissues to the mediastinum.6 Until recently, this syndrome
less limited to modest-sized morsels of food. Another dog had a 40–50% mortality rate and is now treated aggres-
in the surgically explored group, which had sustained sively to expose, debride, and pack open the cervical and
pharyngeal penetration, adjacent to the larynx, remained cranial mediastinal tissues.6,7 Seventy percent of all cases
dysphonic 15 months after initial presentation. of DCM spread via one particular fascial corridor and
the reason for this is unknown.6,7
DISCUSSION Magnetic resonance imaging (MRI) may have pro-
vided greater detail regarding the extent and location
Radiographic evidence of cervical emphysema is a of traumatized structures and the presence of foreign
frequent clinical finding in dogs with acute oropharyngeal material in these dogs. The absence of these imaging
 DORAN, WRIGHT, AND MOORE 785

modalities, particularly in dogs without radiographic cer- 2. Griffiths LG, Tiruneh R, Sullivan M, et al: Oropharyngeal
vical emphysema, was a limitation of our study. Future penetrating injuries in 50 dogs: a retrospective study. Vet
studies using MRI to investigate location of acute pen- Surg 29:383–388, 2000
etrating injury may help our understanding of why dogs 3. Parker NR, Walter PA, Gay J: Diagnosis and surgical man-
with esophageal penetration have a significantly higher agement of esophageal perforation. J Am Anim Hosp As-
soc 25:587–594, 1989
mortality than dogs with pharyngeal penetration. Based
4. Clapper WE, Meade GH: Normal flora of the nose, throat
on our study population, there is a high incidence of ra-
and lower intestine of dogs. J Bacteriol 85:643–648, 1963
diographic cervical emphysema in dogs with suspected 5. Neu HC, Gootz TD: Antimicrobial chemotherapy, in Baron
acute oropharyngeal stick injury, and this clinical finding S (ed): Medical Microbiology (ed 4). The University of
is indicative of deeper cervical tissue damage. Texas Medical Branch of Galveston, Galveston, TX, 1996,
pp 201–219
6. Nikolic I, Bumber Z, Stancic-Rokotov D, et al: Descending
necrotizing mediastinitis. Acta Clin Croat 45:207–211, 2006
REFERENCES 7. Freeman RK, Vallieres E, Verrier ED, et al: Descending
necrotizing mediastinitis: an analysis of the effects of serial
1. White RAS, Lane JG: Pharyngeal stick penetration injuries in surgical debridement on patient mortality. J Thorac Card-
the dog. J Small Anim Pract 29:13–35, 1988 iovasc Surg 119:260–267, 2000