Introduction
Abscesses or infections of the cervical region in dogs occur as firm to fluctuant swellings which can vary in size and appearance as well as in the involvement of associated structures of the neck such as regional lymph nodes. They most commonly occur secondary to trauma from events such as penetrating oropharyngeal injuries, inhaled foreign body migration, external foreign body penetration, or puncture wounds.1–8 A subsequent sterile or septic inflammatory response to foreign material leads to abscess formation.5 Mixed bacterial populations are most commonly isolated from canine septic foreign body reactions and abscesses, including Streptococcus spp, Staphylococcus spp, Pasteurella multocida, Escherichia coli, and Actinomyces spp.2–4
In humans, a similar condition exists called deep neck infection (DNI). Prevalent symptoms for this condition in adult humans include pain and swelling of the neck, while the condition in the pediatric population is characterized by the presence of a neck mass and fever.9 These infections pose critical challenges due to the possibility of causing airway compromise.9–12 The most common cause of this condition in humans is dental infections followed by sialadenitis, pharyngitis, and tonsillitis, although historically pharyngitis and tonsillitis predominated as causes before widespread antimicrobial usage.9,10 Common comorbidities associated with DNIs in humans include diabetes mellitus or an immunocompromised status.9 In humans, CT and ultrasonography are important diagnostic tools for identifying the presence of an abscess and in determining whether medical or surgical treatment is warranted.9,10 Additionally, the neutrophil-to-lymphocyte ratio (NLR) is commonly used to evaluate the inflammatory response and prognosis for many human diseases (cardiac disease, cancers, infections), and studies10,11 have shown that a higher NLR can be used to indicate the need for surgical intervention over medical management. Treatment for this condition initially includes antimicrobial treatment, and in cases that do not respond to antimicrobial treatment or develop abscess formation, surgical debridement is recommended.9–13 Such infections in humans are also commonly polymicrobial in nature, similar to canine infections.9
In the authors’ experience, the term cervical abscess is used broadly in veterinary medicine to characterize cervical swelling secondary to abscess formation or cellulitis affecting the fascial planes of the neck. However, a true abscess may not be evident in all cases. As such, the authors of the present study recommend using the nomenclature of DNI to describe this pathology. In dogs, the initial injury that leads to the development of a DNI is often unknown.1 The primary clinical sign of a dog with a DNI is swelling in the cervical neck region.5 Additional clinical signs include pyrexia, lethargy, anorexia, and signs of pain or sensitivity.1,5 Other differential diagnoses for cervical swellings are neoplasia, salivary mucocele, and sialoadenitis.5 As the treatment modalities and prognoses for these differential diagnoses are highly variable, a multifactorial approach is often required to confirm diagnosis. Imaging modalities such as ultrasonography, CT, and MRI are more sensitive than radiography at identifying potential foreign bodies, helping to determine the extent of the lesion, assessing of local lymph node involvement, and assisting surgical planning.6–8 Additionally, fine-needle aspirate and cytologic examination of the lesion can be used to further differentiate an inflammatory lesion from a neoplastic lesion. Treatment involves systemic antimicrobial treatment, pain management, and surgical debridement of the abscess with removal of any identified foreign or necrotic material.5–7 Bacterial culture and susceptibility testing should be performed, and empirical antimicrobial treatment should be initiated until culture results return.5,9 Fungal cultures may also be considered in some cases.4
A paucity of information is available in the literature on the diagnosis and management of DNIs in dogs. Most studies are focused on the prevalence and predisposing factors for migrating foreign body injuries with limited information specific to DNIs. As such, the objective of this descriptive case series is to describe clinical features, treatment, and outcomes in dogs with DNIs.
Materials and Methods
Medical record review
Medical records from January 1, 2015, through December 31, 2020, at the University of Georgia were searched to identify records of dogs with a DNI that underwent surgical treatment. Inclusion criteria were dogs that had a cervical swelling and one or more of the following: fever, septic suppurative inflammation on cytologic evaluation, or purulent material at surgery. Dogs that did not undergo surgery were excluded. Data collected included signalment, presenting complaint, duration of clinical signs, prior antimicrobial use, working dog status, history of playing with sticks, comorbidities, concurrent use of immunosuppressive therapies, rectal temperature on admission, neutrophil and lymphocyte count on admission, NLR, diagnostic imaging results, cytology results, surgical findings, whether a drain was placed intraoperatively, intra- and postoperative antimicrobial use, histopathology results, bacterial and fungal culture and susceptibility results, presence of foreign material identified via surgery or histopathology, and follow-up. Referring veterinarians and owners were contacted by telephone or email for additional follow-up. On owner contact, questions were asked about the dog’s current health status, recurrence of original clinical sings, and whether the dog frequently chewed on or played with sticks, plant material, or other items before the diagnosis of the DNI.
Statistical analysis
Statistical analysis was performed with a commercially available software (JMP Pro version 16.0.0; SAS Institute Inc). Descriptive statistics were generated. Normality of data was assessed by visual inspection of the histogram and the normal quantile plot. Normally distributed data were expressed as mean ± SD; nonnormally distributed data were expressed as median (range). Spearman correlation was used to compare continuous variables of interest (age, body weight, rectal temperature on presentation, duration of clinical signs before presentation, duration drain left in place, and duration of hospitalization) with NLR. Results for the correlation coefficient (ρ) were interpreted as negligible (0.00 to 0.10), weak (0.10 to 0.39), moderate (0.4 to 0.69), strong (0.70 to 0.89), or very strong (0.90 to 1.00).14 Wilcoxon rank sum was used to compare categorical variables of interest (CT findings [rim enhancement, non–contrast-enhancing fluid attenuation, fluid tracking along fascial planes, presence of an abscess or mass, presence of foreign material], surgical findings [presence of purulent discharge, presence of foreign material, whether a drain was placed], and whether a foreign body was identified on histopathology) with NLR.
Results
Dogs
Nineteen dogs were included. Mean age and body weight were 5.9 ± 4.2 years and 28.2 ± 12.2 kg. There were 9 castrated males, 8 spayed females, and 2 sexually intact males. Dogs were reported as mixed-breed dogs (n = 5), Golden Retrievers (4), Labrador Retrievers (2), Weimaraners (2), German Shepherd Dogs (2), and 1 each of Belgian Malinois, Boxer, German Wirehaired Pointer, and Miniature Poodle.
Presenting complaint, history, clinical signs
A cervical swelling or mass was a component of the owner’s presenting complaint for all 19 dogs. Additional historical findings included lethargy (4/19) and anorexia (3/19). The median duration of clinical signs before presentation was 4 days (range, 1 to 365 days). One dog had a history of right retropharyngeal pyogranulomatous lymphadenitis that had been treated and resolved 2 years prior to presentation. Six of the 19 dogs had received oral administration of antimicrobials (including amoxicillin, amoxicillin–clavulanic acid, cefpodoxime, cefovecin, cephalexin, clindamycin, enrofloxacin, metronidazole, or trimethoprim-sulfadiazine) before presentation. Of these 6 dogs, 5 initially improved but had recurrence of clinical signs, and 1 dog had no improvement. Dogs receiving prior antimicrobial treatment had a median duration of clinical signs prior to surgery of 91 days (10 to 365 days). One of these dogs was evaluated for a neurologic episode that consisted of ataxia, a right-sided head tilt, and circling to the right 3 months before presenting for surgery for the DNI. An MRI was performed, and the dog was diagnosed with left cervical cellulitis and myopathy. Neurologic signs resolved with initial antimicrobial treatment (enrofloxacin and amoxicillin–clavulanic acid), but the cervical swelling returned upon discontinuation of antimicrobials. Two additional courses of antimicrobial treatment (amoxicillin–clavulanic acid) were completed, and the abscess was lanced; however, cervical swelling recurred after each course of antimicrobials.
Owners of 6 of the 19 dogs reported histories of their dogs frequently chewing on sticks and other plant material. Of these 6 dogs, 2 had an owner-witnessed incident of a penetrating oropharyngeal injury while playing fetch with a stick before the development of clinical signs. Eighteen of the 19 dogs were nonworking pets, and the remaining dog was in training to be a hunting dog. Comorbidities were reported in 6 of 19 dogs and were myxomatous mitral valve disease (n = 2) and 1 each of immune-mediated thrombocytopenia, hypothyroidism, acanthomatous ameloblastoma, and chronic diarrhea. The dog with immune-mediated thrombocytopenia was receiving immunosuppressive doses of prednisone, cyclosporine, and mycophenolate at presentation.
On physical examination, all dogs had a cervical neck swelling. Nine of the 19 dogs had a rectal temperature > 39.2 °C. One additional dog had a history of intermittent hyperthermia but was normothermic on presentation. Facial and ventral neck edema, in addition to the cervical swelling, was reported in 4 dogs. Three of the 19 dogs were found to have had mandibular lymphadenomegaly on intake examination. Limited ability to palpate and differentiate mandibular lymph nodes due to cervical swelling was reported in 3 dogs. Two of the 19 dogs had ocular signs reported, including exophthalmos and decreased retropulsion in one dog and entropion, hyperemia, and elevated nictitating membranes in the other. There were no neurologic deficits or respiratory impairments reported on intake examination for the 19 dogs included in the study. One dog had a presenting complaint of labored breathing but was found to be eupneic on intake examination, and another dog had stertorous breathing noted on intake examination.
Diagnostic imaging
Diagnostic imaging was performed for 18 of the 19 dogs and included CT (13/18), ultrasonography (5/18), and MRI (2/18), with 2 dogs having undergone both ultrasonography and CT. Based on results of diagnostic imaging, 15 of the 18 dogs had involvement of the regional lymph nodes, most commonly the mandibular and medial retropharyngeal lymph nodes.
Findings on CT were the presence of a distinct soft tissue abscess or mass (7/13), an abscessed lymph node (4/13), cellulitis without a discrete mass (1/13), and a tooth root abscess (1/13). Additional CT findings included contrast enhancement of the rim, capsule, or periphery of the soft tissue swelling (10/13), non–contrast-enhancing fluid attenuation of adjacent tissues tracking along fascial planes (8/13), intralesional gas (4/13; one of which had placement of a drain prior to referral), and suspected foreign material (4/13; Figure 1). Additionally, 6 of the 13 dogs that underwent CT were noted to have had displacement of the trachea, pharynx, or larynx, with 2 dogs having concurrent compression noted of those structures and 1 dog with a concurrent acute stylohyoid fracture.
Transverse plane postcontrast CT angiographic images (the left side of the dog is on the right side of each image) of 3 of 19 dogs with deep neck infections (panels C and D are from the same dog) evaluated between January 1, 2015, and December 31, 2020, representing commonly identified abnormalities on CT: contrast enhancement of the rim, abscess capsule, or both (solid white arrowheads [A, B, and C]); a non–contrast-enhancing fluid-filled center (asterisks [A, B, and C]); fluid tracking through fat and fascial planes (open arrows [A and B]); intralesional gas (white arrows [A and C]); mineral-dense foreign material (solid black arrow [B]); compression and displacement of the larynx (open arrowheads [C and D]); and severe cellulitis and subcutaneous swelling extending rostrally from the abscess area (dagger [D]).
Citation: Journal of the American Veterinary Medical Association 260, 14; 10.2460/javma.22.04.0169
Transverse plane postcontrast CT angiographic images (the left side of the dog is on the right side of each image) of 3 of 19 dogs with deep neck infections (panels C and D are from the same dog) evaluated between January 1, 2015, and December 31, 2020, representing commonly identified abnormalities on CT: contrast enhancement of the rim, abscess capsule, or both (solid white arrowheads [A, B, and C]); a non–contrast-enhancing fluid-filled center (asterisks [A, B, and C]); fluid tracking through fat and fascial planes (open arrows [A and B]); intralesional gas (white arrows [A and C]); mineral-dense foreign material (solid black arrow [B]); compression and displacement of the larynx (open arrowheads [C and D]); and severe cellulitis and subcutaneous swelling extending rostrally from the abscess area (dagger [D]).
Citation: Journal of the American Veterinary Medical Association 260, 14; 10.2460/javma.22.04.0169
Transverse plane postcontrast CT angiographic images (the left side of the dog is on the right side of each image) of 3 of 19 dogs with deep neck infections (panels C and D are from the same dog) evaluated between January 1, 2015, and December 31, 2020, representing commonly identified abnormalities on CT: contrast enhancement of the rim, abscess capsule, or both (solid white arrowheads [A, B, and C]); a non–contrast-enhancing fluid-filled center (asterisks [A, B, and C]); fluid tracking through fat and fascial planes (open arrows [A and B]); intralesional gas (white arrows [A and C]); mineral-dense foreign material (solid black arrow [B]); compression and displacement of the larynx (open arrowheads [C and D]); and severe cellulitis and subcutaneous swelling extending rostrally from the abscess area (dagger [D]).
Citation: Journal of the American Veterinary Medical Association 260, 14; 10.2460/javma.22.04.0169
Clinicopathologic findings
Fine-needle aspirate and cytology of the affected area were performed in 13 of the 19 dogs prior to surgical intervention. Results were septic suppurative inflammation (7/13), pyogranulomatous inflammation (3/13), reactive lymph node (1/13), hemodiluted sample (1/13), and poorly cellular sample (1/13).
A CBC was performed at admission for 15 of the 19 dogs. Median absolute neutrophil count was 11.48 X 103 cells/µL (range, 5.530 X 103 to 33.02 X 103 cells/µL; reference range, 2.700 X 103 to 8.500 X 103 cells/µL), median lymphocyte count was 1.540 X 103 cells/µL (range, 0.320 X 103 to 3.530 X 103 cells/µL; reference range, 0.500 X 103 to 4.100 X 103 cells/µL), and median NLR was 7.500 (range, 2.130 to 96.00; reference range yet to be determined). There was a moderate association (ρ = –0.548; P = .035) between duration of clinical signs before presentation and NLR, with a shorter duration of clinical signs correlating with a higher NLR. There was also a moderate association (ρ = 0.645; P = .009) between duration of hospitalization and NLR, with a longer duration of hospitalization correlating with a higher NLR. There was no association between NLR and age, body weight, rectal temperature on presentation, and duration a drain was left in place. There was no association between CT, surgical, and histopathologic findings and NLR except that dogs with a higher NLR were more likely to have a drain placed at surgery (P = .049).
Surgery
All 19 dogs underwent surgical exploration of the cervical region. The mean number of days dogs were hospitalized before surgery was 2.3 ± 2.2 days. Empirical antimicrobial treatment (ampicillin-sulbactam, amoxicillin–clavulanic acid, cefpodoxime, and doxycycline) was started at intake for 8 of the 19 dogs. Perioperative antimicrobials were used in 16 dogs and included cefazolin (n = 8) or ampicillin-sulbactam (8).
All 19 surgeries consisted of abscess exploration, drainage of any encountered purulent material, debridement, and lavage. A ventral cervical midline approach was the most common (13/19), and the remaining dogs had an incision directly over the swelling or mass (6/19). In dogs where rim enhancement was described on CT (10/13), purulent material was encountered in 7 dogs (7/10) at the time of surgery. Of the 3 dogs for which rim enhancement was not described on CT, 2 had purulent material encountered at surgery. During surgery, foreign material was identified in 3 dogs and consisted of wood (n = 1), other plant particles (1), and feathers (1); however, only 1 of these 3 dogs had a suspected foreign body identified on CT.
Lymph node extirpation was performed in 10 of the 19 dogs and included mandibular lymph nodes (7/10), the medial retropharyngeal lymph node (5/10), or both (2/10). Drains were placed in 10 of the 19 dogs (7 with closed-suction drains, 3 with passive drains). The median duration of drain placement was 3 days (range, 2 to 8 days).
Aerobic culture and susceptibility testing was performed for all 19 dogs, and 10 had positive bacterial growth. An additional dog had no growth of bacteria, but dead bacteria were identified on Gram stain. Specific isolated bacterial species included Klebsiella pneumoniae, Acinetobacter baumannii, Alcaligenes faecalis, Enterococcus faecium, Escherichia hermannii, Staphylococcus pseudintermedius, Kocuria kristinae, Streptococcus spp, Neisseria animaloris, Staphylococcus schleiferi, and Nocardia spp. Anaerobic bacterial culture was performed for 14 of the 19 dogs, with positive growth in 4 of these 14 dogs; 3 of these 4 dogs also had a positive aerobic culture. Specific bacterial species isolated included Bacteriodes pyogenes, Bacteroides uniformis, and Parvimonas micra. In all, 11 of 19 dogs had a positive bacterial culture, and mixed bacterial populations were isolated from 8 of 19 dogs. Fungal culture was submitted for 7 of the 19 dogs, and all results were negative for fungal growth.
Histopathologic examination was performed for 16 of the 19 dogs and included evaluation of the excised cervical tissue from 15 dogs and lymph nodes from 10 dogs. Histopathologic diagnosis for cervical tissue samples consisted of pyogranulomatous or suppurative myositis, panniculitis, cellulitis, or fasciitis (n = 14), with intralesional bacteria identified in 4 of these 14 dogs and plant material identified in 1 of the 14 dogs. In this dog with plant material identified on histopathologic analysis, visible plant material was not encountered during surgery, and this patient had not undergone diagnostic imaging before surgery. In all, a foreign body was suspected on CT in 4 dogs, was identified at surgery in 3 dogs (including 1 dog in which foreign material was also identified on CT), and was identified on histopathology in 1 dog; thus, foreign material was confirmed in 4 of the 19 dogs. For the remaining dog that had histologic examination of excised cervical tissue, the histologic diagnosis was thyroid follicular cystadenoma with neutrophilic and histiocytic cellulitis; this dog had a concurrent positive aerobic culture. The histopathologic diagnosis for all sampled lymph nodes consisted of pyogranulomatous or necrosuppurative lymphadenitis or lymphoid hyperplasia.
Postoperative management
All dogs survived to hospital discharge with a median duration of hospitalization of 4 days (range, 1 to 12 days). All dogs were discharged with orally administered antimicrobials, the most common of which was single-agent amoxicillin–clavulanic acid (11/19), followed by cefpodoxime, clindamycin, enrofloxacin, and doxycycline in 1 dog each. Four dogs received multiple antimicrobials, including amoxicillin–clavulanic acid and enrofloxacin (3/19) or amoxicillin–clavulanic acid and metronidazole (1/19). Duration of antimicrobial treatment postoperatively was a median of 14 days (range, 10 to 28 days). Adjustments to antimicrobial treatment based on bacterial culture and susceptibility results were made for 6 of 19 dogs, with treatment extended in 3 dogs or antimicrobials changed in 3 dogs. Immediate postoperative complications were reported in 1 dog, which developed a seroma.
Follow-up
One dog died 7 days following surgery due to suspected gastric dilatation-volvulus. The median follow-up time for the remaining 18 dogs was 937 days (range, 68 to 2,138 days). One dog developed swelling in the cervical neck region 1 week following surgery and was assessed by the referring veterinarian. A follow-up bacterial culture and susceptibility test were submitted at that time, which necessitated a change in antimicrobial treatment from amoxicillin–clavulanic acid to enrofloxacin. The swelling resolved following this treatment and did not recur, with a follow-up time of 1,266 days. Another dog also developed mild cervical swelling in the postoperative period. Based on bacterial culture and susceptibility results, metronidazole was added to the previously prescribed amoxicillin–clavulanic acid, and treatment was extended. The swelling resolved following this treatment and did not recur, with a follow-up time of 917 days. The remaining 16 dogs with follow-up information available had no reported recurrence of the original clinical signs.
Discussion
This case series described the clinical management and outcomes of dogs treated surgically for DNIs. This pathology is identified as a localized infection with potential cellulitis and abscess formation affecting the deep fascial planes and potential spaces of the neck.9–13,15–20 Both dogs and humans have superficial and deep neck fascia.9,13,21 In dogs, description exists for the superficial and deep fascia of the neck with the deep fascia subdivided into 3 zones (pretracheal fascia, prevertebral fascia, and endothoracic fascia).21 In humans, superficial and deep fascial planes have been described with the deep fascial plane further subdivided into a superficial, middle, and deep layer. Furthermore, up to 11 potential deep neck spaces have been described in humans, and disease characteristics and prognosis are unique for DNIs affecting each of these spaces.9,14 To the authors’ knowledge, these 11 potential deep neck spaces are not described in dogs; however, the comparative anatomy of the neck between dogs and humans remains similar. This similarity in anatomy strengthens the support for using the nomenclature of DNI to describe this pathology in dogs.
Canine DNIs present unique challenges in their diagnosis and management and in evaluating the need for surgical intervention. In humans with DNIs, there is a serious risk of complications including airway compromise if this condition is not addressed in a prompt manner.9–13,15 As such, surgical drainage is an important tool in the management of DNIs. More recently, several human studies9,16–20 have identified conservative management with parenteral antimicrobials as a treatment strategy in uncomplicated cases of DNIs. No dogs in the present report had airway compromise, although 1 dog had a history of labored breathing but was eupneic at presentation, and a second dog had stertor noted on examination. On CT, compression or displacement of the trachea, larynx, or pharynx was identified in 6 dogs. Although airway compromise was not directly encountered in any of the dogs of the present report, the potential for respiratory compromise and the need for airway management should be considered in all cases of DNI due to the close anatomic relationship that exists with the upper airway and deep fascial planes of the neck.
When evaluating surgical versus medical management, human literature9 suggests that both are viable treatment options and should be considered based on individual patient needs and severity of clinical signs. Medical management is typically only considered in cases of uncomplicated DNIs in people.9–12,16–20 In the present report, 6 dogs were treated with orally administered antimicrobials before referral, with 5 dogs showing improvement but subsequent recurrence after completing treatment. Successful medical or conservative management of uncomplicated DNIs in humans consists of either hospitalization with parenteral antimicrobial treatment and supportive care or a short course of parenteral antimicrobial treatment followed by orally administered antimicrobial treatment.16–19 Because surgery was performed for all the dogs of the present report, the authors were unable to evaluate cases in which medical management alone may have resolved clinical signs. The inability to identify dogs treated with medical management alone was a limitation of the present study. One benefit of medical management may be consolidation of the lesion, which could make surgery more straightforward at a later date. Further investigation to determine criteria for cases that may benefit from medical management alone or prior to surgery is indicated.
A presenting complaint of cervical neck swelling served as an important feature in increasing the index of suspicion for DNIs in the dogs of the present report, but advanced imaging also played a useful role in the ultimate diagnosis of DNI and exclusion of other differential diagnoses. The most commonly used imaging modality in the present study was CT, which allowed for the identification of involved fascial planes, soft tissue changes such as a mass or abscess versus cellulitis or edema, and lymph node involvement. All of these are important to aid surgical planning. One study10 from human literature identified rim enhancement on CT scan as a positive predictor for successful surgical drainage of purulent material from a suspected abscess. Peripheral rim enhancement was a common finding on CT in the dogs of the present report (10/13 dogs) and has been identified as an important marker in evaluating for the presence of a drainable abscess, whereas cellulitis and edema are typically seen as low-attenuating hypodensities without rim enhancement on CT and are typically low yield when aspirated or lanced.22,23 In the present report, purulent material was encountered at the time of surgery in 7 of the 10 dogs in which rim enhancement was described and 2 of the 3 dogs in which rim enhancement was not described. These findings suggested that while rim enhancement seen on CT scan can increase the index of suspicion for a drainable abscess associated with a DNI, the absence of rim enhancement does not negate surgical exploration in cases where a DNI is prioritized.
Another prognostic indicator identified for DNIs in humans is the NLR, which has also been used as marker for infection and as a prognostic indicator for cardiovascular and neoplastic disease in human medicine.10,11 One group11 found that an NLR of ≥ 4.65 in DNIs correlated to a length of hospital stay of ≥ 2 days. Another group identified an NLR of > 8.2 as a positive predictor for drainable abscesses and in determining the need for surgical intervention.10 In the veterinary literature, the utility of NLR for various conditions is still being validated. One group found that NLR was significantly higher in dogs and cats with acute pancreatitis, but they could not confirm a correlation between NLR and disease severity.24 Another group found that in cases of septic peritonitis in dogs, an NLR of ≥ 6 had an 84% sensitivity and 87% specificity in distinguishing dogs in states of systemic inflammatory response from clinically normal dogs; however, no correlation could be found to separate septic and nonseptic causes or relate to duration of hospital stay or patient morbidity.25 In the present report, length of hospitalization was longer in dogs with higher NLRs, which was similar to findings in humans.11 Higher NLRs were also found in dogs that were presented after a shorter duration of clinical signs, which was likely associated with these dogs’ having been presented during the more acute phase of inflammation. Also in the present study, higher NLRs were associated with the placement of a drain during surgery; this finding was also likely related to the more acute phase of inflammation in these dogs, in which a drain may be considered beneficial for residual fluid production. Further studies with larger numbers of dogs are necessary to investigate the utility of NLR for DNIs in dogs.
Although the most common causes of DNI in people are dental infections and sialoadenitis,9,10 the main etiology in dogs may be expected to be different due to the difference in species and canine propensity for chewing on sticks or other foreign material and sniffing of the ground. In the present study, foreign material was suspected in 4 dogs on the basis of CT findings but was only identified at surgery in 1 of these dogs. It could have been that the foreign material was removed at surgery but not identified or that the suspected foreign material identified on CT represented something other than foreign material. Foreign material was identified in an additional 2 dogs at surgery and 1 dog on histopathology that was not previously suspected on CT. Previous reports1–3,5 suggest that penetrating oropharyngeal injuries are a major inciting cause of DNIs in dogs, and it is possible that more dogs in the present study had foreign material as the inciting cause despite the lack of definitive identification of foreign material. Further studies are needed to determine how CT may aid in the identification of foreign material in DNIs in dogs. Similar to humans, odontogenic and oropharyngeal conditions could also play a role in the occurrence of DNIs in dogs. This was evident in 1 dog in the present report for which previous clinical history, physical examination, and CT revealed the presence of a chronic tooth root abscess and subcutaneous fistula formation in the ventral neck region contributing to development of DNI. As could have been expected, large-breed dogs were more common in the present study. Although few dogs were working dogs, many dogs were of breeds commonly used for hunting or retrieving work and may have had a propensity to be in situations where foreign material could enter the body, leading to a DNI. However, the Miniature Poodle was an unexpected finding, as this is not a breed associated with hunting or retrieving work. Dogs of any breed may chew on sticks or other natural materials which could lead to a DNI; thus, consideration should be given to a DNI as a differential diagnosis for any dog presenting with a cervical swelling with or without a fever.
In the present report, histopathology was performed for 16 of 19 dogs. Although most results were consistent with an inflammatory process, 1 dog was diagnosed with a thyroid cystadenoma that may have predisposed to a DNI; thus, histopathology is recommended to identify or rule out an underlying cause precipitating the DNI. Bacterial culture and susceptibility testing were also employed in the management of DNIs, with all dogs in our study having had aerobic bacterial culture and susceptibility testing performed and 14 of the 19 having had anaerobic bacterial culture and susceptibility testing performed. Positive bacterial growth was identified for 10 of the 19 aerobic samples and 4 of the 14 anaerobic samples, with 11 of the 19 dogs having had a positive result on bacterial culture. One consideration for negative results on bacterial culture for dogs in the present study was the use of antimicrobials before sample collection, which may have decreased the likelihood of robust bacterial growth. One dog had a negative result for bacterial culture but dead bacteria evident on gram stain, consistent with this theory. Similar to human DNIs,9 polymicrobial infections were identified in 8 of the 19 dogs. The standard of care in human medicine is to start empirical treatment with antimicrobials while awaiting culture results.9–12,16–20 This strategy was also employed for all dogs in the present report, as all dogs were started on antimicrobials during their surgical visit before culture results were obtained. The most commonly used antimicrobial was amoxicillin–clavulanic acid, either as a single agent or in combination with another antimicrobial, with ampicillin-sulbactam used commonly as a perioperative antimicrobial. Bacterial culture and susceptibility testing results were used to guide treatment, with 3 dogs having a change in antimicrobials and 3 dogs having an increase in length of antimicrobial treatment based on the results of repeated culture. Given these findings, the initial empirical choice of amoxicillin–clavulanic acid appears to be a reasonable choice for first-line treatment while culture and susceptibility results are pending; ampicillin-sulbactam may be considered as a parenteral option in these cases. Fungal cultures were submitted for 7 of the 19 dogs, with no growth reported for all cases. In human medicine, fungal cultures are recommended for all immunocompromised individuals with suspected DNIs.9 One study4 in dogs evaluating bacterial and fungal isolates from inhaled grass foreign bodies of the respiratory tract reported the isolation of fungal organisms for 4 of 41 dogs. Although no fungal organisms were identified in the dogs of the present report, fungal culture should be considered for dogs with suspected DNIs, as the case numbers in this study were low and positive fungal growth would require a change in treatment. The importance of fungal cultures may also depend on the geographic location, with an increased need for fungal cultures in areas with endemic fungal species.
The main limitation of the present study was its retrospective nature; however, many details were able to be obtained from the medical records, including imaging, surgical, and histopathologic findings. Additionally, although some dogs had shorter follow-up times, the overall median follow-up was 937 days. Although only 19 dogs were identified to have undergone surgical treatment of a DNI during the study period, results indicated that empirical treatment with a broad-spectrum antimicrobial (such as amoxicillin–clavulanic acid or ampicillin-sulbactam) was a good option and that long-term resolution was possible with surgical and antimicrobial treatment. The decision to perform lymphadenectomy was surgeon dependent. It was possible that the lymph nodes could have been helping to clear infection locally, but there were also some dogs in which the lymph node itself was abscessed, an indication for lymph node extirpation. A study with larger numbers of dogs with and without lymphadenectomy is necessary to determine the utility of lymph node extirpation in these cases.
To the authors’ knowledge, the present study was the first to detail presenting signs, diagnosis, medical management, surgical technique, and outcomes in dogs with DNIs. Results of the present study demonstrated that in dogs with cervical swelling with or without a history of oropharyngeal trauma, DNIs should be considered on the differential diagnosis list. Cross-sectional diagnostic imaging paired with cytology can prove useful in distinguishing DNIs from other conditions and in evaluating the need for surgical intervention. Treatment consisted of surgical exploration, drainage, debridement of abscessed tissue, and antimicrobials. All dogs with long-term follow-up after surgery had resolution of clinical signs, with no reported recurrence. These results suggested that surgical management combined with antimicrobial treatment is an effective, durable treatment leading to complete, long-term resolution of DNIs.
Acknowledgments
No external funding was used in this study. The authors declare that there were no conflicts of interest.
References
- 1. ↑
Griffiths LG, Tiruneh R, Sullivan M, Reid SW. Oropharyngeal penetrating injuries in 50 dogs: a retrospective study. Vet Surg. 2000; 29(5):383–388. doi:10.1053/jvet.2000.9140
- 2. ↑
Brennan KE, Ihrke PJ. Grass awn migration in dogs and cats: a retrospective study of 182 cases. J Am Vet Med Assoc. 1983; 182(11):1201–1204.
- 3. ↑
Kirpensteijn J, Fingland RB. Cutaneous actinomycosis and nocardiosis in dogs: 48 cases (1980–1990). J Am Vet Med Assoc. 1992; 201(6):917–920.
- 4. ↑
Flisi S, Dall’Aglio M, Spadini C, Cabassi CS, Quintavalla F. Microbial isolates from vegetable foreign bodies inhaled by dogs. Vet Med Int. 2018;2018:3089282. doi:10.1155/2018/3089282
- 5. ↑
Reiter AM, Soltero-Rivera MM. Oral and salivary gland disorders. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine: Diseases of the Dog and the Cat. 8th ed. Elsevier Saunders; 2017:1469–1476.
- 6. ↑
Jones JC, Ober CP. Computed tomographic diagnosis of nongastrointestinal foreign bodies in dogs. J Am Anim Hosp Assoc. 2007; 43(2):99–111. doi:10.5326/0430099
- 7. ↑
Goldsworthy SJ, Burton C, Guilherme S. Parotid duct foreign body in a dog diagnosed with CT. J Am Anim Hosp Assoc. 2013; 49(4):250–254. doi:10.5326/JAAHA-MS-5810
- 8. ↑
Caivano D, Birettoni F, Rishniw M, et al. Ultrasonographic findings and outcomes of dogs with suspected migrating intrathoracic grass awns: 43 cases (2010–2013). J Am Vet Med Assoc. 2016; 248(4):413–421. doi:10.2460/javma.248.4.413
- 9. ↑
Vieira F, Allen SM, Stocks RM, Thompson JW. Deep neck infection. Otolaryngol Clin North Am. 2008; 41(3):459–483. doi:10.1016/j.otc.2008.01.002
- 10. ↑
Ban MJ, Jung JY, Kim JW, et al. A clinical prediction score to determine surgical drainage of deep neck infection: a retrospective case-control study. Int J Surg. 2018;52:131–135. doi:10.1016/j.ijsu.2018.02.024
- 11. ↑
Gallagher N, Collyer J, Bowe CM. Neutrophil to lymphocyte ratio as a prognostic marker of deep neck space infections secondary to odontogenic infection. Br J Oral Maxillofac Surg. 2021; 59(2):228–232. doi:10.1016/j.bjoms.2020.08.075
- 12. ↑
Stalfors J, Adielsson A, Ebenfelt A, Nethander G, Westin T. Deep neck space infections remain a surgical challenge. A study of 72 patients. Acta Otolaryngol. 2004; 124(10):1191–1196. doi:10.1080/00016480410017864
- 13. ↑
Favaretto N, Fasanaro E, Staffieri A, et al. Deep neck infections originating from the major salivary glands. Am J Otolaryngol. 2015; 36(4):559–564. doi:10.1016/j.amjoto.2015.01.003
- 14. ↑
Schober P, Boer C, Schwarte LA. Correlation coefficients: appropriate use and interpretation. Anesth Analg. 2018; 126(5):1763–1768. doi:10.1213/ANE.0000000000002864
- 15. ↑
Warshafsky D, Goldenberg D, Kanekar SG. Imaging anatomy of deep neck spaces. Otolaryngol Clin North Am. 2012; 45(6):1203–1221. doi:10.1016/j.otc.2012.08.001
- 16. ↑
Russell MD, Russell MS. Urgent infections of the head and neck. Med Clin North Am. 2018; 102(6):1109–1120. doi:10.1016/j.mcna.2018.06.015
- 17.
Lawrence R, Bateman N. Controversies in the management of deep neck space infection in children: an evidence-based review. Clin Otolaryngol. 2017; 42(1):156–163. doi:10.1111/coa.12692
- 18.
Oh JH, Kim Y, Kim CH. Parapharyngeal abscess: comprehensive management protocol. ORL J Otorhinolaryngol Relat Spec. 2007; 69(1):37–42. doi:10.1159/000096715
- 19. ↑
Broughton RA. Nonsurgical management of deep neck infections in children. Pediatr Infect Dis J. 1992; 11(1):14–18. doi:10.1097/00006454-199201000-00005
- 20. ↑
Kirse DJ, Roberson DW. Surgical management of retropharyngeal space infections in children. Laryngoscope. 2001; 111(8):1413–1422. doi:10.1097/00005537-200108000-00018
- 21. ↑
Hermanson JW. The muscular system. In: Evans HE, de Lahunta A. Miller’s Anatomy of the Dog. 4th ed. Elsevier Saunders; 2013:185–280.
- 22. ↑
Lazor JB, Cunningham MJ, Eavey RD, Weber AL. Comparison of computed tomography and surgical findings in deep neck infections. Otolaryngol Head Neck Surg. 1994; 111(6):746–750. doi:10.1177/019459989411100608
- 23. ↑
Miller WD, Furst IM, Sàndor GK, Keller MA. A prospective, blinded comparison of clinical examination and computed tomography in deep neck infections. Laryngoscope. 1999; 109(11):1873–1879. doi:10.1097/00005537-199911000-00029
- 24. ↑
Neumann S. Neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios in dogs and cats with acute pancreatitis. Vet Clin Pathol. 2021; 50(1):45–51. doi:10.1111/vcp.12979
- 25. ↑
Hodgson N, Llewellyn EA, Schaeffer DJ. Utility and prognostic significance of neutrophil-to-lymphocyte ratio in dogs with septic peritonitis. J Am Anim Hosp Assoc. 2018; 54(6):351–359. doi:10.5326/JAAHA-MS-6808
