Parotid Duct Sialolithiasis in a Dog

Introduction

In humans, sialolithiasis is fairly common.¹ Clinical signs usually include swelling of the affected gland with or without pain during eating. In animals, salivary calculi are uncommonly reported, with most cases occurring in the parotid duct of the horse.^2–5 To the authors’ knowledge, only five cases of canine sialolithiasis, each involving the parotid duct, have been thoroughly reported.^6–10 These dogs were presented with a history of acute to chronic facial swelling, with or without associated pain. In some cases, the parotid duct had ruptured.^7,8 Diagnosis was made via plain radiography, ultrasonography, or sialography. Treatments varied from medical to surgical intervention, which included sialolith removal via an oral incision and repair of the parotid duct or parotidectomy.^6–10 The purposes of this study were to describe the computed tomographic diagnosis of parotid duct sialolithiasis in a dog and the accompanying histological changes, as well as to discuss potential etiologies and diagnostic and treatment options.

Case Report

A 21.6-kg, 7-year-old, spayed female English bulldog was referred to the North Carolina State University, College of Veterinary Medicine, for evaluation of a recurrent swelling over the lateral aspect of the left cheek, just ventral to the zygomatic arch. Acute, nonpainful facial swelling was first noticed by the owner 3 months previously. There was no known history of trauma. The dog was treated empirically with corticosteroids and antibiotics by the referring veterinarian, and the swelling resolved temporarily. Over the next 3 months, the swelling recurred several times and was treated similarly. The owner reported that with each recurrence, the swelling was larger and became painful. Examination during one of these episodes revealed purulent fluid of unknown origin within the caudal oral cavity. A tooth root abscess was suspected. Ventrodorsal skull radiographs revealed a focal opacity (possible sialolith) lateral to the angle of the mandible and medial to the zygomatic arch [Figure 1]. No other abnormalities were seen on the radiographs other than some mineralization of the horizontal ear canals. The dog was referred for further evaluation.

Physical examination revealed the dog to be bright, alert, and responsive; all vital signs were within normal limits. A small, firm area was palpated on the left side of the dog’s face, in the region of the parotid duct, just caudal to the level of the fourth maxillary premolar. No other abnormalities were detected on physical examination. The dog’s packed cell volume, total solids, blood urea nitrogen, and serum glucose were normal.

Following sedation with medetomidine hydrochloride^a (10 μg/kg intravenously [IV]), computed tomography (CT) of the skull was performed with a third-generation CT scanner^b (3-mm contiguous slices at 120 kVp, 240 mAs). Computed tomography revealed two hyperintense structures [Figure 2A] measuring 1 to 4 mm in width on the lateral surface of the left masseter muscle, at the level of the angle of the mandible. Caudally, they were contiguous with a tubular structure of a similar density to fluid [Figure 2B] that became increasingly dilated, bifurcated, and continued on to the parotid salivary gland. The tubular structure was consistent with the parotid salivary duct. At its widest point, the duct measured 9 × 8 mm. The salivary gland was heterogeneous in appearance, which may have indicated dilated, fluid-filled ductules. The affected parotid gland was enlarged but had similar contrast enhancement when compared to the contralateral side.

An exploratory surgery of the affected area was performed to determine the cause of the swelling. The dog was premedicated with hydromorphone hydrochloride^c (0.05 mg/kg intramuscularly [IM]), induced with thiopental sodium^d (10 mg/kg IV), and maintained under general anesthesia with isoflurane and oxygen. The dog was placed in lateral recumbency, and the skin of the lateral aspect of the left face and neck was prepared aseptically for surgery. Carprofen^e (4.4 mg/kg subcutaneously) was administered preoperatively for pain. Cefazolin sodium^f (20 mg/kg IV) was administered before and 2 hours after induction.

Attempts to cannulate the parotid papilla and duct were unsuccessful. A left-sided, 6-cm skin incision was made over the lateral aspect of the face, parallel to the parotid duct. The parotid duct was isolated, and a focal thickening of the duct was identified approximately 2 cm caudal to the parotid papilla. A longitudinal incision was made into the lumen of the thickened parotid duct. One large (3.3 mm) and several smaller (approximately 1 mm) sialoliths were found to be obstructing the duct. The sialoliths were removed, and the largest was submitted for mineral analysis.^g The distal parotid duct was ligated just caudal to its entry into the oral cavity, and the remaining duct was dissected caudally to the level of the parotid gland. The parotid gland was isolated and removed in its entirety. The gland was firmer than normal in texture and was adherent to the surrounding tissues. A sample of the parotid gland was submitted for aerobic bacterial culture and sensitivity. The remainder of the parotid gland and duct was submitted for histopathological examination. The subcuticular and subcutaneous tissues were closed with 3-0 polydioxanone^h using a simple continuous pattern. Skin closure was performed using staples. Recovery from anesthesia was uneventful. Postoperative pain was managed with hydromorphone (0.05 mg/kg IM as needed for 24 hours) and a transdermal fentanyl patchⁱ (50 μg per hour continuously for 4 days). The dog was discharged on amoxicillin/clavulanate^j (19 mg/kg orally [PO] q 12 hours for 5 days) and carprofen^e (2.2 mg/kg PO q 12 hours for 7 days).

Cultures of the parotid gland yielded no growth. The single sialolith submitted for analysis measured 3.3 × 2.3 × 2.1 mm and weighed 0.01 gm. The surface of the calculus was rough, chalky in appearance, and light tan in color [Figure 3]. Cross-section of the calculus revealed a surface layer of crystals surrounding an interior of fine, light tan, poorly oriented crystals. Based on infrared spectroscopy, both layers were composed of 100% calcium carbonate.

Histopathological findings of the affected parotid gland and duct [Figures 4A, 5A, 6A] were compared to the left parotid gland and duct harvested from a control dog that had been euthanized at a local shelter [Figures 4B, 5B, 6B]. Examination of the affected parotid salivary gland revealed moderate to marked, diffuse atrophy of the glandular lobules, with moderately thickened bands of interlobular fibrous connective tissue (i.e., fibrosis). Serous epithelium was generally composed of low cuboidal cells with less cytoplasmic granularity compared to the normal gland. Scattered throughout the lobules were mild to moderate infiltrates of lymphocytes and occasional plasma cells. Surrounding the larger interlobular ducts were more intense infiltrates of lymphocytes and plasma cells, as well as increased layers of periductal, fibrous connective tissue. The parotid duct epithelium was ulcerated in multiple areas. The periductal connective tissue was expanded by dense, collagenous connective tissue and numerous congested blood vessels with moderate to marked infiltrates of neutrophils, plasma cells, lymphocytes, and macrophages. A focal, circumscribed area of mineralization was noted underlying a region of ulceration in the wall of the duct. A diagnosis of chronic, neutrophilic, and ulcerative sialoangiitis with moderate to marked, diffuse lobular atrophy and mild lymphoplasmacytic adenitis was made.

The dog was seen by the referring veterinarian 5 days after discharge for a fluctuant, nonpainful swelling over the incision site. Blood was aspirated from the site. The incision appeared to be healing normally. The owner was instructed to apply warm compresses and monitor for further swelling. No further problems were noted at the time of skin staple removal or 14 months postoperatively when the owner and referring veterinarian were contacted by telephone.

Discussion

In humans, sialolithiasis is fairly common, accounting for 30% of all salivary abnormalities in adults.¹ Sialolithiasis is much less common in children.¹¹ The mandibular salivary duct (i.e., Wharton’s duct) is most commonly affected (83% to 92%), compared to 6% to 10% for the parotid duct, and the condition is usually unilateral.^12,13 Clinical signs often include swelling of the affected gland, with or without pain during eating. The etiology of sialolith formation is unknown. Numerous theories have been proposed, including the excretion of microcalculi by the acinar cells and the presence of intraluminal mucous plugs or inflammatory debris (possibly from bacterial migration and secondary inflammation) that may act as an organic nidus for deposition of inorganic components.^1,12–14 Bacteria or foreign bodies may also act as a nidus for sialolith formation.¹⁴ Sialolith formation may both be a cause for and result of salivary retention.^1,13 Inflammation arising from the mechanical irritation of the calculus may result in salivary stasis, owing to spasm of the duct.¹³ Anatomical factors that may predispose to salivary stasis and sialolith formation include salivary duct stenosis or diverticuli, hair follicles within the duct, or surgical transposition of the duct.^1,11 If obstruction occurs secondary to sialolith formation, chronic obstructive sialadenitis (the most common form of salivary gland inflammation in humans) may result and contribute to a repetitive cycle of sialolith formation and inflammation.¹⁵

To the authors’ knowledge, only five cases of canine parotid duct sialolithiasis with treatment information have been reported, although 13/48 cervical sialoceles contained sialoliths in one report.¹⁶ Of these five sialoliths, one was composed of calcium carbonate, one was a mixture of calcium and magnesium carbonate, and one was a mixture of calcium carbonate and magnesium ammonium phosphate. The composition of the others was not noted.^6–10 A variety of clinical presentations and treatments were reported. Acute facial pain and swelling occurred in an 8-year-old, male English bulldog that resolved with antibiotics and corticosteroids.⁶ Medical management was similarly attempted for the dog reported here, but resolution of clinical signs was only temporary. In an 11-year-old, spayed female English springer spaniel, an infraorbital fluctuant swelling was detected that was associated with parotid duct rupture 17 months after a duct transposition.⁷ An 8-week-old, spayed female Cavalier King Charles spaniel developed a painful swelling ventral to the ear canal. During surgical exploration, a rupture of the parotid duct was identified and repaired following removal of a sialolith.⁸ In both of these latter cases, surgical repair of the parotid duct resulted in clinical resolution. By comparison, repair of the duct was deemed implausible for the dog reported here, as the duct was grossly thickened and was later confirmed to be ulcerated and fibrotic. In a fourth case, a 10-year-old, spayed female cocker spaniel had clinical signs similar to the current case, which also resolved following surgical resection of the parotid gland and duct.⁹ However, unlike the current case, the parotid duct had ruptured before surgery, and excision of an associated sialocele was required. In the last case, a 7-year-old, castrated male Cavalier King Charles spaniel was presented with a 2-month history of a painful, unilateral, retropharyngeal swelling.¹⁰ The swelling temporarily resolved following removal of a sialolith via an oral incision, but ultimately parotid salivary gland excision was required. It is possible that progressive fibrosis of the duct, as seen in the current case, resulted in recurrence of clinical signs and the need for a second surgery. Although it is interesting to note that of these six canine cases, four were spaniels and two were bulldogs, little can be concluded regarding breed predisposition, because the condition is rare and hospital-based breed frequencies were not reported.^6–10

Diagnosis of canine sialolithiasis was based on results of plain radiography in three cases and ultrasonographic and sialographic evaluations in one case.^6,8,10 Imaging was not performed in the other two cases.^7,9 In humans, imaging modalities commonly used to diagnose sialolithiasis include sialography, sialoendoscopy, ultrasonographic examination of the salivary gland and ducts, and, more recently, CT and magnetic resonance imaging.^17–24 Many (15% to 40%) sialoliths in humans are radiolucent; therefore, sialography was once considered the preferred method of diagnosing sialoliths in people.^12,19,23 Plain radiographs of the dog reported here were suggestive of a sialolith, and sialography was considered; however, the parotid papilla could not be cannulated, so sialography was not an option. More recently, sialendoscopy has become a commonly performed procedure in people, because the sialolith and the duct wall can be visualized while the sialolith is removed.^18,23 This technique, although highly successful in identifying submandibular ductular lesions in people, requires a specialized semirigid endoscope that was not available for the dog of this study. It is not clear if this technique can be performed in the relatively small canine parotid duct; however, it has been used in children and may require incision into the duct or papilla for insertion of the scope.¹¹

Ultrasonography has had variable success (43% to 90% accuracy) in identifying sialoliths in humans.^19,21 Although small stones may be missed, as was the case for one dog where ultrasonography was performed, this modality allows evaluation of the associated gland.¹⁰ Ultrasonography could have been used in this case, but it was felt that CT would provide more information. Computed tomography was performed, because it has been highly accurate for identifying sialoliths in people and also allows simultaneous evaluation of the parotid duct and gland.²¹ In the dog reported here, CT provided more information than plain radiographs regarding the status of the parotid duct and gland, and the exact location of the radiographic opacity (i.e., sialoliths) could be identified.

The parotid gland was removed in this dog, because it was felt the duct would not remain patent after sialolith removal. Because most sialoliths in humans occur in the submandibular duct, distal sialoliths may be removed via an intraoral incision (i.e., papillotomy), and more caudal (proximal) stones may be removed by a variety of techniques, including endoluminal retrieval with the aid of an endoscope or fluoroscopy.^{12,17–19,22,25,26} Larger stones may be fragmented with an endoluminal laser or by either endoluminal or external lithotripsy prior to endoluminal retrieval.^{12,18,22,26,27} Alternatively, sialoliths may be removed via a lateral incision of the face, and if multiple stones or extremely proximal stones are present, the associated gland and duct may be surgically removed.^12,19,22 Despite the high success rate of minimally invasive sialolith removal, sialolithiasis remains the most common reason for submandibular salivary gland excision in people.²²

Because this dog had recurrent clinical signs associated with ductal obstruction from the sialolith, and it was unclear if there was an infectious component to the condition, ligation of the duct to induce glandular atrophy was not considered. Ligation of the parotid duct has been used to treat four dogs and one cat for complications following parotid duct transposition, trauma, or hypersialosis, and it has not been associated with any deleterious side effects.^28–30 The effects of ligation on salivary glands apparently vary by both species and gland, although, in general, glandular atrophy occurs over time.^31–34

Comparing the results of salivary duct ligation studies to the histopathological changes seen in this dog may help to determine whether ductal obstruction by the sialoliths was responsible for or a consequence of the parotid gland pathology. Parotid duct ligation in rats resulted in rapid acinar cell death (12 hours to 5 days).³¹ Dying cells were largely removed by activated intraepithelial macrophages. After 4 weeks of obstruction, only ductules embedded within fibrous tissue were evident.³¹ Extreme glandular atrophy was also noted in cats following parotid duct ligation, although the atrophy was not as rapid (12 days).³² When the mandibular salivary duct was ligated in cats, atrophy was much less predictable, and mucocele formation was commonly reported. Mandibular duct ligations were performed in six dogs in another study.³³ Minimal acinar atrophy occurred in one dog; marked atrophy with interstitial fibrosis occurred in another dog; and no changes were reported in the mandibular glands of the remaining four dogs 3 months after ligation.³³ A larger study involving 24 dogs with parotid duct ligations found fibroblastic interstitial proliferation 45 days postoperatively.³⁴ At 60 to 90 days postoperatively, marked fibroblastic proliferation, ductular epithelial flattening, and acinar collapse were detected.³⁴ Edema and congestion of the parotid glands were also noted. At 120 days postoperatively, advanced fibrous tissue proliferation with acinar atrophy and involution was found, similar to that seen in the dog of this report. Sporadic lymphocytic infiltration was also noted.³⁴

Few specifics regarding the histopathology of canine salivary glands with sialoliths have been reported. Two previous cases had sialadenitis, and fibrous replacement of the parotid gland was noted in one case.^9,10 The acinar atrophy and fibrosis noted in the dog reported here were consistent with the changes seen after experimental ductal ligation, and they were probably related to chronic obstruction by the sialolith. The precipitating cause of this dog’s sialolithiasis was unclear, although an organic core was not identified within the largest sialolith. The history of a progressively painful swelling was compatible with either sialadenitis or intermittent ductal obstruction. It was not known whether the inflammation found on histopathology resulted solely from obstruction of the duct or served as an initiating factor for calculi formation. Bacteria and debris present within the oral cavity may have ascended via the parotid duct and caused localized inflammation, thereby starting the process of calculi formation.^1,12–14 Temporary resolution of the initial swelling with medical therapy did not resolve this question. Since corticosteroids and antibiotics were used simultaneously, it was unclear if the swelling subsided because of antiinflammatory or antibiotic effects, or if the resolution was purely coincidental. No infectious agent was identified on culture of the parotid gland or on histopathological examination of the duct and gland, which made a primary infectious cause less likely. It was also possible that primary obstruction of the duct (e.g., fibrosis, stricture) from an unwitnessed trauma may have provided the nidus for calculus formation secondary to inflammation or stasis of saliva. The presence of the sialoliths likely resulted in the mechanical irritation, epithelial ulceration, local inflammation, and ductal fibrosis seen in this case.

Conclusion

Parotid sialolithiasis was diagnosed with the aid of CT in a dog with unilateral facial swelling. Surgical excision of the parotid gland and duct was curative. Based on the results of this report, CT evaluation of dogs with swelling in the region of the parotid gland or duct should be considered. Parotidectomy appears to be a viable treatment for dogs with parotid sialolithiasis.