Introduction

Bacterial infections of the central nervous system (CNS) in small animals are uncommon.¹ In a study evaluating inflammatory and infectious diseases of the CNS in dogs, only 7% of dogs with meningoencephalitis had a bacterial etiology.² In veterinary literature, only a few retrospective reviews on bacterial CNS infections have been published.^2,3 The bacterial organisms that cause meningitis in dogs are varied, with the most commonly isolated species in one retrospective study of dogs being Escherichia coli, Streptococcus spp., and Klebsiella spp.³ Other bacteria that have been isolated include Staphylococcus spp., Pasteurella spp., Proteus spp., Salmonella spp., Actinomyces spp., Nocardia spp., and Mycoplasma spp., as well as anaerobes, such as Prevotella oralis, Fusobacterium spp., Bacteroides spp., Peptostreptococcus spp., Eubacterium spp., Flavobacterium breve, and Propionibacterium spp.^1,3–6 Bacterial infections can spread to the CNS via multiple routes, including hematogenous spread from a distant site, contiguous spread (nasal cavity, retrobulbar space, oral cavity, paranasal sinuses, or otitis media and interna), or direct penetration (penetrating foreign material, trauma, or bite wounds and along nerves or nerve roots).^1,3,6–8 Here, we describe the clinical characteristics and outcome of enterococcal meningoencephalitis in a 13 yr old spayed female Yorkshire terrier. To the authors’ knowledge, this case report is the first case in veterinary literature of Enterococcus spp. leading to meningoencephalitis in a dog.

Case Report

A 13 yr old 1.1 kg spayed female Yorkshire terrier was presented for evaluation of a 1 wk history of progressive lethargy, a 1 day history of anorexia, and diarrhea with melena. The patient had a history of a traumatic T5 cranial endplate fracture and subluxation 2 yr before that was medically managed, leading to a persistent, mild T3-L3 myelopathy.

On physical examination, the dog was dull, had a rectal temperature of 39°C, and was cachectic. She had severe periodontal disease with gingival recession and diarrhea with melena on rectal examination. Historical, stable findings included bilateral, hypermature cataracts and ambulatory paraparesis. The rest of the physical examination was within normal limits.

Complete blood count^a revealed a mild normocytic, hypochromic, non- or preregenerative anemia (hematocrit 38.7%; reference range 41–58%) and a moderate leukocytosis at 22,250 cells/μL (reference range 5700–14,200 cells/μL) characterized by a neutrophilia at 18,690 cells/μL (reference range 2700–9400 cells/μL) with a left shift at 450 band neutrophils/μL (reference range 0–100 cells/μL), and a monocytosis at 1780 cells/μL (reference range 100–1300 cells/μL). Moderate toxic change and reactive lymphocytes were noted. The platelet count was within the reference range. Serum biochemistry^b revealed an elevated blood urea nitrogen at 66 mg/dL (reference range 5–30 mg/dL) with a normal creatinine at 0.8 mg/dL (reference range 0.7–1.8 mg/dL), and hypoglycemia at 58 mg/dL (reference range 65–112 mg/dL). A baseline cortisol was performed in order to rule out hypoadrenocorticism and was high at 6.85 μg/dL. Urine aerobic culture was negative for bacterial growth.

Abdominal ultrasound revealed moderate gastric wall thickening, particularly of the mucosa, which was mildly hyperechoic, and mild fluid distention of the stomach. Subjectively decreased motility was noted. The duodenum and occasional segments of the jejunum were mildly thickened and moderately corrugated. The colonic wall was mildly thickened ≤2 mm and contained echogenic material. Otherwise, the ultrasound was generally unremarkable. Thoracic radiographs revealed generalized mild cardiomegaly and tracheal collapse but was otherwise unremarkable. A noninvasive Doppler blood pressure was normal (110 mm Hg).

The dog was hospitalized, and therapy was initiated for suspected gastrointestinal hemorrhage and presumed gastroenterocolitis with associated functional ileus. Therapy included IV fluid therapy with dextrose supplementation; an anti-emetic, maropitant^c (1 mg/kg IV q 24 hr); a proton pump inhibitor, pantoprazole^d (1 mg/kg IV q 12 hr); a prokinetic, metoclopramide^e (2 mg/kg/day IV continuous rate infusion); analgesia, methadone^f (0.1 mg/kg IV q 6 hr); and antibiotic therapy, enrofloxacin^g (10 mg/kg IV q 24 hr) and metronidazole^h (10 mg/kg IV q 12 hr) because of the concern for bacterial translocation. Despite therapy, the dog remained anorexic, showed signs of nausea, including hypersalivating and lip-licking, and became progressively dull the next day. Recheck blood work after 24 hr of therapy revealed persistent hypoglycemia at 64 mg/dL (reference range 65–112 mg/dL), hypernatremia at 158 mmol/L (reference range 140–150 mmol/L), hyperchloremia at 123 mmol/L (reference range 109–120 mmol/L), a progressive normocytic, hypochromic, nonregenerative anemia (hematocrit 31.6%; reference range 41–58%), and a mildly improved leukocytosis at 21,070 cells/μL (reference range 5700–14,200 cells/μL) characterized by a mature neutrophilia at 18,330 cells/μL (reference range 2700–9400 cells/μL) with a resolved band neutrophilia. The patient appeared euhydrated and had one episode of diarrhea with melena during this time.

A nasogastric feeding tubeⁱ was placed without complication in order to quantify residual gastric contents and provide nutritional support. Liquid nasogastric feedings^j at one-fourth of the dog’s daily resting energy requirement and water supplementation (24 mL/day) were initiated and tolerated well. An additional anti-emetic, ondansetron^k (0.5 mg/kg IV q 8 hr); gastrointestinal mucosal protectant, sucralfate^l (250 mg via feeding tube q 8 hr); and subcutaneous cyanocobalamin^m (25 μg/kg subcutaneously once) were initiated.

The next day, the dog remained anorexic and was assessed to be progressively dull, mildly hypothermic (36.7°C), and hypertensive (Doppler 170 mm Hg). Additionally, the patient continued to intermittently pass small amounts of loose stool with melena. Recheck of electrolytes revealed progressive hypernatremia at 162 mmol/L (reference range 140–150 mmol/L) and hyperchloremia at 128 mmol/L (reference range 109–120 mmol/L). These electrolyte derangements were suspected to be due to continued hypotonic losses within the gastrointestinal tract. Shortly thereafter, the dog became acutely, progressively neurologic, exhibiting a stuporous level of consciousness, bilateral thoracic limb rigidity, and bilateral pelvic limb flexion concerning for pathology of the caudal cranial fossa. Evaluation by our neurology service and further diagnostics were discussed, including advanced imaging and cerebrospinal fluid tap. At this point, the owners elected to euthanize because of the need for further advanced diagnostics and continued supportive care with an unknown prognosis.

Necropsy revealed severe, suppurative, acute meningoencephalitis, ventriculitis, hypophysitis, and otitis interna (Figure 1). The sella turcica and leptomeninges overlying the hypothalamus, just caudal to the pituitary gland, contained a small amount of thick, green, purulent exudate. A sterile sample of this exudate was submitted for aerobic culture because of academic interest. Aerobic culture revealed heavy growth of Enterococcus spp. The species was identified as either E casseliflavus or E gallinarum, which cannot be differentiated using a biochemical identification system.ⁿ Antimicrobial susceptibility testing was not performed because this is not routine with postmortem specimens. Anaerobic culture was negative for bacterial growth. Acute, focal, ulcerative pharyngitis was also noted. The ventricular system contained sheets of neutrophils and mononuclear phagocytes that occasionally contained intracytoplasmic coccoid bacteria.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

Enterococci are α- or nonhemolytic gram-positive cocci that can form short chains and often possess Lancefield group D antigens. Common species include Enterococcus faecalis, Enterococcus faecium, and Enterococcus hirae. They are commensal bacteria of the gastrointestinal tract, skin, oral cavity, and nasal cavity of dogs and cats.^1,6

Enterococci are intrinsically resistant to antimicrobials that are generally chosen to target gram-positive organisms, including most β-lactams, due to the expression of low-affinity penicillin-binding proteins, aminoglycosides due to limited drug uptake, and fluoroquinolones due to multiple mechanisms, including mutations in the DNA gyrase and topoisomerases genes.^9–13 Even though enterococci may appear susceptible in vitro to potentiated sulfonamides, they have documented limited in vitro activity because of the bacteria’s ability to evade the in vivo folate synthesis block.^10,14 Acquired resistance to lincosamides and macrolides are also common.^10,15

Enterococcal meningitis is rare in humans, making up only 0.3–4.2% of cases of bacterial meningitis.¹⁶ Enterococcal meningitis occurs most commonly in humans after a neurosurgical procedure. A majority of these cases have CNS devices in situ at the time of development of meningitis.^16–17 Less commonly, spontaneous, often community-acquired meningitis can occur and is generally associated with severe underlying diseases, immunosuppression, or another associated enterococcal infection. A majority of cases have an acute clinical course.¹⁶ When the species is identified, E. faecalis accounts for the majority of cases in humans whereas a smaller proportion of cases are associated with vancomycin resistant E faecium.^16–17 Most cases are treated with ampicillin, a carbapenem, linezolid, or vancomycin, with or without aminoglycosides.⁹

The enterococcal species identified in the described patient (E casseliflavus or E gallinarum) are intrinsically resistant to vancomycin. Recent literature regarding the treatment of these enterococcal species in humans recommend treating CNS infections with ampicillin in combination with an aminoglycoside. The recommended second-line treatment is linezolid.¹⁸

Despite the fact that both enrofloxacin and metronidazole penetrate the CNS effectively, these empirically chosen antibiotics used in this patient would not have treated an enterococcal infection because of its intrinsic resistance to fluoroquinolones and the lack of activity metronidazole has against enterococcal species.^1,19–20

The source of the bacterial infection remains unknown in our patient. Possibilities discussed included local extension from otitis interna, periodontal disease, hematogenous spread from an unknown source, or extension from the ulcerated nasopharyngeal region. These areas of concern (middle ear, inner ear, nasopharynx) were not cultured, and histopathology did not identify bacteria. Irritation of the pharyngeal mucosa secondary to placement of the nasogastric feeding tube is a consideration for the ulcerative pharyngitis given the lack of bacteria noted on histopathology. The patient did not have a history of known recent trauma. An antemortem diagnosis in this patient would have required cerebrospinal fluid culture, which was not obtained, because this diagnostic was declined by the owner.

Conclusion

To the authors’ knowledge, this case report is the first case in veterinary literature of Enterococcus spp. leading to meningoencephalitis in a dog. In cases of acute-onset neurological symptoms, bacterial meningitis should be considered. When possible, cerebrospinal fluid culture should be obtained, because empiric therapy for the most common causes of bacterial meningitis may not be effective against Enterococcus spp.