Materials and Methods

Case selection—Medical records from 1998 to 2007 were reviewed for pet rabbits admitted to the Ontario Veterinary College Veterinary Teaching Hospital Avian and Exotics Service with dental abscesses. Data on rabbits were included in this study if there was documentation of the treatment outcome for a period of > 6 months after use of a wound-packing treatment protocol.²⁸

Pretreatment evaluation—Rabbits received a general physical examination before they were anesthetized with isoflurane or sevoflurane. Once anesthetized, a more thorough dental and physical evaluation were performed that included obtaining 4 radiographic views of the skull (ventrodorsal, laterolateral, and 2 obliques), CBC, plasma biochemistry profile, and endoscopically guided dental examination by use of a 2.7-mm, 30°-viewing rigid endoscope.^a Premolars, molars, and incisors were trimmed or equilibrated as necessary with a fine, medium, or coarse flame-shaped diamond composite burr^b and a low-speed, straight handpiece.^c Teeth that were sufficiently loose were extracted intraorally with a small luxator. Abscess treatment and endoscopically guided dental examinations were repeated weekly until the abscess had resolved.

Surgical procedures—Surgical treatment of abscesses was performed with the wound-packing technique previously described by Taylor.²⁸ The skin over the abscess was shaved, aseptically prepared, and incised, and the abscess capsule was lanced over its most dependent portion. The abscess cavity was cleaned and flushed with approximately 10 mL of sterile saline (0.9% NaCl) solution. Care was taken to avoid extensive dissection of the capsule from the skin. Inspection of the cavity was performed to locate the origin of the draining tract of the periapical infection if possible. Excess margins of the abscess capsule and skin were trimmed before the cavity was filled with strips of 3- to 5-mm-diameter sterile gauze^d aseptically cut from the folded edge and impregnated with the selected antimicrobial. The skin was closed with nonabsorbable suture material. This procedure was repeated weekly with removal and evaluation of the old packing and replacement with new gauze strips containing an appropriate antimicrobial. The volume of gauze used in the packing could usually be decreased each week as the abscess cavity retracted because of progressive wound healing. Treatment was discontinued when the abscess cavity was almost completely replaced by granulation tissue and no purulent material remained. As this endpoint was reached, the packing material frequently became adherent to the abscess wall. All the rabbits received systemic treatment with antimicrobials for the duration of the local treatment.

Sample collection for bacterial culture—Samples of purulent materials for aerobic and anaerobic^e bacterial culture were submitted to the Animal Health Laboratory, University of Guelph. Antimicrobial susceptibility testing with the disk diffusion method was performed for aerobic organisms. Antimicrobial susceptibility testing for cultured anaerobic organisms was not performed. Gram-stained smears were also evaluated.

Antimicrobial treatment—Ampicillin at a dose of 30 mg/kg (13.6 mg/lb) was the antimicrobial most commonly used for the first packing procedure. Choice of antimicrobial for local treatment was altered as necessary on the basis of results of bacterial culture and antimicrobial susceptibility testing, gram staining of purulent material, and patient response to treatment. If bacterial culture results indicated that large numbers of anaerobic bacteria were present and there was poor response to ampicillin, clindamycin at a dose of 20 mg/kg (9.1 mg/lb) was substituted. Other antimicrobials used to pack the wounds included cefazolin (20 mg/kg), cefoxitin (30 mg/kg), gentamicin (5 mg/kg [2.3 mg/lb]), and amikacin (10 mg/kg [4.5 mg/lb]).

The initial antimicrobial combination used for systemic treatment was trimethoprim-sulfamethoxazole (30 mg/kg, PO, q 12 h) with metronidazole (30 mg/kg, PO, q 12 h). Systemic treatment was also adapted according to bacterial culture results and patient response. Other antimicrobials used for systemic treatment included azithromycin (30 mg/kg, PO, q 24 h) and enrofloxacin (10 mg/kg, PO, q 12 h). In 2 rabbits, a cephalosporin-clindamycin AIPMMA bead was placed to treat focal osteomyelitis of a bony defect in the mandible that remained after the abscess in the cavity had resolved. All rabbits also received meloxicam (0.3 mg/kg [0.14 mg/lb], PO, q 12 to 24 hours) for a minimum of 3 days after wound packing; meloxicam administration was continued depending on the patient's apparent degree of comfort and appetite on clinical assessment.

Follow-up—All rabbits were reassessed after the abscess was considered to have resolved on a regular basis depending on the frequency of routine dental examinations and client compliance. Time to follow-up was defined as the time between the removal of the last packing and the last recheck. One rabbit successfully treated with the technique developed another abscess at a different site 2 years later and was included in the study as a separate case (ie, instance of disease). An additional rabbit failed to heal and thus was not included in calculations of the time or number of wound-packing procedures required for resolution of infection.

Statistical analysis—Data from only 12 (13 abscesses) of the 13 rabbits were included in calculations for number of wound-packing procedures needed and time to follow-up. Data from 1 rabbit were excluded from calculations for number of wound-packing procedures and time to follow-up as the treatment was not successful. All data were tested for normality independently by use of the Shapiro-Wilk test (α = 0.05), which revealed a normal distribution. Mean ± SD and 95% CI estimate values were then obtained.

Results

Thirteen rabbits (3 males and 10 females) with a total of 14 dental abscesses met the criteria for inclusion in this study. One of the 13 rabbits developed a second anatomically separate abscess 2 years after resolution of its first abscess.

Pretreatment evaluation—At the time of initial evaluation and diagnosis, the rabbits' ages ranged from 1 to 5 years. All rabbits had radiographic changes at the site of the abscess that consisted frequently of focal osteolysis, sometimes accompanied by osteoproliferation, thinning and remodeling of mandibular bone cortices, and dental root elongation (Figure 1). All rabbits had evidence of mild to moderate dental disease on endoscopically guided dental examinations characterized by crown elongations and deformities, dental spurs, and sometimes tooth loosening. In 1 rabbit, extensive bony lysis and periosteal proliferation of the mandibular bone were present. The 2-month-recheck radiography for this rabbit revealed the presence of 2 bony sequestra. Fractured teeth were not detected in any rabbit. Findings on CBC occasionally revealed mild nonregenerative anemia or mild heterophilic leukocytosis. Results of plasma biochemistry profiles were within reference ranges. At debridement, most abscesses were found to have a draining tract toward a periapical site. Multiple sites of drainage were not recorded for this series of abscesses.

View Full Size

Right lateral oblique radiographic view of the skull of a rabbit with a dental abscess. Soft tissue swelling is visible as well as bone lysis, proliferation, and remodeling of the mandibular bone (white arrows). The premolar and molar roots are also elongated and irregular (black arrowheads).

Citation: Journal of the American Veterinary Medical Association 237, 12; 10.2460/javma.237.12.1444

• Download Figure
• Download figure as PowerPoint slide

Bacterial culture results—Bacterial culture of samples from the 14 abscesses resulted in identification of various combinations of anaerobic (1 Bacteroides fragilis, 1 Bacteroides sp, 3 Prevotella melaninogenicus, and 4 that were not further identified) and aerobic (1 Proteus vulgaris [with suspected anaerobic gram-positive bacteria], 1 Pasteurella sp, 1 Streptococcus sp, 2 Actinomyces sp, and 1 Escherichia coli) bacteria from 6 and 5 abscesses, respectively. Bacterial culture yielded no growth in samples from 4 of the 14 abscesses; however, bacteria were identified on Gram-stained smears of the purulent material from each of these 4 abscesses, suggesting the presence of difficult to culture anaerobic bacteria (3 abscesses with suspected mixtures of gram-positive and -negative bacteria and 1 with suspected gram-positive cocci). Taking these 4 abscesses into account, 10 of the 14 abscesses contained or were suspected to contain anaerobic bacteria.

Antimicrobial treatment and follow-up—Ampicillin was used in the initial packing in 13 of the 14 abscesses and in all wound-packing procedures in 6 abscesses. Antimicrobials used in packings subsequent to the initial packing included clindamycin (3 abscesses), cephalosporins (cefazolin or cefoxitin; 3 abscesses), and aminoglycosides (gentamicin or amikacin; 2 abscesses).

Antimicrobials used initially for systemic treatment were the combination of trimethoprim-sulfamethoxazole with metronidazole (12 abscesses), enrofloxacin with metronidazole (1 abscess), and azithromycin (1 abscess). For 4 of the 14 abscesses, the initial antimicrobial used for systemic treatment was changed to azithromycin (3 abscesses) for an increase in antimicrobial spectrum and an increase in tissue penetration; for the treatment of 1 abscess, the antimicrobial was changed to enrofloxacin.

In addition to the wound-packing techniques, AIPMMA beads were used in 2 abscesses as the last treatment performed. In 1 of these 2 rabbits, the abscess resolved after 4 wound-packing procedures but recurred 1 month later and, at that time, was draining into the oral cavity. The rabbit received systemic treatment with an antimicrobial (azithromycin) for 2 weeks. Thirty-two months after the end of the treatment, the abscess recurred again at the same site. One clindamycin-impregnated gauze strip was placed in the abscess cavity and 1 week later was removed and replaced with a cefoxitin-clindamycin AIPMMA bead that was used to fill a small defect in the lateral mandibular bone. The rabbit was then followed for a period of 28 months without evidence of additional recurrence of the abscess.

The abscess of 1 rabbit from which E coli was isolated initially improved with treatment, but the abscess failed to completely resolve despite weekly wound packing and the use of various antimicrobials on the basis of susceptibility test results. At the 2-month follow-up examination, bacterial culture of the abscess confirmed the continued persistence of E coli and skull radiography revealed 2 sequestra in the mandibular bone. Surgical debridement of the bone and placement of gentamicin AIPMMA beads was suggested to the owner but was declined. The abscess was considered not cured, and the rabbit was lost to follow-up 16 weeks after the beginning of the treatment.

Mean ± SD number of procedures and packing procedures to obtain complete resolution of 13 abscesses was 4.8 ± 2.2 with a minimum of 1 and a maximum of 9 procedures (95% CI, 3.5 to 6.2). Rabbits were followed up and remained abscess-free for a period ranging from 6 to 78 months (mean ± SD, 32.6 ± 24.3 months; 95% CI, 17.9 to 47.3 months) after the removal of the final packing. Complete resolution was achieved for 13 of 14 abscesses.

Discussion

Our results on dental abscesses in rabbits treated with a wound-packing technique and followed up for a minimum period of 6 months indicate that minimal surgical intervention combined with antimicrobial-impregnated gauze packing of the abscess and systemic treatment with an appropriate antimicrobial is an effective treatment option. Bacterial culture of material from the dental abscesses in this study revealed the predominance of polymicrobial bacterial infections. Most organisms were anaerobes. These findings are consistent with those of the 2 studies^9,14 investigating the bacterial pathogens of dental infections in rabbits. While the physiologic oral flora of rabbits is unknown, in humans and other herbivore species, the mucosal and epithelial surfaces including the oropharynx are colonized by numerous aerobic and anaerobic bacteria with a large preponderance of anaerobic species.¹¹ In humans and herbivores, polymicrobial mixed aerobic-anaerobic infections are found in most periodontal infections, with anaerobic bacteria predominating.^{11–13,16,17} Routine culture of anaerobic bacteria is challenging and seldom carried out by veterinary laboratories. Receiving a result of no growth from a rabbit abscess specimen should raise the suspicion of unrecovered anaerobic organisms. We encourage clinicians to perform Gram staining on samples collected for bacterial culture of dental abscesses.

Antimicrobials with a spectrum of activity against anaerobic and gram-positive bacteria were used locally and systemically in our study in the management of dental abscesses in these rabbits and yielded good results. Clinically important anaerobes can vary widely in their antimicrobial susceptibilities.¹⁷ We could only find 1 report⁹ of antimicrobial susceptibilities for some commonly cultured bacteria from rabbit abscesses that included anaerobic isolates. As long as antimicrobial susceptibility testing is not routinely performed for anaerobic organisms, the clinical decision making for treatment of anaerobic pathogens will remain empirical. The study by Tyrell et al⁹ revealed a high susceptibility of isolated anaerobic organisms to azithromycin, chloramphenicol, and clindamycin (19/19 isolates); cefazolin, ceftriaxone, and penicillin G (18/19 isolates); and metronidazole and tetracyclines (15/19 isolates). Metronidazole was selected as a first-line antimicrobial for systemic treatment for 12 of 14 abscesses in our study. Metronidazole reaches sufficient concentrations in bony tissue, exceeding the minimal inhibitory concentrations for many anaerobic bacteria in rodents with experimentally induced bone infections.²⁹ Therefore, we believe metronidazole is a safe, inexpensive, and effective antimicrobial for the treatment of dental infections in rabbits. Azithromycin was used in our study in the treatment of 4 of 14 abscesses and was used as a second-line antimicrobial for persistent anaerobic infections that had poor response to metronidazole. Azithromycin reaches tissue concentrations higher than serum concentrations in rabbits.³⁰ It was found to be effective against 19 of 19 anaerobic and 11 of 12 aerobic bacterial isolates from dental abscesses in rabbits.⁹ We found this antimicrobial to be safe and effective for the treatment of difficult or recurrent dental infections in rabbits. The rabbits in this study did not develop any gastrointestinal adverse effects during antimicrobial treatment.

The site of specimen collection from within the abscess seemed to have little effect on bacterial culture or results of Gram staining. This has been suggested as a reason for no growth culture results from abscesses in rabbits⁸; however, lack of access to appropriate anaerobic culture techniques is a more likely explanation. This is supported by the fact that we identified bacteria on Gram-stained preparations of purulent material that did not grow in culture.

Several reports state that the affected tooth is frequently fractured.⁷ However, our study did not support that statement on the basis of interpretation of skull radiographs and endoscopically guided dental examination. For the rabbits in our study, where the tooth of origin could be identified, we frequently found that extraction was not required for successful resolution. Tooth extraction was only performed if the tooth was or became sufficiently mobile or dystrophic in appearance over the course of treatment. The general recommendation to extract teeth involved in dental abscesses during the initial debridement is questionable, and we believe that teeth that do not show substantial pathological changes and associated mobility should not be extracted during the initial surgery. Rather, teeth of questionable status should be reassessed over the course of the abscess management protocol as some will heal in response to the presence of continuously active germinal tissue. This recommendation is also consistent with the human literature, where it is generally recommended that tooth extraction that requires advanced surgical techniques should be delayed until the bacterial infection has been successfully controlled.^1,16 If an extraction or complicated debridement is deemed necessary in a rabbit, the wound-packing technique can be used to achieve control of the infection by reducing tissue inflammation and the bacterial load.

The wound-packing technique was not curative for an abscess of 1 rabbit in our study from which E coli was persistently cultured and aggressive bony changes were recorded. In our experience, E coli is uncommonly cultured from dental abscesses in rabbits, and it was not found in the study by Tyrrell et al.⁹ Escherichia coli is also known to be of increased pathogenicity in bone, compared with other bacteria commonly isolated from dental abscesses.³¹ The response to treatment of the rabbit with the E coli infection was substantially different than that of the other rabbits in our study, suggesting a more guarded prognosis when this bacteria is present.

Currently promoted therapeutic techniques^4,7,20,21 for the treatment of dental abscesses in rabbits have several drawbacks, in our opinion. First, complete removal of the abscess is technically difficult and may lead to iatrogenic damage to adjacent structures. Attempts at removal also carry a significant risk of spreading the bacterial inoculum into surrounding tissue. The surgical debridement and marsupialization technique^4,7,20,21 results in an open wound with the need for extensive daily wound care and also requires frequent anesthesia for repeated wound debridement. Pus from rabbits is usually highly viscous; therefore, passive drainage and flushing performed by owners may not result in effective removal of purulent debris. Daily flushing of wounds in the facial region is chronically stressful and unpleasant. The unattractive cosmetic appearance of the method and the risk of contamination of the wound area by food or substrate material can also be problematic for rabbits and their owners.

The AIPMMA beads can be difficult and expensive to manufacture. Each batch of AIPMMA must be mixed and packaged under sterile conditions with the addition of only 1 or at most 2 thermostable antimicrobials. Additional antimicrobials require completely new polymethylmethacrylate kits and repeat of the mixing and packaging process. The AIPMMA bead technique lacks the flexibility to change implants on the basis of bacterial culture results, antimicrobial susceptibility test results, and patient response to treatment.

The last technique involving the use of calcium hydroxide has been reported in the only other retrospective study²⁷ on long-term outcomes of dental abscesses in rabbits. This technique is not currently recommended because of the severe soft tissue reactions and skin sloughing that frequently develop at the surgery site.⁴

In contrast, the protocol described here results in a closed surgical wound with reduced requirements for client compliance beyond daily administration of antimicrobials for systemic treatment and a commitment to weekly rechecks. Wound management and assessment of response to treatment are performed under general anesthesia by the veterinarian allowing for more thorough examination and debridement while reducing discomfort and stress for the patient. This technique is flexible, permitting changes in the antimicrobial packing at low cost with ease of availability, compared with that with the use of AIPMMA beads. The use of AIPMMA beads was found to be a valuable adjunct to the wound-packing technique for management of those abscesses with focal bone defects and early osteomyelitis. The simplified surgical debridement used in our study resulted in a short anesthesia time with a low degree of postoperative pain and low risk of contamination to surrounding tissues. However, the use of this technique requires multiple short anesthetic procedure at frequent intervals that may have an effect on the function of the gastrointestinal tract of the rabbit.