Enucleation is a common orbital surgery performed in dogs to remove painful, non-visual eyes or for treatment of diseases such as intraocular neoplasia that are not amenable to alternative medical or surgical therapies. Post-operative complications include hemorrhage, draining fistulas from incomplete removal of secretory tissues within the orbit, and orbital emphysema in brachycephalic breeds (1). Despite the fact that enucleation is considered a clean-contaminated procedure (2) and there are inherent challenges in maintaining effective hemostasis and minimizing dead space, surgical site infection (SSI) is considered a rare complication (1). However, SSI is still important to consider as it can increase morbidity and treatment cost and potentially lead to delays in adjunctive therapy, such as administration of chemotherapy.

SSIs following procedures without implants typically occur within 30 days after the surgical procedure and up to a year following implant placement. Criteria for diagnosis include the presence of pain, redness, swelling, heat, and purulent discharge. Additionally, if an aseptic culture is obtained, there must be growth of a microbial organism (3). The overall incidence of SSIs in dogs and cats undergoing surgical procedures in veterinary teaching hospitals has been reported to be 5.1–8.7%, with clean-contaminated procedures developing SSIs in 4.5–8.6% of cases (4, 5).

The specific incidence and risk factors associated with SSI after enucleation in dogs have not been reported. In a report of 186 dogs that had a silicone orbital implant placed after enucleation of 215 eyes, confirmed infection occurred in two eyes (1%) and cellulitis without a culture obtained occurred in an additional two eyes (6). Similarly, a study on owner perceptions of bilateral enucleation in dogs reported a 2% incidence of SSIs (7). Alternatively, in 117 cases of enucleation in horses, the rate of SSIs was 7.5%. In that study, implant placement and standing procedures were associated with increased risk of SSIs (8). The goal of this retrospective study was to determine the incidence of SSIs and risk factors for SSI following enucleation with or without implant placement in dogs. Based on anecdotal experience, the authors hypothesized infection rates would be higher than previously reported. Additionally, the authors sought to determine if there is a benefit to post-operative antibiotic administration in this specific population.

Two hundred and eighty dogs were included in this study. The mean age of all dogs was 9.1 ± 3.7 years. There were 18 (6.4%) intact female dogs, 124 (44.3%) spayed female dogs, 23 (8.2%) intact male dogs, and 115 (41%) neutered male dogs. There were 61 breeds represented with the most common being mixed breed dog (43/280, 15.3%), shih tzu (32/280, 11.4%), Boston terrier (18/280, 6.4%), cocker spaniel (13/280, 4.6%), and Yorkshire terrier (13/280, 4.6%).

One hundred and forty-one dogs (50.4%) had the left eye enucleated, 121 dogs (43.2%) had the right eye enucleated, and 18 dogs (6.4%) had both eyes removed for a total of 298 enucleations performed. The primary indication for enucleation was associated with a clean eye in 161 dogs (57.5%), a non-clean eye in 98 dogs (35%), and neoplasia in 21 dogs (7/5%) (Table 1). Of the 98 dogs with a non-clean eye, 42 (42.9%) received pre-operative antibiotics (31/42 topical, 9/42 topical and systemic, and 2/42 systemic only). Comorbidities, including cardiac disease, kidney disease, liver diseases, m neurological disease, systemic hypertension, non-ocular neoplasia, diabetes mellitus, hypothyroidism, and hyperadrenocorticism were identified in 37/280 (13.2%) with diabetes mellitus being the most common comorbidity in 19/280 (6.8%) of dogs.

Surgical approach was recorded in 211 dogs with a transpalpebral approach used in 142/211 (67.3%) of dogs and subconjunctival used in 69/211 (32.7%) of dogs. For 200/280 (71.4%) of dogs, the surgeon was considered novice, while for 80/280 (28.6%) of dogs the surgeon was considered experienced. Use of local anesthesia was recorded in 141 dogs with 108/141 (76.6%) receiving a retrobulbar block and 19/141 (13.5%) receiving a splash block. Use of absorbable gelatin sponge (Gelfoam, Pfizer, New York, NY, USA) for hemostasis was recorded in 54/171 (31.6%) of dogs. Thirty-seven dogs had a concurrent ocular surgery performed, including phacoemulsification, canthoplasty, conjunctival graph placement, tarsorrhaphy, cryotherapy, and mass biopsy. Sixty-three dogs (22.5%) had a prosthetic implant placed. Information regarding the type of suture used for deep tissues and skin was recorded in 172 dogs with Polydioxanone (PDS, Ethicon, Bridgewater, New Jersey) being the most common used for deep tissues (161/172, 93.6%) and nylon (Ethilon, Ethicon, Bridgewater, New Jersey) being the most commonly used for skin closure (158/172, 91.9%). Information regarding length of anesthesia; length of surgery; and presence of hemorrhage, hypotension, and hypothermia was recorded in 275 dogs. The average duration of anesthesia for all cases was 118.4 minutes and the average duration of surgery for all cases was 63 minutes. Seventy-seven cases (28%) had intraoperative hemorrhage with 9 cases (1.2%) having lost greater than 1.3% of blood volume (9). Hypotension (systolic blood pressure less than 90 mmHg) occurred in 101 cases (36.7%) with an average lowest systolic blood pressure of 88 mmHg. Hypothermia (body temperature less than 98°F) occurred in 147 cases (53.5%) with an average lowest temperature of 97.3°F. Information on peri-operative antibiotic usage was recorded in 251 dogs. Perioperative antibiotics administered included cefazolin (233/251, 92.8%), ampicillin-sulbactam (17/251, 6.8%), and cefoxitin (1/251, 0.4%). Perioperative antibiotics were chosen based on clinician preference or aerobic culture and sensitivity results (n = 5) and administered 30 min prior to skin incision and every 90 min for the duration of surgery.

Two hundred and two dogs were administered systemic post-operative antibiotics based on clinician preference or aerobic culture and sensitivity results (n = 5), including amoxicillin-clavulanate (168/202, 83.2%), cephalexin (16/202, 7.9%), cefpodoxime (14/202, 6.9%), enrofloxacin (2/202, 1%), ampicillin (1/202, 0.5%), and clindamycin (1/202, 0.5%). Two hundred and thirty-four dogs received non-steroidal anti-inflammatory (203/234, 86.8%) or immunomodulatory (prednisone or chemotherapy) (31/234, 13.2%) medications within 2 weeks of surgery.

Time to follow-up ranged from 5 to 420 days in all dogs (median 14 days) and 5 to 420 days in dogs diagnosis with SSI (median 14 days). Fourteen dogs (14/280, 5%) developed an SSI. The median time to diagnosis of an SSI was 14 days (range 3–420). Documented clinical signs of infection included pain (8/14, 57.1%), redness (10/14, 71.4%), swelling (10/14, 71.4%), drainage from the surgical site (10/14, 71.4%) that was characterized as purulent (8/14, 57.1%) or hemorrhagic (2/14, 14.3%), dehiscence (3/14, 21.4%), or displacement of the implant (2/14, 14.3%). Five had the infection confirmed by culture with the following organisms identified: Staphylococcus aureus, multi-drug resistant (MDR) Staphylococcus psuedointermedius, MDR Staphylococcus intermedius, and MDR Staphyloccosus species.

The following were not associated with increased risk for development of an SSI: age; gender; weight; presence of comorbidities; concurrent diabetes mellitus; whether the reason for enucleation was clean, not clean, or neoplasia; administration of pre-operative antibiotics; length of anesthesia or surgery; presence of hypotension or hypothermia; unilateral or bilateral enucleation; surgical approach (transpalpebral vs. subconjunctival); experience of the surgeon; procedure for administration of local anesthesia (retrobulbar or splash block); use of absorbable gelatin sponge in surgery; concurrent ocular surgery; placement of a prosthetic implant; the type of suture used for deep tissues and skin; type of peri-operative antibiotic used; administration of post-operative antibiotics; or administration of anti-inflammatory or immunomodulatory medications.

A higher percentage of dogs that received cephalexin post-operatively developed an SSI vs. those that received other post-operative antibiotics (p = 0.045). However, when looking at overall post-operative antibiotic administration, dogs that received antibiotics were not less likely to develop an SSI (Table 2).

In the retrospective study reported here, the overall incidence of SSI following canine enucleation was 5%. This is similar to previously reported infection rates of 4.5–8.6% for clean-contaminated surgeries performed in veterinary teaching hospitals (4, 5) and similar to a reported SSI incidence of 7.5% in horses undergoing transpalpebral enucleation (8). However it is higher than two canine enucleation studies that reported an SSI incidence of <2 (6, 7). The SSI incidence of 5% reported here may be an overestimation. Four hundred eleven cases of enucleation were identified in the medical records search but only 280 had adequate follow-up information to allow inclusion. It is possible that the dogs without follow-up did not return to a veterinarian because they did not have any complications. Equally, only 5/14 (35.7%) of cases had confirmation of infection by culture so it is possible that the reported 5% is an overestimation.

Numerous previously reported SSI-risk factors, including pre-operative variables such as reason for enucleation; peri-operative variables such as surgical approach, surgeon experience, and surgical approach; as well as post-operative variables such as use of post-operative antibiotics or anti-inflammatory medications were assessed in the patient population presented here. Overall, no risk factors for development of SSI after enucleation were identified. The only finding that reached statistical significance was that dogs administered cephalexin post-operatively were more likely to develop an infection than those that received other post-operative antibiotics. However, the clinical significance of this finding is unclear. First, it is important to note that administration of post-operative antibiotics regardless of type did not reduce the risk of SSI. Additionally, a large number of variables were assessed in this study making it possible that this significant finding occurred due to chance. The authors do not believe this finding should change clinical decisions made for canine patients undergoing enucleation.

In equine patients undergoing enucleation, standing surgery and implant placement were associated with increased risk of SSI. All patients in our population were placed under general anesthesia for enucleation, and, although a minority had implants placed, implant placement was not associated with an increased risk for infection. It is possible that with longer follow-up times of at least a year, additional SSIs in the dogs that had implants placed would have been identified. Ultimately, a prospective study with larger case numbers may help identify risk factors associated with enucleation in dogs.

One impetus for this retrospective study was to determine if post-operative antibiotics reduce the risk of SSI following enucleation in dogs. As mentioned, administration of post-operative antibiotics did not decrease the risk of SSI in this study. This is consistent with a metaanalysis in people of 52 randomized controlled trials with 19,273 participants that did not find conclusive evidence that prophylactic post-operative antibiotics reduce the risk of SSI (10). Additionally, although only a small number (n = 5) of cases reported here had positive bacterial cultures, 60% (3/5) grew multi-drug resistant (MDR) organisms. This is in line with reports of human SSI where 37–65% of isolated organisms have a MDR profile (11, 12). Based on the fact that post-operative antibiotics did not reduce the risk of SSI and can increase the risk of development of MDR pathogens (13, 14), prophylactic post-operative use of antibiotics following enucleations in dogs is not supported.

There are several limitations to this study. It is retrospective in nature so some information in the medical records was incomplete and as a result certain previously reported risk factors such as increased duration of anesthesia and surgery as well as intraoperative hypotension could not be assessed (15–19). Although the study included a large number of dogs overall, only a small number of cases had SSI, which may have influenced our ability to identify risk factors. Not all cases of enucleation for the time period of the retrospective study were able to be included because they lacked adequate follow-up information; this may have skewed the SSI incidence and ability to identify risk factors. Also, as previously mentioned, not all cases of infection were confirmed with aerobic culture, which might have influenced SSI incidence.

Based on this report, risk of SSI after enucleation in dogs is 5%. No risk factors were identified to guide clinical decision-making for prevention of SSI. However, lack of confirmation that systemic post-operative antibiotics reduce SSI risk in combination with concerns about increasing numbers of MDR pathogens involved in SSI, indicates prophylactic antibiotics should not be used be after surgery in routine cases of canine enucleation.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

SD was responsible for data acquisition. SD and RB were responsible for data analysis. All authors were responsible for study design, manuscript preparation, and approved the submitted version of the manuscript.