Incidence and risk factors of surgical site infection in ruminant species following internal fixation for orthopedic injury: 81 cases (2010–2023)

Abstract

Objective

To determine the prevalence and risk factors of surgical site infection (SSI) following internal fixation in ruminants.

Study design

Retrospective multicenter study across five veterinary teaching hospitals.

Animals

A total of 81 ruminants undergoing internal fixation for orthopedic disease.

Methods

Medical records over a 13‐year period were analyzed for ruminants presenting with orthopedic disease that was surgically managed with internal fixation. Patient signalment, limb affected and type of injury, surgical method, orthopedic implants used, total anesthetic duration, antimicrobial use, development of SSI as well as short‐ and long‐term survival were reported.

Results

The incidence of SSI was 17.3% (14/81). Overall, 92.6% (75/81) of cases survived to discharge; 85.7% (12/14) of cases with SSI and 94% (63/67) of cases that did not develop SSI. Increased duration of antimicrobial use (p = .004) and fractures distal to the tarsus/carpus (p = .04) were associated with an increased risk of SSI but had no effect on overall survival. The use of intramedullary pins for surgical repair was associated with a reduced likelihood of survival to discharge (p = .008).

Conclusion

In conclusion, the SSI incidence in ruminants following internal fixation was comparable to the infection risk in horses and camelids. SSI had a higher incidence in fractures distal to the tarsus/carpus and in animals that were administered antibiotics for a prolonged period.

Clinical significance

The use of orthopedic implants in ruminants is associated with a good outcome with 92.6% of cases surviving to discharge and a comparable incidence of surgical site infection as other large animal species.

1. INTRODUCTION

Surgical site infection (SSI) remains a major complication of orthopedic procedures in both human and veterinary medicine. ¹ , ² , ³ , ⁴ Infections that develop in either the incision or deep tissue at the surgical site up to 1 year following surgery are considered to be surgical site infections following the use of orthopedic implants. ⁵ , ⁶ , ⁷ In the veterinary literature, the incidence of SSI associated with orthopedic implants in both equine and small animal species has been well reported. ⁴ , ⁸ , ⁹ , ¹⁰ , ¹¹ A retrospective study in 2010 identified SSI in 28% of equine long bone fractures, ¹² and a more recent study identified an improved SSI incidence of 14.2% when looking at both arthrodesis and long bone fractures. ³ Horses that developed SSI were 12 times less likely to survive to hospital discharge and implant infection remains the leading cause of mortality following orthopedic procedures. ³ , ¹³ In canines, the incidence of SSI following surgery involving the use of orthopedic implants is widely reported with a high degree of procedural variation, with rates of 0.3%–21.1% reported. ¹⁴ SSI in livestock species following internal fixation has been sparsely published in the literature, including one previous study which evaluated complications following orthopedic procedures in llamas and camelids. ¹⁵ The retrospective study reported postoperative complications including infection, poor fracture healing or soft tissue injury in 87% of cases. ¹⁵

The most common indication for the use of internal skeletal fixation in ruminants is for the treatment of long bone fractures; however, other less common indications include joint arthrodesis for traumatic injuries as well as management of specific congenital and degenerative diseases. ¹⁶ , ¹⁷ With the increasing popularity of small ruminants as pets, in addition to high commercial value of some ruminants, the demand for surgical methods to repair orthopedic diseases in these animals is increasing. However, despite this, there are currently no studies on the prevalence or risk factors associated with surgical site infection in ruminant species that have had internal fixation of an orthopedic disease. Surveillance data on SSI is essential to both improving practices to prevent the development of infection and providing accurate information to guide treatment recommendations. The objective of this study was to determine the prevalence of SSI in ruminant cases requiring orthopedic implants. Additionally, we aimed to identify risk factors associated with the development of surgical site infection. We hypothesized that the prevalence of SSI is higher than those published for similar internal fracture fixation in horses. We also hypothesized that systemic and local antimicrobial therapy would be protective against the development of surgical site infection and that open fractures or longer surgical times would be risk factors for surgical site infection.

2. MATERIALS AND METHODS

Medical records of all ruminant species (bovine, ovine, caprine, other) that received internal fixation using orthopedic implants between January 2010 and December 2023 were obtained from five veterinary teaching hospitals (Johnston Family Equine Hospital, Colorado State University; Galbreath Equine Center, The Ohio State University; Morrie Waud Large Animal Hospital, University of Wisconsin‐Madison; University of Georgia Veterinary Teaching Hospital, University of Georgia; William. R. Pritchard Veterinary Medical Teaching Hospital, University of California‐Davis). An individual was responsible for gathering data from medical records from each institution to ensure consistency and data was collected into a standardized computer‐based spreadsheet (Microsoft Office Excel, Microsoft Corporation, Redmond, Washington). The criteria for the diagnosis of SSI was the presence of at least two of the following: positive bacterial culture, presence of swelling, drainage of serosanguinous or purulent material from the incision, radiographic evidence of osteomyelitis in addition to clinical signs, ultrasonographic evidence of infection or the development of a fever as previously described. ³ , ⁶ , ¹² , ¹⁸ Only cases undergoing internal fixation of injuries of the appendicular limbs were included. Any cases that had infection present at the surgical site prior to the implantation of internal implants were excluded from analysis. Open fractures were considered to be contaminated not infected and therefore were included in the study. Cases were only excluded if there was evidence of active infection including gross purulence and/or osteomyelitis. Open fractures were classified as such based upon visual inspection and/or diagnostic imaging. Any cases that underwent limb amputation due to complications of SSI, failure of fracture healing or other reasons were noted. For non‐surviving cases the reason for euthanasia was recorded.

Data retrieved from medical records included patient signalment (species, breed, age, weight, sex), days from presentation to surgery, injury characteristics, the surgical repair method (open or closed reduction and internal fixation), orthopedic implants used (plates/screws, intramedullary [IM] pins, screws alone, interlocking nails, wires alone) and surgical duration. Any complications related to the fracture or fixation were noted. Systemic and local antimicrobial used was recorded which included the antimicrobial agents utilized, the length of time they were administered for and the method of local antimicrobial delivery. In cases that developed SSI, the time at which it was recognized postoperatively and the results of any culture and sensitivity testing were recorded. For all cases, survival to discharge from the hospital was recorded, and where available the long‐term survival was also noted in addition to any pertinent information about the case. Follow‐up was obtained by veterinarians from each institution by either phone consultation with the owners, or repeat examinations of the animal.

For the purposes of the study, short‐term survival was defined as survival to hospital discharge. Long‐term survival was defined as survival beyond discharge until a minimum of 6 months follow‐up. Possible short‐term outcomes included the presence of SSI, complications related to the fixation, complications unrelated to the initial injury or surgery, and survival or non‐survival. Long‐term outcomes included survival, the presence of SSI, complications related to the fixation and complications unrelated to the fixation.

2.1. Statistical analysis

Analysis was performed using R software (R Foundation for Statistical Computing, Vienna, Austria). Three outcomes; the presence of SSI, short‐term survival (STS), and long‐term survival (LTS) were assessed. For categorical predictors, counts and proportions were calculated. For continuous predictors, means and SD were calculated; however, where skewed data was present, the median was also presented.

For SSI and STS, univariate logistic regression models were utilized. Each variable's contribution to the risk of SSI or STS was assessed using likelihood ratio tests (LRTs). The p‐value of the LRTs is reported in the results as it relates to the specific variable being tested. Variables of interest in the univariate model included hospital (5 hospitals included), age (continuous in years), sex (male or female), species (ovine, caprine, bovine, other), diagnosis (fracture or other), limb location (front or hindlimb), location of injury on the limb (proximal to the carpus/tarsus or distal including carpus/tarsus), surgical repair method (ORIF or other), primary implant (plate, screw, wire, interlocking nail or IM pin), surgery time (continuous in minutes), days between presentation and surgery (continuous), duration of antimicrobial use (continuous in days), local antimicrobial use (yes/no), open/closed injuries (open or closed).

Variable selection for the multivariate model was performed by the following procedure. Predictors were initially screened using the LRTs from the univariate model. Predictors with a LRT p < .10 were moved forward and considered in the multiple logistic regression model. Backwards elimination (p < .05) was used for final model selection. For the SSI model, the hospital variable was included as a blocking variable (regardless of p‐value) to help account for variability between institutions. However, one hospital was missing data relating to the antimicrobial duration, leading to a sample size of 67 for the SSI model, rather than the full 81 cases. For STS hospital was excluded from the model due to low event numbers. SSI was considered as a predictor for modeling and variable selection purposes. Due to the low variety in responses for primary implant type against STS, they were regrouped in the regression into either “pin” or “other”. For LTS, summary statistics were evaluated and reported. A Fisher's exact test was performed to determine the effect of SSI on long term survival. A p‐value of <.05 was considered statistically significant.

3. RESULTS

A total of 81 cases that met the inclusion criteria were identified from the medical records. One further case was identified that had fracture fixation with orthopedic implants; however, this case was excluded due to the presence of chronic infection and osteomyelitis prior to surgical fixation. SSI developed in 14 of the 81 cases (14/81, 17.3%), with a mean time between surgery and diagnosis of 34 ± 29.2 days and a median time of 21 days. Overall, 75 of 81 cases survived to discharge (75/81, 92.6%), which included 12 of the 14 cases with a SSI (12/14, 85.7%) and 63 of the 67 cases that did not develop a SSI (63/67, 94%). From the six cases that did not survive to discharge; two were euthanized for reasons unrelated to the fracture, one was euthanized due to complications associated with SSI, one was euthanized due to failure of fracture fixation with concurrent SSI and two cases were euthanized due to failure of the fracture fixation in the absence of infection. Of the 14 ruminants that developed SSI, amputation of the affected limb was performed in two cases (2/14, 14.2%). In cases where SSI was not identified, limb amputation was performed in one case (1/67, 1.6%) due to a nonunion at the fracture site.

3.1. Demographics

The average age of ruminants presenting for orthopedic injury requiring internal fixation was 1.64 ± 2.3 years old. Ruminants that developed SSI were on average 1.98 ± 2.4 years of age, the animals without SSI were 1.44 ± 1.94 years of age (p = .38). A total of 51 of the ruminants were female (51/81, 63%), 30 were male (30/81, 37%). Of the ruminants that developed SSI, seven were female (7/14, 50%) and seven were male (7/14, 50%), there was no significant differences between sex in respect to the development of SSI (p = .385).

The study included 38 bovines (38/81, 46%), 33 caprines (33/81, 41%), nine ovines (9/81, 11%), one reindeer (1/81, 1%), and one antelope (1/81, 1%). There were no significant differences between species in respect to the development of SSI (p = .233).

3.2. Injury characteristics

Of the 81 cases assessed in the study, 59 injuries occurred in hindlimbs (59/81, 72.8%) and 22 occurred in forelimbs (22/81, 27.1%). SSI occurred in eight hindlimbs (8/14, 57.1%) and six forelimbs (6/14, 42.9%). The proportion of animals developing SSI did not differ by injury location (hind vs. front limb) (p = .16). A total of 59 internal fixation procedures (59/81, 72.8%) occurred in the proximal limb, and 22 (22/81, 27.1%) occurred in the distal limb (Table S1). Of the cases that developed SSI, there were seven involving the proximal limb (7/14, 50%) and seven involving the distal limb (7/14, 50%). The increased risk of animals developing SSI in the distal limb was significant (p = .043).

A total of 73 cases had fractures and eight had other reasons for internal fixation including luxation or subluxation (n = 4), arthrogryposis (n = 3), and arthrodesis due to severe degenerative joint disease in one case. Four of the fractures were classified as open (4/73, 5.5%) and the remainder were classified as closed (69/73, 94.5%). Open fractures included two comminuted mid‐diaphyseal fractures of the right metatarsus, a long oblique mid‐diaphyseal fracture of the left tibia and a comminuted right humeral fracture. These occurred in three caprine and one ovine. No SSI was reported in the cases with open fractures.

Plates and screws were used as the primary fixation in 40 cases (40/81, 49.4%), intramedullary (IM) pins were used in 20 cases (20/81, 24.7%), screws alone were used in 13 cases (13/81, 16%), interlocking nails in five cases (5/81, 6.2%), and wires alone in three cases (3/81, 3.7%). In the 14 cases that developed SSI, seven used plates and screws (7/14, 50%), three used IM pins (3/14, 21.4%), two used screws alone (2/14, 14.3%) and two used interlocking nails (2/14, 14.3%). Table 1 outlines the cases organized by implant type used. There was no significant difference between implant used and the development of SSI (p = .613).

TABLE 1.

Summary of case information relating to all 81 cases included in the study and the outcomes of these cases organized by implant type.

Implant type	Total cases	SSI	Isolates identified	Open injury	Closed injury	ORIF	CRIF	Systemic antimicrobials	Local antimicrobials	Euthanized due to SSI	Euthanized not due to SSI	Amputation performed
Plate	40	7	E. faecalis (1), coagulase negative Staphylococcus (1), E. coli (1), T. Pyogenes (1)	2	38	40	0	40	12	1	2	2
IM pin	20	3	n/a	1	19	16	4	19	0	1	6	0
Screws	13	2	n/a	0	13	10	3	13	2	1	1	1
Wire	3	0	n/a	1	2	3	0	3	1	0	1	0
Interlocking nail	5	2	E. faecalis (2), alpha hemolytic streptococcus (1)	0	5	5	0	5	0	2	0	0
Total:	81	14		4	77	74	7	80	15	5	10	3

Abbreviations: CRIF, closed reduction and internal fixation; E. coli, Escherichia coli; E. faecalis, Enterococcus faecalis; IM pin, intramedullary pin; ORIF, open reduction and internal fixation; SSI, surgical site infection; T. pyogenes, Trueperella Pyogenes.

3.3. Antimicrobial therapy

Prophylactic systemic antimicrobials were used in 80 of 81 cases (80/81, 99%). The most common antimicrobial used was ceftiofur (including ceftiofur sodium, ceftiofur hydrochloride and ceftiofur crystalline free acid) in 18 cases (18/80, 22%), followed by a combination of penicillin (including potassium penicillin G and procaine penicillin G)/ceftiofur in 13 cases (13/80, 16%) and florfenicol in 12 cases (12/80, 15%). Data on the prophylactic antimicrobials used in all cases is displayed in Table 2. It is important to note that the prophylactic antimicrobial use reported in this study is not a clinical recommendation, antimicrobials classified as critically important were extensively utilized in this cohort. ¹⁹ Data on the duration of antimicrobial use were available for 67 cases. The mean duration of antimicrobial use was 10.93 ± 12.88 days.

TABLE 2.

Prophylactic systemic antimicrobial use.

Antimicrobial type	Total cases	Bovine	Ovine	Caprine	Other	SSI	Bacterial isolates
Ceftiofur a	18	3		15		2	Staphylococcus spp.
Penicillin b /ceftiofur a	13	10		3		3	E. faecalis; streptococcus spp.
Florfenicol	12	4	3	5		3
Florfenicol/penicillin b	8	5	2	1		1
Florfenicol/ampicillin c	6	4		2		1
Penicillin b	5	1	2	2
Ampicillin c	3	3				1
Ceftiofur a /sulfadimethoxine	3	1		2
Ceftiofur a /florfenicol	3	2	1			2	E. faecalis
Ceftiofur a /tulathromycin	2	1		1
Ceftiofur a /ampicillin c	2	1		1
Ceftiofur a /gentamicin	1	1
Tulathromycin/penicillin b	1	1
Florfenicol/sulfadimethoxine	1			1
Penicillin b /gentamicin	1				1	1	E. coli; T. pyogenes
Florfenicol/tulathromycin	1		1
Florfenicol/ampicillin c /tulathromycin	1	1
None	1				1

Abbreviations: E. faecalis, Enterococcus faecalis; SSI, surgical site infection; T. pyogenes, Trueperella pyogenes.

^a Including ceftiofur sodium, ceftiofur hydrochloride and ceftiofur crystalline free acid.

^b Including potassium penicillin G and procaine penicillin G.

^c Including ampicillin sodium and ampicillin trihydrate.

Local antimicrobial therapy was used prophylactically in 15 cases, with some cases receiving several different methods of local antimicrobial therapy. Prophylactic antimicrobial use included antibiotic irrigation solution in seven cases (lincomycin in 4 cases, ampicillin sodium/polymyxin B/neomycin in 3 cases), regional limb perfusions in five cases (ceftiofur in 3 cases, imipenem in 1 case and florfenicol in 1 case), placement of antimicrobial impregnated beads in two cases (amikacin in 1 and ampicillin/sulbactam in 1), placement of bone cement containing amikacin at the time of surgery in one case and local infiltration of the surgical site with amikacin in one case.

3.4. Bacterial identification

Out of the 14 cases in which surgical site infection was diagnosed, bacterial culture and sensitivity were performed in five cases. The species in which cultures were performed included two bovines, two ovines and one caprine. Enterococcus faecalis was cultured in two cases, Enterococcus faecalis and alpha hemolytic streptococcus were cultured in one case, coagulase negative Staphylococcus and Escherichia coli/Trueperella Pyogenes were each cultured in one case.

3.5. Logistic regression

3.5.1. Surgical site infection

An overview of the SSI logistic regression model is shown in Table 3. This model included 67 cases due to missing information on antimicrobial duration from one hospital. The limb location (proximal vs. distal) and duration of systemic antimicrobial use were found to be significant predictors of SSI at the p < .1 level for the univariate regressions and therefore moved onto the backwards selection. They both were found to be significant predictors of SSI at the p < .05 level in the multivariate model. Longer antimicrobial duration was associated with increased odds of SSI (OR 1.187, likelihood ratio test‐statistic = 2.87, p = .004); meaning that a one day increase in the duration of antimicrobial administration was associated with a 1.187 multiplicative increase in the odds of SSI, controlling for other variables in the model. For limb location there was an estimated odds ratio of 0.148 (likelihood ratio test statistic = −2.02, p = .04). The odds of SSI in the proximal limb is estimated to be 0.148 times the odds for distal limbs, controlling for other variables in the model. Therefore, proximal limb fractures had lower chance of developing SSI when compared to distal limb fractures.

TABLE 3.

Overview of the SSI logistic regression model.

Variable	p‐value
	SSI	STS
Age	.383	.160
Antimicrobial duration	.042 b	.885
Antimicrobial local use	.308	.109
Days between surgery and presentation	.619	.095 a
Fracture vs. other	.561	.254
Open vs. closed fracture	.212	.427
Hospital	.982	.170
Limb	.161	.529
Proximal vs. distal location	.043 b	.046 a
ORIF vs. CRIF	.287	.199
Primary implant	.613	.004 b
Sex	.385	.693
Species	.233	.268
Surgery time	.526	.129

Abbreviations: CRIF, closed reduction and internal fixation; ORIF, open reduction and internal fixation; SSI, surgical site implant; STS, short‐term survival.

^a Denotes a variable moved on to backwards selection but was eliminated.

^b Denotes a variable moved on to backwards selection and was retained.

3.5.2. Short‐term survival

The STS model used a total sample size of 81 cases including the same variables as the SSI model. An overview of the STS logistic regression is found in Table 3. The limb location, primary implant, and days between presentation and surgery were found to be significant at the p < .1 level for the univariate regressions and moved onto the backwards selection. Having a pin as a primary implant was associated with decreased odds of survival to discharge (OR 0.05, likelihood ratio test‐statistic = −2.64, p = .008); meaning that the expected odds of survival to discharge for a pin is .05 times the odds for any other implant.

3.5.3. Long‐term survival

A total of 46 cases were alive at the time of long‐term follow‐up (61%), 20 cases were lost to follow‐up (26%) and nine cases had been euthanized (12%). Cases were lost to follow‐up due to loss of contact with the owner or the animal being sold. Time between surgery and follow‐up ranged from 1 year to 15 years. Of the animals euthanized following discharge, five had a SSI and four did not have a SSI. Of the five cases with SSI, three were euthanized due to progression of the SSI with the presence of osteomyelitis, one was euthanized due to severe angular limb deformities and one was euthanized due to aspiration during recovery from limb amputation. These animals were euthanized on average 6 weeks following discharge. Of the four cases without SSI, three cases were euthanized due to gastrointestinal disease, one due to fracture fixation failure. The development of SSI in the postoperative period was associated with reduced long‐term survival (p = .035).

4. DISCUSSION

This retrospective study found that ruminants undergoing internal fixation for orthopedic injury had a short‐term survival rate of 92.6%, and an overall SSI prevalence of 17.3%. These findings were contrary to our hypothesis that the prevalence of SSI in ruminants following internal fixation would be higher than that reported in equine patients. Additionally, we did not find that open fractures or longer surgical times were risk factors for the development of SSI in this population. However, it is possible the study was underpowered to find this difference due to the relatively low number of open fractures. Studies investigating SSI in livestock species are limited, and no previous studies evaluating SSI prevalence or risk factors in ruminants undergoing internal fixation has been published to the authors' knowledge. The results of the current study are comparable to previous studies in the equine literature assessing SSI following internal fixation of long bone fractures in which SSI infection incidence varies from 10% to 28%. ³ , ¹² , ²⁰ We initially hypothesized the incidence of SSI would be higher in ruminants when compared to equines as a reflection of a widespread belief among veterinarians in the authors' practice area. Possible reasons for this include the challenge in controlling the environment in which ruminants live, increasing the risk of environmental contamination and consequential SSI; as well as the possibility for transient subclinical bacteremia from translocation of ruminal microflora. Unlike in the previous equine studies, where patients with a SSI were significantly less likely to survive to discharge, ³ the presence of SSI was not found to have a statistically significant effect on survival to discharge in this study. However, due to the relatively low case numbers analyzed in this study, it is possible the study was underpowered to determine a significant relationship between SSI and short‐term survival. Potential hypotheses for the higher survival of ruminants compared to horses with SSI are ruminants improved ability to ambulate on three limbs and their increased time spent in recumbency. This makes them less likely to succumb to complications such as support limb laminitis, and also cycle implants or osteolytic bone less during the convalescent period. The presence of SSI had a significant effect on the long‐term survival in this population, highlighting the need to improve practice at both preventing and treatment this complication.

There are several disadvantages to internal fixation in food producing animals, including the placement of hardware in animals intended for food production. This makes the operated region inappropriate for consumption and unsafe for slaughterhouse workers if not alerted to the implants. The cost of internal fixation is also often significantly increased when compared to more conservative management practices. As a result of these disadvantages, more conservative management practices have been used more commonly and several studies exist on the outcomes of less invasive repair methods for management of fractures in ruminant patients, including external coaptation and transfixation pin casting. Furthermore, various external skeletal fixator devices have been developed for use in livestock species. ²¹ , ²² , ²³ Reported survival rates where external coaptation is used alone range from 80% to 94.1%. ¹⁷ , ²⁴ In ruminants, transfixation pin casts for fracture fixation have been investigated, with an overall short term survival rate to pin cast removal of 79%. ²² A total of 50% of the non‐surviving animals in this study were euthanized due to osteomyelitis and a further 15% of the surviving cases developed a SSI. ²² While reported outcomes from these less invasive treatment strategies are comparable to those seen in the current study, the fracture configurations amenable to these repair methods are also more limited. The information on the outcomes of internal fixation provided in this study continues to add to the available literature to optimize decision‐making on the best treatment strategy for orthopedic disease in ruminants.

In the current study, 24.7% of cases were repaired with IM pins and the use of this implant type was associated with a significant reduction in the odds of survival to discharge. In orthopedic surgery of other species, the use of the IM pin has steadily declined with the availability of newer implants and devices. However, their use remains popular in ruminant surgery due to reduced costs, ease of removal in case of infection and as an option in juvenile animals with a low cortical density. A review article by Nuss analyzed the use of various implants and repair techniques for a range of specific fracture configurations. They found that there may be benefits to the use of intramedullary pins for fixation of humeral fractures in calves and mature bovines; however, in most situations these were outperformed by plate osteosynthesis. ²⁵ Additionally, the review raises the concerns about the lack of consideration surrounding the disruption of the intramedullary blood supply upon placement of the intramedullary pin, and the potential consequences this may have. ²⁵ While the current study cannot conclusively state that IM pins alone cause the reduction in the odds of survival due to the presence of confounding factors, the high complication rate in the current study suggests that more information is needed when recommending the use of IM pins. Their use should be carefully considered especially when this method is chosen over other repair options.

In the current study, fractures of the distal limb were found to have a significantly higher incidence of SSI than fractures of the proximal limb. We hypothesized that the reduced soft tissue coverage and blood supply to the distal limb may result in an increased risk of infection due to a reduction in the concentration of antimicrobials at the surgical site, as well as reducing the local immune response of the patient. Furthermore, a reduction in blood supply to the distal limb is more likely to promote a hypoxic environment, delaying healing and increasing the risk of infection. A higher incidence of distal limb SSI may also be due to the proximity of the surgical site to the ground and an increased risk of environmental contaminants. Further investigation is warranted to justify these hypotheses. Additionally, many simple distal limb fractures are treated with casts or transfixation casts in ruminants. Therefore, our selection criteria of internal fixation could have inadvertently selected the more severe distal limb fractures with comminution and/or open fractures, which may have contributed to the relatively higher proportion of SSI in distal limb fractures.

Open fractures are typically associated with a higher incidence of infection leading to delayed healing and increased costs associated with treatment. ²⁶ , ²⁷ In the current study, we did not find a difference between SSI in open compared to closed fractures. This is likely due to the relatively small number of open fractures compared to closed fractures leading to the study being underpowered to determine this difference. Additionally, the open fractures were not graded, so it is possible that these fractures were all of a low grade. However, a previous study assessing equine fractures similarly identified no difference in SSI in horses with open fractures compared with closed fractures. ³ The four animals with open fractures were included in the analysis of the current study as we considered them to be contaminated and not infected at the time of presentation. Additionally, we thought it important to investigate the outcome of these cases to determine if good outcomes can be obtained in the face of an open contaminated fracture. While no firm conclusion can be drawn regarding the association of open fractures and SSI in the current study due to the limited number of open fractures, it is important to note that successful repair of open fractures in select cases is possible.

In the current study, a prolonged duration of antibiotic treatment was associated with an increased risk of SSI. It is highly plausible that this is due to reverse causation. Cases at risk of developing SSI may be identified early in the postoperative period, due to the presence of specific risk factors, or the development of either non‐specific or overt clinical signs of infection, meaning that antibiotic therapy was continued. Therefore, the prolonged antimicrobial use was a response to SSI and not a cause. Alternatively, it is possible that an extended period of antimicrobial use had a negative impact on the gastrointestinal microbiome, influencing the likelihood of infection with enteric bacteria, which were predominantly isolated in the limited cases where this was performed. While these theories are purely speculative, the negative impact of systemic antimicrobial drugs on the microbiome is well‐established, while the potential detrimental consequences this disruption may cause to these patients remains unknown. ²⁸ , ²⁹ , ³⁰ The average number of days antimicrobials were continued for postoperatively was 10.93 days. In both humans and horses, the current recommendations for prophylactic antimicrobial use in orthopedic procedures involves administering a broad‐spectrum antimicrobial 30–60 min prior to the initial incision, with antimicrobial administration beyond 24 h deemed unnecessary. ³¹ , ³² , ³³ However, there are few evidence‐based guidelines for antimicrobial use in open contaminated fractures, or for prophylactic use in ruminants.

Wide variation in the choice of antimicrobial drugs utilized by clinicians existed in this study. Increasingly strict guidelines on the use of antimicrobial drugs in food producing animals are used globally in an effort to reduce the development of antimicrobial resistance, and the World Health Organization (WHO) categorizes antimicrobial classes based on their medical importance to human medicine and the risk of development of antimicrobial resistance. In the United States, the USA Food and Drug Administration (FDA) is responsible for approving antimicrobials for use in food animals, and guidelines for extra‐label drug use are provided by the Food Animal Residue Avoidance Databank (FARAD). Judicious use of antimicrobial drugs used in this study was frequently lacking when you consider the laws and guidelines provided by these organizations and several of the drug products used prophylactically in these cases are considered a last resort or sole therapy for treating serious multidrug‐resistant infections in humans. Ceftiofur was by far the most common antimicrobial reported to be used in the current study and is classified by WHO as a highest priority critically important antimicrobial for use in humans. ¹⁹ This finding highlights the importance and need for continued improvement in antimicrobial use in livestock. The use of critically important antimicrobials as prophylactic medications must be firmly discouraged, it undermines the effectiveness of these antimicrobials in human medicine and poses a significant problem in the growing threat antimicrobial resistance poses to global health. The wide variation in antimicrobial protocols used in the current study, is in part indicative of the challenges faced by veterinarians performing high risk specialty surgery on animals considered to be food producing species, but who are used as companion animals with a high emotional as well as economical value. However, even with these challenges, significant room for improvement exists in the development of appropriate prophylactic protocols in these cases and further research to identify the optimal use and duration of antimicrobials following fracture fixation is required, including the use of local antimicrobials.

The limitations of this study are inherent to the retrospective nature of the data collection. These limitations include the risk of unmeasured confounding factors affecting the results in addition to reliance on historic medical records. Additionally, one hospital was missing data relating to the duration of antimicrobial use meaning that these cases were excluded from the multivariate model. While this is the largest retrospective study looking at fractures in livestock, there are still limited numbers of cases developing SSI which may have resulted in a lack of power when assessing certain variables. This lack of power was particularly apparent when assessing open fractures, since only four were included in the study. Additionally, the information gathered from cases was limited to specific variables in order to maintain consistency between the centers, this could have meant that information was missed if it did not fit into one of the analyzed categories. Finally, the duration of the study period made long‐term follow up challenging. It was difficult to contact owners that had moved away during the study period, and also hard to trace animals that were sold at the beginning of the study period. This meant that the long‐term follow up data was highly variable which precluded the use of formal statistical analysis for all variables. The long‐term data was therefore limited to reporting descriptive statistics and analysis of the association between SSI and LTS.

5. CONCLUSION

In conclusion, the SSI incidence in ruminants following internal fixation of fractures was comparable to the infection risk in horses and camelids. We associated a higher SSI risk in distal limb fractures and animals that were administered antibiotics for a prolonged period. Fractures stabilized utilizing intramedullary pins had a significant reduction in the odds of survival to discharge.

AUTHOR CONTRIBUTIONS

Barton CK, BVetMed, MS, DACVS (Large Animal): Contributed to the design of the work, acquisition, analysis and interpretation of data, drafting and revision of the manuscript. Lozier JW, DVM, MS, DACVS (Large Animal): Contributed to the acquisition of data and revision of the manuscript. Merkatoris PT, DVM, DACVS (Large Animal): Contributed to the acquisition of data and revision of the manuscript. Jordan B, DVM, MS: Contributed to the acquisition of data and revision of the manuscript. McCormack JO, MVB, DACVS (Large Animal): Contributed to the acquisition of data and revision of the manuscript. Vanhoy G, DVM, MS, DACVIM (Large Animal): Contributed to the acquisition of data and revision of the manuscript. Look B: Contributed to analysis and interpretations of data, revision of manuscript. Nottle B, BVSc, DACVS (Large Animal): Contributed to the design of the work, acquisition, analysis and interpretation of data, revision of the manuscript. All authors provided critical review of the manuscript and endorse the final version. All authors are aware of their respective contributions and have confidence in the integrity of all contributions.

FUNDING INFORMATION

The project did not have any funding.

CONFLICT OF INTEREST STATEMENT

There are no conflicts of interest to report.

Supporting information

Supplementary Table S1: The proportions of SSI when assessed by proximal versus distal limb location in 81 cases.

Supplementary Table S2: Table to show continuous variables within the SSI model.

Barton CK, Lozier JW, Merkatoris PT, et al. Incidence and risk factors of surgical site infection in ruminant species following internal fixation for orthopedic injury: 81 cases (2010–2023). Veterinary Surgery. 2025;54(8):1520‐1529. doi: 10.1111/vsu.70029