Proximal interphalangeal joint arthrodesis is indicated for horses that have lameness as a result of osteoarthritis or instability of the PIPJ or P2 fracture caused by trauma. Several methods for PIPJ arthrodesis have been described in the literature, including transarticular screws placed in lag fashion, plate-screw combinations, laser-facilitated ankylosis with transarticular screws placed in lag fashion, and chemically induced ankylosis.1–8 Use of an LCP in conjunction with locking head screws for internal fracture fixation was first described in the human medical literature.9 The LCP was introduced as an alternative to LC-DCPs and DCPs, with purported benefits in improved biomechanical strength, compared with the latter devices.5
Studies10–12 have compared the biomechanical properties of LCPs with those of LC-DCPs and DCPs in vitro. Results of these studies10–12 indicate that the mechanical properties of LCPs matched or exceeded those of DCPs and LC-DCPs when used with 2 transarticular cortex screws for arthrodesis of the PIPJ, except that under axial compression in single cycle-to-failure testing, DCPs had greater stability.10
Locking compression plates are placed with screws that lock directly into the plate and reduce the chance for loss of fracture reduction secondary to implant loosening, making the construct less dependent on bone-to-plate contact.9,13 Currently a 3-hole, 4.5-mm, narrow LCP specifically designed for equine PIPJ arthrodesis is commercially available. The first report5 describing LCP placement in horses was a clinical case series that included 8 patients (1 with a comminuted P2 fracture, 1 with PIPJ subluxation, and 6 with degenerative joint disease) treated by PIPJ arthrodesis. Seven of those horses were considered sound or pasture sound at the time of follow-up, and 1 was participating in horse shows. A more recent retrospective study14 to evaluate the effects of body type and performance discipline on outcomes following PIPJ arthrodesis included 19 horses that underwent arthrodesis with LCP placement; 14 had osteoarthritis, 3 had P2 fractures, and 2 had subluxations of the PIPJ. Fifteen of the 19 horses had successful outcomes, defined as being sound for performance, competition, or other intended use, in that study.14
Double-plate fixation of an unstable PIPJ that included LCP placement was described in another retrospective case series of 30 horses with 31 PIPJ injuries.8 In that study,8 2 narrow DCPs were used for repair in 26 cases, the combination of a DCP and LCP was used in 4 cases, and 2 LCPs were used in 1 case of PIPJ injury. However, outcomes for horses that were treated with an LCP were not separately reported. To the best of the authors' knowledge, no clinical case series assessing outcomes specifically in regard to PIPJ arthrodesis with LCPs has been published. The objective of the study reported here was to describe the clinical use of LCPs as a means of fixation for horses undergoing PIPJ arthrodesis and to compare outcomes for these horses with those described in previous reports2,4,5,7 that had arthrodesis performed by other means.
Materials and Methods
Case selection
Electronic medical records of all horses that underwent PIPJ arthrodesis between January 1, 2008, and December 31, 2014, at the Colorado State University Veterinary Teaching Hospital and the University of California-Davis Veterinary Medical Teaching Hospital were reviewed. Search terms for the electronic database included “arthrodesis,” “middle phalanx,” “locking compression plate,” and “fracture.” To be included in the study, horses had to have undergone PIPJ arthrodesis of a single limb with placement of an LCPa within the predetermined study period. Criteria for inclusion were a minimum 1-year follow-up period after surgery and complete medical records including the following information as a minimum: signalment, use, duration of clinical signs, diagnosis, complete surgical report, and details of postoperative care while hospitalized.
Medical records review
Information obtained from the medical records included signalment, performance discipline or activity level, limb affected, circumstances surrounding the injury (if applicable), duration of lameness, reason for arthrodesis, preoperative and postoperative diagnostic imaging reports, medical treatments, details of the surgical technique, surgical time, postoperative care, and complications.
One of the authors (RRS) conducted telephone interviews with horse owners ≥ 1 year after the date of surgery. A standardized telephone questionnaire, developed for this study, was used to collect information on the duration of rehabilitation, the horse's use or activity at the time of the follow-up, and owner satisfaction.
Procedures
All horses underwent preoperative radiographic imaging. At a minimum, a standard 4-view radiographic series of the affected limb (lateromedial, dorsopalmar or dorsoplantar, 45° dorsolateral-palmaromedial or plantaromedial oblique, and 45° dorsomedial-palmarolateral or plantarolateral oblique views) was obtained.
All patients were administered an NSAID (phenylbutazone, 4.4 mg/kg [2 mg/lb], IV) and a tetanus toxoid vaccine if indicated prior to surgery. Preoperative antimicrobial treatment was selected according to the surgeon's preference and consisted of penicillin G potassium (22,000 U/kg [10,000 U/lb], IV, q 6 h), penicillin G procaine (22,000 U/kg, IM, q 12 h), or cefazolin sodium (11 mg/kg [5 mg/lb], IV, q 12 h), given in combination with gentamicin sulfate (6.6 mg/kg [3 mg/lb], IV, q 24 h).
General anesthesia was induced and maintained with a protocol determined by the attending veterinary anesthesiologist. A standard approach to the PIPJ followed by internal fixation as previously described15 was used for all patients. In all cases, cartilage was removed by use of a curette or motorized burr followed by osteostixis with a 2.5-mm drill bit. Surgeon preference and patient size were determinants for plate size selection in each case. All plates were contoured as needed to improve bone-to-plate contact. The affected limb was bandaged and a cast was placed during the anesthetic episode, or a bandage and splint were applied and a cast was placed on the standing horse shortly after anesthetic recovery, according to the surgeon's discretion. Horses were monitored through recovery from anesthesia, and the contralateral limb was placed in a custom wooden clog, a supportive boot,b a therapeutic support shoe,c or a customized cuffed support shoe.c
All horses received a tapering dose of NSAIDs after surgery. The type and duration of this treatment and administration of additional analgesic treatment were determined by the clinician in charge of the case and were dependent on signs of patient discomfort. Postoperative radiographs were obtained at the time of the first cast change, and a second cast, bandage cast, or splint was then applied, depending on the apparent comfort level of the horse.
After surgery, horse owners were advised to maintain strict stall rest of patients for 2 months, followed by stall rest with hand-walking for an additional 2 months. Gradual return to exercise over a period of 60 to 90 days was recommended after the initial 4 months of rest.
Patients with suspected implant-associated infection were administered penicillin G procaine (22,000 U/kg, IM, q 12 h) and gentamicin sulfate (6.6 mg/kg, IV, q 24 h). Intravenous regional limb perfusions were performed with amikacin (2 to 4 mg/kg [0.9 to 1.8 mg/lb], IV, q 24 h) until lameness improved. As lameness improved, treatment was transitioned to doxycycline (10 mg/kg [4.5 mg/lb], PO, q 12 h) or minocycline (4 mg/kg, PO, q 12 h) for long-term treatment until sufficient healing had occurred and implants could be removed. Following implant removal or when results were available from culture of an intraoperative sample, systemic and local antimicrobial treatments were selected on the basis of culture and susceptibility testing.
Statistical analysis
For statistical analysis, horses were divided into 2 groups according to the reason for arthrodesis (fracture vs other). A nonparametric Mann-Whitney test was used to investigate potential associations between age and group. A Fisher exact test was used to compare categorical factors between groups, including sex, athletic outcome, presence of infection, and development of laminitis. Values of P < 0.05 were considered significant for all analyses.
Results
Horses
Twenty-nine horses met the study selection criteria. American Quarter Horses were the most commonly represented breed in the study population (17/29 [59%]). Other breeds represented were American Paint Horse (4 [14%]); Arabian (3 [10%]); and Saddlebred, Oldenburg, Appaloosa, Holsteiner, and Thoroughbred–Quarter Horse cross (1 [3%] each). There were 18 males (14 castrated and 4 sexually intact) and 11 females (all sexually intact). Fourteen horses underwent surgery because of traumatic fracture (all involving the P2), and 15 had the procedure performed for other reasons, including PIPJ osteoarthritis (n = 12 [including 3 with osteochondrosis of the first phalanx]), subluxation and osteoarthritis of the PIPJ (2), and chronic subluxation of the PIPJ (1). There was no significant (P = 0.71) difference in sex distribution between the 2 groups (OR, 1.5; 95% CI, 0.37 to 6.61). Median age of all horses at the time of surgery was 6 years (range, 1 to 25 years). Median age of patients with fracture was 5 years (range, 3 to 15 years). The median age of horses undergoing surgery for a reason other than fracture was 12 years (range, 1 to 25). There was no significant (P = 0.19) difference in age between groups.
In addition to the standard radiographic imaging, 6 horses underwent CT examination under general anesthesia immediately prior to surgery. All these horses had P2 fractures.
Arthrodesis was performed on 16 hind limbs (9 right and 7 left) and 13 forelimbs (9 right and 4 left). In cases of fracture, all horses had acute non–weight-bearing lameness at the time of injury and were examined by a veterinarian within 24 hours after onset of lameness was detected. Twelve of 14 horses with P2 fracture were referred to the study hospitals where repair was performed ≤ 24 hours after the onset of lameness, and the remaining 2 were managed conservatively for 3 and 5 months before referral for surgery.
Eighteen of the 29 horses were used for western-style riding, including roping (n = 5), reining (8), barrel racing (2), versatile ranch use (2), cutting work (1), or western pleasure riding (2). Nine horses were used for various other functions, including racing (n = 1), dressage (1), endurance riding (1), hunter-jumper (3), and trail riding, breeding, or other use (2). Some horses were used in > 1 discipline. Two horses (a 1-year-old American Paint and a 3-year-old Oldenburg) were considered juvenile and had not yet begun training at the time of surgery.
Fixation methods
For the 15 horses that had arthrodesis performed for reasons other than P2 fracture, a single LCP positioned on the dorsal midline was used in combination with 2 transarticular, 5.5-mm cortical screws placed in lag fashion (Figure 1). For 13 of these 15 horses, a 3-hole, 4.5-mm narrow LCP was used, and for the remaining 2, a 4-hole, 4.5-mm narrow LCP was used.
Postoperative radiographs (dorsopalmar and mediolateral views on the left and right side, respectively, of each panel) obtained following placement of LCPs in horses for arthrodesis of the PIPJ. A—A 4.5-mm, narrow, 3-hole LCP placed in a patient with PIPJ osteoarthritis. B—A 4.5-mm, narrow, 4-hole LCP placed as treatment for PIPJ osteoarthritis. C—A single 4.5-mm, narrow, 3-hole LCP placed in a patient with a uniaxial palmar eminence P2 fracture. D—Two 4.5-mm, narrow, 4-hole LCPs placed as treatment for a comminuted P2 fracture. Panels A, B, and D depict placement of 2 transarticular cortical screws in lag fashion outside of the LCP. In panels C and D, 1 or more cortical screws placed in lag fashion for fracture fixation outside of the LCPs are shown.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Postoperative radiographs (dorsopalmar and mediolateral views on the left and right side, respectively, of each panel) obtained following placement of LCPs in horses for arthrodesis of the PIPJ. A—A 4.5-mm, narrow, 3-hole LCP placed in a patient with PIPJ osteoarthritis. B—A 4.5-mm, narrow, 4-hole LCP placed as treatment for PIPJ osteoarthritis. C—A single 4.5-mm, narrow, 3-hole LCP placed in a patient with a uniaxial palmar eminence P2 fracture. D—Two 4.5-mm, narrow, 4-hole LCPs placed as treatment for a comminuted P2 fracture. Panels A, B, and D depict placement of 2 transarticular cortical screws in lag fashion outside of the LCP. In panels C and D, 1 or more cortical screws placed in lag fashion for fracture fixation outside of the LCPs are shown.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Postoperative radiographs (dorsopalmar and mediolateral views on the left and right side, respectively, of each panel) obtained following placement of LCPs in horses for arthrodesis of the PIPJ. A—A 4.5-mm, narrow, 3-hole LCP placed in a patient with PIPJ osteoarthritis. B—A 4.5-mm, narrow, 4-hole LCP placed as treatment for PIPJ osteoarthritis. C—A single 4.5-mm, narrow, 3-hole LCP placed in a patient with a uniaxial palmar eminence P2 fracture. D—Two 4.5-mm, narrow, 4-hole LCPs placed as treatment for a comminuted P2 fracture. Panels A, B, and D depict placement of 2 transarticular cortical screws in lag fashion outside of the LCP. In panels C and D, 1 or more cortical screws placed in lag fashion for fracture fixation outside of the LCPs are shown.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Of the 14 horses in which arthrodesis was performed for fracture fixation, 3 had a single 3-hole, 4.5-mm narrow LCP placed dorsally, and 10 had 2 dorsally located 4.5-mm narrow LCPs affixed (Figure 1). In the remaining horse, the fracture was repaired with a 4.5-mm narrow LCP and a 4.5-mm LC-DCP, both placed dorsally. Two cases of fracture repaired with a single dorsal plate had additional stabilization with two 5.5-mm transarticular cortical screws placed in lag fashion, and 1 case of fracture had a single 5.5-mm transarticular cortical screw placed in lag fashion. Eleven of 14 fracture repairs included the use of 1 or more 3.5- or 4.5-mm cortical screws placed in lag fashion outside the plate for fracture fragment stabilization.
Twenty-six of 29 horses had a half-limb cast placed on the affected limb during general anesthesia. The other 3 horses had sterile bandages and a distal limb splint placed at the time of surgery, followed by cast application immediately after recovery from surgery with the limb bearing weight. The contralateral limb was placed in a wooden clog (n = 6), a supportive boot (12), a therapeutic support shoe (8), or a customized commercial cuff shoe (3).
Postoperative care
Phenylbutazone (2.2 to 4.4 mg/kg [1 to 2 mg/lb], IV or PO, q 24 h) was the most commonly administered NSAID (29/29 horses) after surgery. Ten horses (6 with P2 fractures and 4 that had the surgery for other reasons) received opioid treatment for up to 3 days after surgery; 2 were given butorphanol tartrate (0.02 mg/kg [0.009 mg/lb], IM) as needed, and 8 were given morphine sulfate (0.1 to 0.15 mg/kg [0.045 to 0.068 mg/lb], IM or IV) as needed. Two horses with a hind limb affected (a 4-year-old American Quarter Horse with a plantar eminence fracture of P2 and a 3-year-old American Quarter Horse with osteoarthritis of the PIPJ secondary to osteochondrosis at the distal aspect of the first phalanx) received preservative-free morphine sulfate by caudal epidural administration (0.1 mg/kg, once) ≤ 24 hours after surgery.
The median time to the first cast change was 20 days (range, 4 to 52 days), and times were similar between horses that had surgery because of P2 fracture (median, 21.5 days) or for reasons other than fracture (median, 18 days). Sutures were removed at the time of the first cast change or ≥ 14 days after surgery if the cast was removed prior to this time. Following radiography at the first cast change, 10 horses (7 that had fractures of the P2 and 3 that had surgery for other reasons) had a second cast placed prior to transitioning to a splint or bandage.
Complications
Eleven of 29 (38%) horses had complications directly related to the procedure. One horse developed facial nerve paralysis after recovery from anesthesia, and this condition resolved 3 days after surgery. Implant infection occurred in 5 horses (3 that had P2 fractures and 2 that had osteoarthritis of the PIPJ). All horses that developed infections had surgery at the same institution. Median surgical time for these 5 horses was 215 minutes (range, 147 to 255 minutes). Median surgical time for the 24 horses that did not develop infection was 202.5 minutes (range, 90 to 298 minutes). No association between the presence of infection and surgical time was identified. Bacterial culture results were positive for Corynebacterium spp on routine intraoperative culture of a sample for 1 of the 5 horses; 3 horses had positive culture results for Staphylococcus spp at the time of implant removal. The fifth horse had infection suspected on the basis of persistent lameness, radiographic evidence of excessive periosteal reaction around the implants, and a positive response (improvement in signs of lameness) to antimicrobial treatment. The latter 4 affected horses had all implants removed ≤ 4 months after surgery. Following the removal of implants used in 1 of these 4 horses, a methicillin-resistant Staphylococcus sp was isolated, and the patient was euthanized because of a poor prognosis for soundness as a result of osteomyelitis. There was no significant (P = 1.0) difference in infection rates between horses that had P2 fractures and those that had surgery for other reasons (OR, 1.6; 95% CI, 0.29 to 10.2).
After surgery, 2 horses that underwent PIPJ arthrodesis of a hind limb developed signs of laminitis in the contralateral limb with radiographic evidence of displacement of the distal phalanx within the hoof capsule. Another horse that had surgery on a forelimb developed severe persistent lameness of the contralateral limb without radiographic evidence of laminitis. Of the 2 horses with radiographic evidence of laminitis, 1 had undergone surgery immediately following an acute P2 fracture, and mild evidence of rotation was detected 6 days later when the horse began to have signs of lameness in the supporting limb. Radiographic changes stabilized, and the lameness had improved prior to discharge from the hospital. The horse was reevaluated at the study hospital 52 days after surgery for acute onset of lameness in the same limb and was subsequently euthanized owing to the severity of the displacement of the third phalanx within the hoof capsule. The other horse had a 6-month history of lameness at a walk secondary to PIPJ osteoarthritis at the time of referral, and laminitis of the contralateral limb was diagnosed by the referring veterinarian at a 34-day postoperative recheck examination; this horse was subsequently lost to follow-up. The horse that was lame in the contralateral limb without radiographic evidence of laminitis was euthanized at home 4 months after hospital discharge. There was no significant (P = 1.0) difference in the rates of postoperative laminitis between horses with fractures and those that had surgery for other reasons.
Implant removal was performed for 1 horse without evidence of infection 5 months after surgery because of persistent grade 3/5 lameness.16 Radiographs revealed complete fusion of the PIPJ, and there was no radiographic evidence of infection; the lameness was localized to the pastern region with perineural anesthesia and resolved after implant removal. One horse had persistent mild lameness after surgery and had a single screw replaced 63 days after the original procedure because of suspected impingement on the deep digital flexor tendon. The lameness in this horse resolved after the second surgery.
Implant failure occurred in 1 horse. A locking head screw in the distal hole of the plate was noted to be broken on radiographic examination 8 months after surgery (Figure 2). The last radiograph available prior to detection of implant failure had been obtained 5 weeks after surgery. The broken screw was not removed.
Lateromedial radiographs of the PIPJ (right forelimb) of an 8-year-old American Quarter Horse that underwent placement of a 3-hole, 4.5-mm LCP as treatment for subluxation and osteoarthritis. A—In an image obtained during follow-up examination 5 weeks after surgery, notice the appearance of the head of the distal locking head screw (black arrow), compared with that of the proximal locking head screw (white arrow), relative to the plate surface. B—In a similar radiographic view obtained 8 months after surgery, breakage of the distal locking head screw is evident.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Lateromedial radiographs of the PIPJ (right forelimb) of an 8-year-old American Quarter Horse that underwent placement of a 3-hole, 4.5-mm LCP as treatment for subluxation and osteoarthritis. A—In an image obtained during follow-up examination 5 weeks after surgery, notice the appearance of the head of the distal locking head screw (black arrow), compared with that of the proximal locking head screw (white arrow), relative to the plate surface. B—In a similar radiographic view obtained 8 months after surgery, breakage of the distal locking head screw is evident.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Lateromedial radiographs of the PIPJ (right forelimb) of an 8-year-old American Quarter Horse that underwent placement of a 3-hole, 4.5-mm LCP as treatment for subluxation and osteoarthritis. A—In an image obtained during follow-up examination 5 weeks after surgery, notice the appearance of the head of the distal locking head screw (black arrow), compared with that of the proximal locking head screw (white arrow), relative to the plate surface. B—In a similar radiographic view obtained 8 months after surgery, breakage of the distal locking head screw is evident.
Citation: Journal of the American Veterinary Medical Association 253, 11; 10.2460/javma.253.11.1460
Wound dehiscence occurred in 1 horse 29 days after surgery, partially exposing the surgical implant. The surgical site was managed as an open wound and subsequently developed a draining tract. The horse was treated with systemic antimicrobial administration as well as IV regional limb perfusions with amikacin (2 to 4 mg/kg) alone or with ticarcillin (14 mg/kg [6.4 mg/lb]). The horse did not develop radiographic evidence of implant failure or osteomyelitis, and the draining tract resolved prior to discharge from the hospital, but the patient was lost to further follow-up.
Outcome
Postoperative follow-up time for the 27 horses with information available ranged from < 1 month to 6 years. Three horses (2 that had surgery for fractures of the P2, and 1 that had surgery for other reasons [chronic subluxation of the PIPJ]) were euthanized because of previously described orthopedic complications; 1 had methicillin-resistant Staphylococcus sp infection and osteomyelitis, another had laminitis of the contralateral limb, and the third had persistent severe lameness of the contralateral limb without evidence of laminitis. One other horse that had surgery because of PIPJ osteoarthritis was euthanized after discharge from the hospital because of colic during the convalescent period. Of the remaining 25 horses, 2 (1 in each group) were lost to follow-up after hospital discharge.
At the time of follow up, 8 of 23 (35%) horses (1/11 treated for fracture of the P2 and 7/12 that had surgery for other reasons) had returned to the same or higher level of work, compared with that prior to surgery (including the 1 horse that had implant failure). Uses of these horses included all-around pleasure riding or hunting, dressage, roping work, general ranch use, reining, and breeding (a brood mare). Horses that had surgery for reasons other than fracture were significantly (P = 0.03) more likely to return to their previous level of work (OR, 13.0; 95% CI, 1.42 to 155.3). Ten (43%) horses had returned to work at a lower level (7 and 3 that had fractures of the P2 and surgery for other reasons, respectively). Four (2 in each group) horses were pasture sound or retired for reasons other than lameness, and 1 (treated for a P2 fracture) was retired with persistent lameness. There was no significant (P = 0.70) difference in the rate of return to a lower level of activity between the 2 groups (OR, 0.53; 95% CI, 0.13 to 2.55).
Of the 22 horses that were not lame at the time of follow-up, the median convalescent period as defined by a return to working under saddle or free exercise without restriction was 7 months (range, 4 to 21 months). When evaluated separately, the median convalescent period for horses that underwent arthrodesis for P2 fracture repair was 11 months (range, 4 to 21 months; n = 11), and that for horses that had the procedure for other reasons was 6.75 months (range, 4 to 15 months; 11).
Twenty-two of 24 owners stated they would have the procedure performed again under similar circumstances (this number excludes 3 horses euthanized as a direct result of surgical complications and 2 horses lost to follow-up). Ten of 11 owners of surviving horses treated for a P2 fracture and 12 of 13 owners of horses that had the surgery to treat other conditions stated they would have the procedure performed again.
Discussion
For the population of horses in the present study, use of an LCP for surgical arthrodesis of the PIPJ as treatment of P2 fracture or degenerative disease of the PIPJ was found to be an appropriate means of fixation in most cases. Of 29 horses enrolled, 4 were euthanized and 2 were lost to follow-up; 18 of the remaining 23 (78%) returned to their previous function, with 8 considered to perform at the same or higher level and 10 reported as performing at lower levels. Outcome following surgical arthrodesis of the PIPJ in horses has been reported in a number of studies,2,4,5,8,14 and success rates ranging from 4 of 1614 to 8 of 85 have been described. In those investigations, successful outcome was variably defined as owner satisfaction, return to soundness, and return to previous performance level or intended athletic use. Additional variations in breeds of horses, limb affected, their uses, and the fixation methods applied likely contribute to the differences in reported success rates among these studies.
In our retrospective case series, success was defined by owner satisfaction or return to intended use, similar to previous studies.2,4,5,14 Owner satisfaction was high, with 22 of 24 owners contacted stating they would opt for surgery again under similar circumstances. Of 23 surviving horses with follow-up available, 22 were sound for their intended use (including 4 horses that were retired for reasons other than lameness). If success was strictly defined as a return to the same or higher level of function, 8 of 27 horses for which data were available had a successful outcome (1/13 that had arthrodesis because of P2 fracture and 7/14 that had the procedure for other reasons). This result was lower than previously described2–5,7; however, there were several instances in which the owners did not attempt to return a horse to its previous use. Of the 10 horses working at a lower level compared with that before surgery, 7 had owners who reported they felt there was potential for increased athletic ability but that they did not attempt to pursue it.
Implant failure (breakage of the distal locking head screw) was noted in 1 horse between 5 weeks and 8 months after surgery. A lateromedial radiograph obtained immediately after surgery revealed that the locking head of the screw that ultimately failed was not exactly horizontal with the plate surface. Therefore, it was likely the screw was improperly threaded into the plate, resulting in abnormal cycling of the screw and subsequent breakage. Deviation from a perpendicular position has been shown to result in significant decreases in bending load-to-failure and push-out force.17 In this patient, the PIPJ was subluxated prior to surgery with rupture of the straight distal sesamoidean ligament, and this might have placed increased stress on the implants after surgery. An increase in implant stress when palmar or plantar support is decreased because of soft tissue injury has been previously described.18,19 In the case described in our study, the screw was not found to cause clinical lameness and was left in place, and the horse subsequently returned to the same or higher level of performance (roping work) as before surgery.
Implant infection was identified in 5 of 29 (17%) horses, with 1 subsequently euthanized. Two of the remaining 4 returned to their previous use with performance at the same or higher level, 1 was reportedly working at a lower level, and 1 was sound but retired at the time of follow-up. The overall infection rate in this case study was similar to that in a previous report4 of horses that had PIPJ arthrodesis. Three of the 14 horses with P2 fractures had implant-associated infection, and all had been treated at 1 institution. We believe this was possibly related to differences in rules regarding students and other personnel in place during this type of surgery at the different institutions. After the study, changes were made at that institution to limit the number and movement of people (including radiology technicians performing imaging procedures) in the operating room during such procedures.
Implant removal was performed for 1 horse that had no evidence of infection but was persistently lame 5 months after surgery. Radiographs revealed a stable arthrodesis, and the surgeon opted to remove the implant. At the time of follow-up 11 months after implant removal, the horse had returned to its previous use in team roping and was actively competing.
The most common method of fixation in horses that had arthrodesis for reasons other than P2 fracture (13/15) was placement of an axial, 3-hole, 4.5-mm narrow LCP designed specifically for PIPJ arthrodesis with 2 transarticular cortical screws placed in lag fashion. This particular LCP method was evaluated in vitro and found to result in a stiffer construct, compared with an LC-DCP method.11 Methods of fixation for fracture of the P2 in the present study were more variable and were often determined by fracture configuration and surgeon preference. In some cases, the ability to use cortical screws through dual-use holes of the LCP (which allow placement of standard bone screws in one part or threaded conical locking screws in the other) enabled the surgeon to achieve some fracture reduction, thereby reducing the number of cortical screws needed outside the plate. In 1 horse that had both an LCP and LC-DCP placed, the surgeon believed that the LC-DCP allowed better positioning to place plate screws in lag fashion across the fracture lines and reduce the overall number of implants used. Implants that allow variation of locking head screw insertion angle have not yet become widely available in equine orthopedics but would likely be useful in some cases.
Transarticular screw fixation with or without articular laser treatment, plate-screw combinations including use of a DCP or LC-DCP, and double-plate fixation in patients with an unstable PIPJ have all been successfully used to achieve joint fusion in horses.2,4,6,8,14 As an inherently more stable construct, the LCP should result in less bone reaction, morbidity, and time in external coaptation, compared with other methods, and thereby improve postoperative recovery times in horses. Decreased time in external coaptation, attributable to improved stability, was shown when plate-screw fixation was introduced and compared with transarticular screw methods.2,4 Rehabilitation time may not differ substantially when an LCP (vs an LC-DCP or DCP) is used in horses undergoing arthrodesis for reasons other than fracture. However, in patients with PIPJ instability or implant-associated infection, an LCP provides an important advantage over nonlocking implants. The rigid fixation of the implant imparts stability and, if infection develops, can allow surgeons to manage the infection until sufficient healing has occurred and the implants can be removed.
Limitations of the present study included its retrospective nature and the limited number of horses enrolled. To the author's knowledge, this report included the largest group of horses undergoing PIPJ arthrodesis by means of LCP placement and was the largest case series describing double locking-plate fixation for fractures of the P2 at the time of writing. Despite adherence of surgeons from both study institutions to AO (ie, Association for the Study of Internal Fixation) principles for fracture fixation, variations existed between hospitals in surgical technique as well as postoperative management, facilities, and equipment. The challenge of comparing the results from the present case series with those previously published was substantial, and we attempted to avoid direct comparisons with the other studies and, instead, describe the procedures and results in our equine patients.
The LCP system used in the present study requires technical expertise and understanding of the various nuances involved with proper application prior to its use. Despite results of a recent investigation14 that found no significant difference in performance outcomes between horses treated by use of an LCP versus other methods of arthrodesis, the authors of the present study believe that the benefits of the LCP in patients that have PIPJ instability or develop infection justify its use even where the financial cost is greater.
Acknowledgments
The authors have no sources of funding or financial interest to disclose. The authors declare that there were no conflicts of interest.
Presented in part as an abstract at the American College of Veterinary Surgeons Surgery Summit, Seattle, October 2016.
The authors thank Dr. Carrie J. Finno for assistance with statistical analysis and Meghan Monahan for assistance with data collection.
ABBREVIATIONS
| CI | Confidence interval |
| DCP | Dynamic compression plate |
| LC-DCP | Limited contact-dynamic compression plate |
| LCP | Locking compression plate |
| P2 | Middle phalanx |
| PIPJ | Proximal interphalangeal joint |
Footnotes
DePuy Synthes Vet, West Chester, Pa.
Soft-Ride Inc, Vermilion, Ohio.
NANRIC Inc, Lawrenceburg, Ky.
References
1. Schneider JE, Carnine BL, Guffy MM. Arthrodesis of the proximal interphalangeal joint in the horse: a surgical treatment for high ringbone. J Am Vet Med Assoc 1978;173:1364–1369.
2. Schaer TP, Bramlage LR, Embertson RM, et al. Proximal interphalangeal arthrodesis in 22 horses. Equine Vet J 2001;33:360–365.
3. Caston S, McClure S, Beug J, et al. Retrospective evaluation of facilitated pastern ankylosis using intra-articular ethanol injections: 34 cases (2006–2012). Equine Vet J 2013;45:442–447.
4. Knox PM, Watkins JP. Proximal interphalangeal joint arthrodesis using a combination plate-screw technique in 53 horses (1994–2003). Equine Vet J 2006;38:538–542.
5. Levine DG, Richardson DW. Clinical use of the locking compression plate (LCP) in horses: a retrospective study of 31 cases (2004–2006). Equine Vet J 2007;39:401–406.
6. Watts AE, Fortier LA, Nixon AJ, et al. A technique for laser-facilitated equine pastern arthrodesis using parallel screws inserted in lag fashion. Vet Surg 2010;39:244–253.
7. MacLellan KN, Crawford WH, MacDonald DG. Proximal interphalangeal joint arthrodesis in 34 horses using two parallel 5.5-mm cortical bone screws. Vet Surg 2001;30:454–459.
8. McCormick JD, Watkins JP. Double plate fixation for the management of proximal interphalangeal joint instability in 30 horses (1987–2015). Equine Vet J 2017;49:211–215.
9. Gautier E, Sommer C. Guidelines for the clinical application of the LCP. Injury 2003;34(suppl 2):B63–B76.
10. Sod GA, Riggs LM, Mitchell CF, et al. A mechanical comparison of equine proximal interphalangeal joint arthrodesis techniques: an axial locking compression plate and two abaxial transarticular cortical screws versus an axial dynamic compression plate and two abaxial transarticular cortical screws. Vet Surg 2011;40:571–578.
11. Ahern BJ, Showalter BL, Elliott DM, et al. In vitro biomechanical comparison of a 4.5 mm narrow locking compression plate construct versus a 4.5 mm limited contact dynamic compression plate construct for arthrodesis of the equine proximal interphalangeal joint. Vet Surg 2013;42:335–339.
12. Zoppa AL, Santoni B, Puttlitz CM, et al. Arthrodesis of the equine proximal interphalangeal joint: a biomechanical comparison of 3-hole 4.5 mm locking compression plate and 3-hole 4.5 mm narrow dynamic compression plate, with two transarticular 5.5 mm cortex screws. Vet Surg 2011;40:253–259.
13. Wagner M. General principles for the clinical use of the LCP. Injury 2003;34(suppl 2):B31–B42.
14. Herthel TD, Rick MC, Judy CE, et al. Retrospective analysis of factors associated with outcome of proximal interphalangeal joint arthrodesis in 82 horses including warmblood and Thoroughbred sport horses and Quarter Horses (1992–2014). Equine Vet J 2016;48:557–564.
15. Auer JA. Principles of fracture treatment. In: Auer JA, Stick JA, eds. Equine surgery. 4th ed. St Louis: Elsevier-Saunders, 2012;1047–1081.
16. Kester W. Definition and classification of lameness. In: Guide for veterinary services and judging of equestrian events. 4th ed. Lexington, Ky: American Association of Equine Practitioners, 1991;19.
17. Kääb MJ, Frenk A, Schmeling A, et al. Locked internal fixator: sensitivity of screw/plate stability to the correct insertion angle of the screw. J Orthop Trauma 2004;18:483–487.
18. Elce YA, Goodrich L. Deep digital flexor tendon rupture in two horses: a potential complication of comminuted second phalangeal fractures. Equine Vet Educ 2015;27:65–70.
19. Lescun TB. Equine fractures: the importance of the soft tissues. Equine Vet Educ 2015;27:71–74.
