Inclusion Criteria

Medical records from three private surgical practices (California Veterinary Specialists, Southern California Veterinary Surgery, and Inland Valley Veterinary Specialists) and one academic veterinary hospital (University of Missouri Veterinary Health Center) were searched to identify all dogs examined during the period from January 1, 2007, to June 30, 2017, with the diagnosis of coxofemoral joint luxation. Cases were included in the study if open reduction and toggle rod stabilization had been performed and at least 4 wk of follow-up data were available.

Procedures

Data collected from the medical records included patient signalment; body weight; history; results of initial diagnostic testing, including hematologic testing and radiographs; luxation cause and side; concurrent injuries; and if closed reduction had been attempted. Surgery-specific information obtained included surgical approach, toggle rod type and size, suture type used, and treatment of other injuries. Surgeries were performed by residents, a university residency–trained surgeon, or a board-certified surgeon. The duration of surgery was recorded and categorized as being less than or greater than 2 hr. The total anesthesia time was recorded and categorized as being less than or greater than 3 hr. The patient’s willingness to use the limb 24 hr postoperatively was recorded. Time from surgery to recheck examination was documented. Occurrence of complications was determined and categorized as either minor or major. Minor complications were considered those that resolved without the need for additional surgery (seroma, hematoma, bruising, incisional irritation, and chronic lameness). Major complications include those that required additional surgery or were perceived to potentially affect long-term outcome (surgical site infection leading to wound dehiscence, infection requiring removal of implants, and reluxation). For the purposes of this study, surgical success was defined by the following two criteria: (1) the patient not reluxating the affected hip by at least 4 wk postoperatively as determined by re-examination by a veterinarian; and (2) the patient not requiring additional surgical procedures for any major complication. The significance of those factors that Demko et al. described as being risk factors for either postoperative reluctance to bear weight or reluxation were assessed and included body weight, gender, surgery time, anesthesia time, and if multiple surgical procedures were done under a single anesthetic episode.¹²

Client Questionnaire

A survey of clients was completed, which was intended to identify other previously undetermined client-associated risk factors that may affect outcome as well as ascertain owner satisfaction with the procedure. Attempts were made to contact the owners of all patients by phone to request completion of an emailed questionnaire (Appendix A) in order to obtain additional information on short- and long-term outcomes, as well as overall client satisfaction. The questionnaire was modeled after one described by Zamprogno et al.³⁷ Questions pertained to nine specific functions and four behaviors that were designed to assess the degree of function and activity level. A five-point descriptive score was used for each function/behavior, where 0 = “unable to perform the function/behavior” and 5 = “function/behavior is normal.” In addition, owners were asked to indicate the severity of lameness upon presentation, at the time of discharge, at the time of postoperative re-examination (usually between 4 and 8 wk), and at the time of questionnaire completion if the patient was still living, using a scale from 0 to 5, where 0 = “no lameness” and 5 = “non–weight-bearing lameness.” Owners were also asked to describe the degree to which they had complied with postoperative discharge instructions, including confinement and rehabilitation, where 0 = “instructions were not followed, the patient was allowed immediate normal activity, rehabilitation not performed” and 5 = “discharge instructions were strictly followed and the owner performed or sought out rehabilitation.” Postoperative complications, as perceived by the owners, were also documented and categorized as either minor or major as previously defined. Additional questionnaire information included primary use or lifestyle of the patient (family pet, sporting, show, working, or other athletic use), any medications or special diet that the pet was on, and owner overall satisfaction with the outcome of surgery.

Statistical Analysis

Results from the client questionnaire were tabulated and percentages for each of the questions were calculated. Descriptive statistics were generated using a commercially available software program^a. Data were reported as mean ± standard deviation and range. A variety of statistical analyses were used to examine for the relationship between various patient- and client-associated factors and outcome. A Wilcoxon rank-sum test was used to examine the effect of body weight, surgery time, and anesthesia time on complications including reluxation. χ² analysis was used to determine the effect of gender, possessing multiple injuries requiring surgery, suture type (mono- versus multifilament), and weight-bearing status 24 hr postoperatively on reluxation. A Fisher exact test was used to examine the effect of duration from the time of injury to surgery on reluxation. A Spearman’s rank correlation was used to analyze the relationship in client survey answers and the occurrence of complications. Significance was set at P < .05.

Patient Data

Fifty-eight cases met the criteria for inclusion in the study. Mean age of all patients was 4.8 yr (range 8 mo to 13 yr). Nine of the patients were sexually intact males, 27 were castrated males, 1 was a sexually intact female, and 21 were spayed females. Mean body weight was 23 kg (range 4.1–50.9 kg). Breeds of dogs included 12 mixed-breed dogs, 7 Labrador retrievers, 4 border collies, and 3 German shepherd dogs; the remaining patients represented a variety of breeds. Trauma was the most common cause of the coxofemoral luxation, with motor vehicular accidents representing 66% of cases. In 14 cases, the cause of the coxofemoral luxation was unknown. The time interval between luxation and presentation to the hospital was <2 days for 36% of cases, between 3 and 7 days for 50%, and >7 days for 14% of cases. The right coxofemoral joint was affected in 48% of the patients, and the left coxofemoral joint was affected in 52%. Closed reduction was attempted in 24% of the dogs prior to open reduction and toggle rod stabilization. No patient underwent previous open surgical reduction prior to presentation to the respective hospitals surveyed.

Preoperative Diagnostic Imaging

Pelvic radiographs were obtained prior to surgery in all cases. In 30 of the 58 cases (52%), abnormalities in addition to the coxofemoral joint luxation were identified. The most common concurrent pathologies noted were a variety of fractures (45%), sacroiliac joint luxation (17%), and musculoskeletal soft tissue injures (9%) including disruption of single or multiple stifle ligaments and rupture of the prepubic tendon. Of the 26 dogs with concurrent fractures, 24 dogs had fractures of the appendicular skeleton, 10 had pelvic fractures, and 4 had fractures elsewhere. Pelvic limb fractures made up 50% of the total fractures and included the ipsilateral femur (6), tibia (1), and metatarsal bones (1) as well as the contralateral femur (2), tibia (2), and tarsal bones (1). The pelvis was involved in 13 cases (50%), with the ipsilateral pelvis (9 cases; 69%) occurring more often. Osteoarthritis of the luxated hip joint or both hips, consistent with canine hip dysplasia, was observed in 10% of cases. Thoracic radiographs were performed in 44 patients. Abnormalities were seen in 4 of the 44 patients (9%) and included pulmonary contusions (2), pneumothorax (1), and diaphragmatic hernia (1). Abdominal ultrasound was performed in 11 of the 58 patients. Abnormalities were seen in 4 of the 11 patients (4%) and included abdominal wall hernia (2), ruptured urinary bladder (1), prepubic tendon rupture (1), and diaphragmatic hernia (1). Computed tomography was performed in 3 patients. Abnormalities were seen in all 3 cases and included rupture of the urinary bladder (1), ruptured kidney (1), and abdominal hernia (1).

Surgery

A craniolateral approach to the coxofemoral joint was performed in 54 of the 58 cases. The surgical approach was unknown or not reported in 4 cases. Surgical time was <2 hr in 58% of the patients and >2 hr in 42% of cases. Mean and standard deviation of the dogs included in this study was 2.18 ± 1.08 hr. Surgical procedures under the same anesthesia as the completion of the toggle rod stabilization were performed in 47% of the cases, including fracture stabilization and correction of various soft tissue injuries. Surgeries were performed by a board-certified surgeon in 31 cases (54%), a university residency–trained surgeon in 18 cases (31%), and residents in the remaining 9 (15%) cases. Other hip joint stabilization techniques were used to augment the toggle rod stabilization at the discretion of the surgeon including the placement of an antirotational suture via bone tunnels (2) and the addition of stabilizing sutures secured with either bone screw or bone anchor (3).

The type of toggle rod used was recorded in all 58 cases. Commercially available toggle rods were used in 55 of the 58 patients in this study^b. A button that is part of a commercially available system was used in the other 3 cases^c. In all but 2 dogs, a single toggle rod was used in each coxofemoral joint. In two cases, a second toggle rod was placed after the first one was lost as a result of breakage or slippage of the suture. Of the 55 cases that received a toggle rod, 4.0, 3.2, and 2.7 mm rods were used in 28, 24, and 3 cases, respectively. Suture material used for the toggle rod stabilization procedure was monofilament nylon, #5 OrthoFiber^d, and polyblend FiberWire^e, in 66, 22, and 12% of cases, respectively. Among patients receiving monofilament nylon suture, 80 lb suture was used in 47%, 60 lb suture was used in 18%, and 40 lb suture was used in 34%. Various techniques were used to secure the suture lateral to the femoral bone tunnel. A polypropylene suture button was used most commonly (48), whereas other techniques included TightRope^f system buttons (5), titanium suture buttons (4), and an Ormrod^g button in 1 case. Immediate postoperative radiographs confirmed appropriate reduction in all cases.

Outcome

In 47 patients, a follow-up examination was performed at one of the participating hospitals, whereas in the remaining cases, recheck examinations were performed by the referring veterinarian. All cases had at least one follow-up examination performed between 4 and 8 wk after surgery at which time complications, if present, were recorded. During this time, postoperative complications were observed in 25 (43%) of the patients with nine different conditions observed and 5 cases having both minor and major complications occurring together. Complications were associated with reluxation in the present study (P < .001). Sixteen (28%) patients developed minor complications, which did not require additional surgery, and included primarily incisional issues (superficial irritation [6], seroma [5], partial dehiscence as a result of a skin suture or staple [4], and hematoma [1]). Muscle atrophy and chronic lameness was reported in one case each and concomitant with another minor complication. Two of the 16 patients with reported minor complications were diagnosed with subluxation of the operated hip diagnosed by an increase in joint space on radiography, neither of which required additional surgery. Radiography reports for these patients did not describe any signs indicating progression of osteoarthritis or evidence of tunnel widening. Major complications were reported in nine cases (15%). The most common major complication was reluxation, which occurred in 6 of the 58 patients (10%). One of the remaining three cases developed chronic lameness of the limb attributed to subluxation of the coxofemoral joint and required a femoral head and neck ostectomy. An additional case required further surgery on the contralateral tibia fracture that likely placed strain on the coxofemoral luxation hip, leading to subluxation and lameness; the third case required additional surgery to remove the suture as a result of a late, deep infection. Of the 6 patients with reluxation, none had radiographic evidence of underlying hip dysplasia. The cause of reluxation was attributed to failure of the suture in 5 patients and could not be determined in the remaining case. No reluxation in any patient occurred as a result of toggle rod failure. In 5 of the patients with reluxation, monofilament suture had been used and was recorded as having broken. FiberWire was used in the remaining reluxation case; however, the mode of failure was not recorded. Time from surgery to reluxation was between 2 and 8 wk in 5 of the patients, whereas in the remaining case, time to reluxation was not recorded. Five of the 6 patients with reluxation were bearing weight on the affected limb at the time of discharge.

Duration of luxation prior to surgical stabilization was not significantly related to reluxation (P = 1.0). Neither the existence of multiple injuries (P = 1.0) nor the performance of multiple surgical procedures under the same anesthetic episode were related to reluxation (P = .8). Patients whose surgery time was <2 hr were not significantly less likely to have a reluxation than those whose surgery time was >2 hr (P = 1.0). The weight-bearing status of the patient postoperatively was not significantly related to reluxation (P = .17). Neither owner compliance to discharge instructions nor any previous illness or injury were associated with reluxation (P = 1.0). It was noted that dogs for whom surgery times were ≤2 hr were less likely to bear weight on the affected limb 24 hr after surgery than those dogs for whom the surgery time was >2 hr, although no significance was noted (P = .16). The type of suture used (mono- versus multifilament) was not related to reluxation (P = .3).

Client Questionnaire

Client questionnaires were returned by 49 of the 58 patients (85%), with the time from surgery to completion of the questionnaire ranging from 6 mo to 11 yr. Mean score for severity of lameness at the time of discharge (0 = “no lameness” and 5 = “non–weight-bearing lame”) was 2.9, and mean score for degree to which owners complied with recommendations for postoperative confinement and rehabilitation (0 = “none” and 5 = “complete”) was 4.36. Postoperative complications were reported by 24 of the 49 (46%) respondents, with pain being the most commonly reported complication (17%), followed by reluxation (10%), dehiscence (6%), seroma (4%), infection (4%), complications involving an unrelated surgery completed at the time of the toggle rod procedure (2%), and chronic lameness (2%). When asked to rate limb function at the time of first recheck (0 = “no lameness” and 5 = “non–weight-bearing lame”), 44 of the 49 (90%) indicated a score of 0 or 1, and the remaining 9 indicated an average score of 2.7. When asked to rate current limb function (0 = “no lameness” and 5 = “non–weight-bearing lame”), 43 of the 49 (88%) indicated a score of 0 or 1, and 1 respondent indicated a score of 5. The remaining 4 owners reported an average score of 3.5. A complete breakdown of the more relevant questions is provided in Table 1.

TABLE 1 The Distribution of the More Informative Questions from the Client Questionnaire

View Fullscreen

When owners were asked if they were satisfied with the outcome of the surgical procedure (0 = “agree” and 5 = “disagree”), 43 of the 49 (88%) owners provided a score of 1 or less (“good” to “excellent agreement”), 1 owner (2%) provided a score of 2 (“fair agreement”), 2 owners (4%) reported a score of 2–3 (“fair to poor agreement”), and 3 owners (6%) provided a score of 3 (“poor agreement”). Three of the 5 owners who provided a score of “fair to poor agreement” or “poor agreement” with respect to their satisfaction with the procedure had patients in whom reluxation had occurred. It was determined that the more serious the complications experienced by the patient, the lower the owner satisfaction (Rs = –0.48, P < .01). Conversely, the higher the limb function as judged by the owners, the higher their satisfaction (Rs = 0.45, P < .02).