Tibial Tuberosity Advancement for Treatment of CrCL Injury: Complications and Owner Satisfaction

Introduction

Rupture of the cranial cruciate ligament (CrCL) is the most common cause of hind limb lameness in dogs and is over a billion dollar per year industry in the United States.^1–3 Surgical techniques described for the cranial cruciate deficient stifle include exploratory arthrotomy or arthroscopy to inspect the meniscus and address meniscal injury, evaluation of the severity of degenerative joint disease and other joint pathology, and potential debridement of the ruptured CrCL, followed by a procedure to stabilize the joint. Many procedures and adaptations have been developed since Paatsama's first description in 1952.⁴ Surgical procedures are classified according to whether the stabilization is static or dynamic. The static procedures are further classified by whether the stabilization is performed within the joint capsule (intracapsular; e.g., the “over-the-top” technique), or outside of the joint capsule (extracapsular; e.g., the lateral fabellar suture technique).^5,6 Some of the geometric, dynamic stabilization techniques include a tibial osteotomy to indirectly stabilize the joint, as is performed in tibial plateau leveling osteotomy (TPLO), tibial tuberosity advancement (TTA), and cranial closing wedge osteotomy (CCWO), as well as several other procedures.^7–9 To date, one procedure has not been proven to be superior to another in terms of complication rates and clinical outcome.^2,10

The TTA is one of the newest techniques used to establish stability in the CrCL deficient stifle. First described by Dr. Slobodan Tepic at the University of Zurich in 2002, it is based on both a human procedure reported by Dr. Paul Maquet in 1975 to help decrease patellofemoral compressive stresses in people with stifle arthritis and chondromalacia of the patella as well as human models regarding tibio-femoral forces during weight bearing.^1,8,11,12 The procedure attempts to neutralize the abnormal forces in the CrCL deficient stifle by advancing the tibial tuberosity cranially, and therefore, changing the angle between the patellar tendon and the tibial plateau slope to the perpendicular, which reduces tibiofemoral shear forces.^8,12–15

According to Slobodan Tepic in 2002, early discovery of a high rate of complications in dogs at the University of Zurich before the procedure was publicly released caused the investigators to redevelop the design. According to Phil Stearns, to this date in 2010, Kyon^a has sold over 80,000 implants to over 750 veterinary surgeons. Although widely performed, few studies to date have described the complications, clinical outcome, and owner satisfaction associated with TTA. Some previously reported complications included subsequent meniscal tears, fractures, infections, incisional complications, patellar luxations, implant failures, and intraoperative procedural errors involving improper technique.^16–18 None of these studies to date analyzed risk factors for complications associated with the procedure, and only one other study by Lafaver et al. evaluated owner satisfaction by way of telephone interview.¹⁸

The purpose of this study was to describe the complications associated with TTA as defined in this study in a large cohort of dogs and determine the patient and procedure specific factors that might be associated with the occurrence of those complications. In addition, through the use of a questionnaire, the owners' global satisfaction with the procedure was assessed as well as their willingness to have the procedure performed again in the future should it become indicated. It was hypothesized that TTA was an acceptable technique for treatment of CrCL injury in dogs, which was justified by having similar complication rates and owner satisfaction to other currently practiced procedures.

Materials and Methods

Results

Signalment

The medical records of 198 consecutive TTA procedures were identified, of which, 193 had sufficient data recorded to be included in the study. The five dogs that were excluded did not have a complete medical record that could be located. Twenty-two dogs were represented twice in the cohort because they had the TTA procedure performed in opposite stifles on separate occasions. Signalment and presurgical condition of the dogs are presented in Table 1. In addition to the 12 different breeds in which complications occurred, there were 25 other breeds represented in which no complications were documented. There were 53 Labrador retrievers, 18 rottweilers, 15 Pitt bull terriers, 10 mastiffs, 10 golden retrievers, 6 mixed breed dogs, and 5 or fewer of the following dogs: Afghan hound, Ainu, Akita, American bulldog, Australian cattle dog, beagle, Belgian sheepdog, Bernese mountain dog, Bouvier, boxer, Chesapeake Bay retriever, chow, cocker spaniel, collie, Dalmation, English bulldog, field spaniel, German shepherd, Great Dane, Great Pyrenees, husky, komondor, Newfoundland, pointer, Rhodesian ridgeback, Russian hound, Samoyed, springer spaniel, St. Bernard, standard poodle, and Weimaraner. There were 109 spayed females, 5 intact females, 75 castrated males, and 4 intact males. The mean age was 5.86 yr (range, 0.9–15 yr). The mean body weight was 38.9 kg (range, 18.2–77.3 kg).

Table 1 Signalment and Presurgical Condition in 193 Dogs Undergoing Tibial Tuberosity Advancement, Stratified by the Occurrence of Postsurgical Complications

*

Complications were postoperative meniscal tears, incisional complications, implant failures, fractures, patellar luxations, and infections.

View Fullscreen

Complications

Complications were documented in 21 (11%) cases including meniscal tears (n=10), incisional complications (n=5), implant failure (n=2), fracture (n=2), patellar luxation (n=1), and infection (n=1). All complications were postoperative and no intraoperative complications were documented. Of the 104 dogs that had an intact meniscus at surgery, 10 (9.6%) had a subsequent meniscal tear. All of the meniscal tears were diagnosed surgically, and cases with meniscal injury had a partial meniscectomy performed. The implant failures consisted of one plate fracture leading to loss of reduction and avulsion of the tibial crest (Figure 1), in which the plate was surgically replaced, and one screw loosening from the caudal ear of the cage, which was surgically removed and not replaced. The two cases of fractures were nondisplaced tibial crest fractures that healed with external coaptation. The patellar luxation was a medial grade two and was surgically repaired with a wedge recession trochleoplasty. The infection, which was confirmed by culture and sensitivity testing, was associated with the plate and resolved with implant removal and antimicrobial therapy. Eighteen (86%) of 21 of the cases with documented complications and 18 of 193 (9.3%) of the total cohort received surgical intervention, and 3 (14%) of 21 of cases with documented complications received medical management.

View Figure

• Download as Powerpoint
• Download Figure

On univariate analysis, factors associated with the presence of postoperative complications in the 193 procedures with a P value <0.2 included the Labrador retriever breed (P=0.10), partial versus complete cruciate rupture (P=0.11), presence of meniscal tear (P=0.04), body weight 30.9–45.5 kg compared with <30.9 kg (P=0.04), body weight >45.5 kg compared with <30.9 kg (P=0.13), and decreasing cage size (P=0.18). On multivariate analysis, the combination of body weight and cage size revealed that cage size was a positive confounder to the association of body weight and complications in Group 3 (the heaviest) dogs. The coefficient for body weight changed by 35% for Group 3 dogs when basket size was added to the model. There was no substantial change in the coefficient for the Group 2 dogs. Because cage size was a confounder in the association with body weight, it was retained in the model despite the fact that the P value was 0.06 (Table 2). The other factors of breed, completeness of tear, and meniscal tear had P values >0.05 and did not change coefficients by >15%. Because these factors did not have a direct of confounding association with the incidence of complications, they were not retained in the model. Controlling for cage size, dogs weighing 30.9–45.5 kg had 11.4 greater odds of complication (95% CI 1.4–90.7) and dogs >45.5 kg had a 10.8 greater odds of complication (95% CI 1.1–107.5) compared with dogs weighing <30.9 kg. Body weight group and cage size adjusted fitted values for the probabilities of complications are presented in Table 3.

Table 2 Factors Associated with the Development of Postoperative Complications* in 193 Dogs that Underwent Tibial Tuberosity Advancement

*

Complications were postoperative meniscal tears, incisional complications, implant failures, fractures, patellar luxations, and infections.

†

Compared with dogs weighing <30.9 kg.

CI, confidence interval; OR, odds ratio; z, Wald statistic.

View Fullscreen

Table 3 Body Weight Group and Cage Size Adjusted Fitted Values for the Probabilities of Complications* in 193 Dogs that Underwent Tibial Tuberosity Advancement

*

Complications were postoperative meniscal tears, incisional complications, implant failures, fractures, patellar luxations, and infections.

View Fullscreen

Owner Assessment

Eighty-five surveys (44%) were returned for evaluation of owner satisfaction of the TTA procedure. This represented 44% of the total cohort and 48% of the cases (10 of 21) in which a complication was documented. The median length of time from surgery to mailing of the questionnaire was 562 days (range, 324–896 days). Baseline characteristics of the dogs from 85 owners who did and 108 owners who did not respond to the questionnaire are presented in Table 4. There were no significant differences in these baseline characteristics between dogs of owners that did respond to the survey and the dogs of owners that did not respond. Although 92% of owners who responded to the survey were satisfied overall with the outcome of the TTA procedure, only 84% stated that they would be willing to have the TTA procedure performed in the future. Owners of Great Pyrenees dogs were significantly more likely to be unwilling to have the TTA performed again than owners of other breeds of dogs (OR 21.2; 95% CI 1.6–286.5; P<0.001). Owners with dogs that had a postoperative complication were significantly more likely to be unwilling to have the TTA performed again than owners with dogs that did not have complications (OR 5.0; 95% CI 1.1–23.0; P=0.02). Owners who believed their dogs were currently uncomfortable on the leg were significantly more likely to be unwilling to have the TTA performed again than owners who believed that their dogs was currently comfortable (OR 26.2; 95% CI 3.0–228.3; P<0.001).

Table 4 Baseline Characteristics of the Dogs from 85 Owners Who Did and 108 Owners Who Did Not Respond to the Tibial Tuberosity Advancement Questionnaire

*

Complications were postoperative meniscal tears, incisional complications, implant failures, fractures, patellar luxations, and infections.

View Fullscreen

Discussion

Previous studies regarding tibial osteotomy techniques other than the TTA for dynamic stabilization after CrCL injury reported complication rates between 18.8% and 77.8%.^7,9,20–25 The disparity of complication rates might be explained by differing inclusion criteria and that several studies divided the data into major and minor complications, which could also have different definitions between studies. Sometimes a major complication was defined as requiring a surgical treatment and minor requiring a medical treatment. Talaat et al. reported a complication rate for the TPLO/CCWO in dogs with excessive tibial plateau angles as 77.80%, with patellar tendon thickening occurring in 61.1% of all cases.⁹ None of the dogs with patellar tendon thickening required any type of therapy. Without including the Talaat et al. study, the median complication rate among the other studies was 25.35%.^7,9,20–25

Since the introduction of the TTA in 2002, there have been few clinical reports in both peer-reviewed journal articles and abstracts regarding complication rates of this procedure. The reported complication rates ranged from 17.1 to 59%.^{16–18,26,27} As reported by Hoffman et al., there could be many procedural errors.¹⁷ In that report, there were a total of 55 technical errors in 65 operations, which suggested that there might be a steep learning curve associated with the procedure; many of the reported complications might be avoidable with more experience.

The complication rate in this study (11%) seemed to be less than that reported in previous TTA research. The three major TTA studies in the veterinary literature were retrospective in design and had a total of 236 cases, and a combined major complication rate of 22%.^16–18 This major complication rate was suggested by Stein and Schmoekel and seemed to be the most reasonable way to compare studies rather than accounting for every documented complication.¹⁶ This is because some studies reported incisional bruising, swelling, and edema in as high as 59.6% of cases, which might be an expected result of most surgeries with low clinical significance. These previous publications all reported similar weight, signalment, surgical findings, and surgical technique to this study.

One possible explanation for the lower complication rate in this study was that all procedures were performed by a single experienced surgeon, so many of the technical errors that might be associated with complications might have occurred less frequently compared with a study in which the same number of procedures might have been performed by a number of surgeons, many of whom might have had less experience with the technique. Another possibility was that some complications occurred but were not documented in the dogs' record. The radiographs taken at 6 wk were not evaluated by a radiologist, so the only complications documented from radiographs that would be included in this report were those that were significant enough to be documented in the medical record. Radiographic complications that were not of clinical significance might be underreported.

Unlike previous reports, the authors' goal was not only to document complications but to determine whether risk factors for complications could be identified. The data indicated that increasing weight was a risk factor for complications. It was unlikely that the reason for this was that heavier dogs unduly stressed the implants because implant failure was not a common complication. This result could have been because of the greater forces placed on the stifle postoperatively in larger dogs. Regardless of cause, this finding suggested that heavier dogs, in particular, needed to be closely monitored for the development of postoperative complications after TTA or perhaps alternative techniques should be considered.

In addition to increasing body weight, smaller cage size was associated with a higher risk of complications when a 6 mm cage was used compared with using a 9 or 12 mm cage. Although cage size selection of each implant was based on the radiographic measurement and template, the smaller implants might not cause adequate advancement of the tibial tuberosity, and therefore, not adequately neutralize the forces in the stifle. This could then lead to continued tibial thrust and subsequent complications. Specifically when looking at the data, it seemed as though there were a relatively small amount (5/193) of 12 mm cages used, which suggested that some dogs might have been undersized. This could have contributed to the finding that a smaller cage size or a larger dog was associated with a higher risk of complications. These data would suggest that the larger cage should be selected if a measurement on preoperative radiographs was between 6 and 9 mm, keeping in mind the variation in preoperative planning used in the cases supporting these data.

Subsequent meniscal tears and the performance of prophylactic meniscal release or partial meniscectomy is still under much debate.²⁸ Listed in the present study as a complication due to the recommendation for reoperation, with current treatment strategies, there is always a risk for subsequent meniscal tears. Previous research showed that subsequent meniscal tears will occur even with a meniscal release during the TPLO.²⁸ Other studies showed that both meniscal release and partial meniscectomy could be associated with higher stress concentration in the stifle, which might increase the risk of osteoarthritis progression.²⁹ More recent studies by Kim et al. suggested that pressure distribution on the meniscus and femoro-tibial contact points were better reestablished with TTA than TPLO in the CrCL deficient stifle.^30,31 Therefore, in a stifle with an intact meniscus, there was no definitive consensus as to whether the meniscus should be released or removed at time of the original surgery. In the present study, only two dogs had prophylactic meniscal releases performed, and neither of them had a subsequent meniscal tear reported. The overall subsequent meniscal tear rate was 5.2% for all dogs and 9.6% for dogs with an intact meniscus at the time of TTA, which is in line with other TTA studies in which the subsequent meniscal tear rate was between 6.3 and 29% after surgery for CrCL disease.^{16–18,26,27} As with all studies, it was possible that there were meniscal tears that were missed at time of the TTA and identified as a postoperative complication.

Owner reported overall satisfaction was found to be good or excellent in 92% of cases. Previous reports of owner assessed outcome had good to excellent results in 90–93% of cases.^16–18 These results were similar to owner assessed outcome after TPLO and CTWO, which reported 86–94% good to excellent results.^7,9,22,23 It was interesting that 10% of owners who said they were satisfied overall with the procedure would be unwilling to have it performed again. Owners who perceived their dogs had an unsatisfactory comfort level on the leg at the time of the survey, owners of dogs that had a postoperative complication, as well as owners of Great Pyrenees dogs were less likely to be willing to have the procedure performed again. Interestingly, cost was not associated with this decision in this group of owners. This finding suggested that owners cared most about an uncomplicated procedure and an end result that left their dogs comfortable. This was not surprising as complications and dogs that were persistently uncomfortable could cause owners frustration and grief. The authors did not have an explanation for why the Great Pyrenees dog owners were unwilling to have the procedure performed again. Only one of the four dogs in the cohort had a complication, none of the dogs were considered uncomfortable at the time of the survey, and none of them were dissatisfied with the cost of the procedure.

There were several limitations to this study that must be considered when evaluating the data. Because there were so few of each type of complication, a statistical analysis evaluating risk factors for individual complications was not justified. Instead, the risk factors identified were for all complications. Also, it was possible that there were more implant-related complications, bone healing complications, or technical errors than reported; however, if present, dogs were not symptomatic for the failure and did not require any additional therapy within the follow-up time described. Because the TTA is a relatively new procedure performed at the Oradell Animal Hospital, there is no long-term follow up more than 2 yr. The complications involved with a long-term follow-up study would likely involve progression of osteoarthritis and implant-related complications.

An additional limitation was that the preoperative planning radiographs were not positioned with a goniometer, but were just positioned in as much extension using sandbags to hold the leg. As described in the original procedural instructions, the leg should be positioned in 135° of extension, which is a weight-bearing position.^1,32–34 It was possible that variable positioning of stifle angle and the presence or absence of cranial drawer could lead to variability in the calculation of advancement of the tibial tuberosity, and therefore, continued stifle instability. Future studies could investigate the variability of stifle angles in relation to the calculations of TTA using the standard template to determine the impact of this preoperative planning limitation.

As with any study using a survey, one must consider whether the surveys that were returned were representative of the entire cohort. This study had a 44% return of surveys from the entire study population and 48% return of surveys from the owners of dogs that had complications. There were no significant differences in the measured characteristics of the dogs and their surgeries between the “responders” and “nonresponders” to the surveyor, which would suggest that the responding cohort could be representative of the entire cohort. It is possible, however, that there were some unmeasured factors that could cause people to be more or less likely to respond, which could bias the results. In addition, the survey was a subjective outcome measure, and subjective measures used with no blinding and no control inherently had the potential for some bias.

Conclusion

The TTA procedure has a complication rate and owner satisfaction similar to other tibial osteotomies for the surgical correction of CrCL disease. Also, procedures utilizing the smaller cage size and those performed in heavier dogs were associated with higher complication rates. The surgeon must be aware of long-term comfort level as perceived by the owner because this likely will have an effect on their willingness to have the procedure performed again. Prospective, long-term cohort studies, as well as controlled clinical trials are needed to further determine which factors are associated with a successful outcome of the TTA procedure and which animals may benefit from TTA compared with other available options for treatment of the cranial cruciate deficient stifle.