Complications Associated With 696 Tibial Plateau Leveling Osteotomies (2001–2003)
A total of 696 tibial plateau leveling osteotomy (TPLO) procedures were performed over a 30-month period following TPLO training. The overall complication rate was 18.8%.
Complications were classified as perioperative (1%), short-term (9.3%), and long-term (8.5%). Examples of complications encountered during the study were hemorrhage, swelling at the incision site, premature staple removal by the dog, tibial tuberosity fracture, patella tendon swelling, and implant complications. Based on the rate of complications observed, clinical outcomes of TPLO procedures within 30 months of TPLO training were considered good.
Introduction
In the normal canine stifle, an intact cranial cruciate ligament (CCL), composed of cranial medial and caudal lateral bands, works to oppose cranial tibial thrust.1,2 Cranial tibial thrust is an internally generated force that causes cranial translation of the tibia during weight bearing. This force is a function of both tibial compression and the slope of the tibial plateau.2,3 Rupture of the CCL results in stifle instability secondary to cranial tibial thrust.
Sequelae of CCL rupture have been well documented.4–6 Stifle instability results in inflammation from the release of inflammatory mediators such as interleukins, leukotrienes, collagenases, and elastases resulting in degradation of cartilaginous matrix and usually progresses to degenerative joint disease.4–7 To avoid such sequelae, numerous types of surgical procedures have been devised for stabilizing the affected stifle.8–11 Retrospective studies of extracapsular repairs, intracapsular repairs, and fibular head transpositions have been reported.12–15
Tibial plateau leveling osteotomy (TPLO), as developed by Slocum in 1993, attempts to eliminate cranial tibial thrust.1 Slocum et al. proposed that decreasing the tibial plateau angle would neutralize cranial tibial thrust and increase the effectiveness of the hamstring and biceps femoris muscles. The goal of the TPLO procedure is to decrease the tibial plateau angle to approximately 6° by relocating the proximal tibia.1 Initially, TPLO was a patented procedure with strict legal requirements for its use.a This patent has expired, making the procedure increasingly available for more veterinarians to perform.
The purpose of this retrospective study is to report complications associated with TPLO procedures performed during the first 30 months after TPLO training.
Materials and Methods
Data Collection
Medical records of dogs that underwent TPLO for treatment of CCL rupture at the Veterinary Specialty Center of Indiana between June 2001 and December 2003 were reviewed. Diagnosis of CCL rupture was made on the basis of stifle joint effusion, cranial tibial thrust, cranial drawer motion, and/or radiographic evidence of degenerative joint disease with concurrent joint effusion. Information collected included signalment, duration of lameness, and whether there was unilateral or bilateral involvement. If the CCL rupture was unilateral, the affected leg was noted. If the condition was bilateral and both limbs were operated, the time between operations was noted.
Operative data evaluated included surgical time, whether or not arthrotomy was performed (and if so, its extent), and if a medial meniscus-releasing incision or partial meniscectomy was performed. Preoperative, immediate postoperative, and 8-week follow-up radiographs were reviewed. Tibial plateau angle was recorded for both stifles preoperatively and for operated stifles postoperatively.
Complications were defined as any undesirable consequence associated with the TPLO procedure. Complications were classified as perioperative, short-term (i.e., postsurgical recovery to 14 days), or long-term (i.e., 15 days to 30 months after surgery).
Surgical Procedures
For the first 150 cases, the first and third authors performed the TPLO procedures as a team. All subsequent surgeries were performed by one of these authors alone. Both surgeons performing the procedure had previously attended the TPLO courseb at the same time.
All dogs were routinely anesthetized and maintained with isofluranec in oxygen. The rear limb was clipped and prepared for surgery in a routine manner. Each dog was placed in dorsal recumbency with the pelvis over the end of the table. Each dog received two doses of cefazolind (22 mg/kg intravenously [IV]), one prior to surgery and the other 2 hours postoperatively.
All procedures were performed according to the TPLO technique as described by Slocum, et al.1 If a medial meniscus-releasing incision was performed, either a medial arthrotomy or a limited incision caudal to the medial collateral ligament was used. All osteotomies were secured by either a 2.0-, 2.7-, or a 3.5-mm TPLO bone plateb and six screws, based on the size of the dog and the discretion of the surgeon. Immediately postoperatively, the dogs were radiographed and the tibial plateau angle was determined. Morphine epiduralse (0.1 mg/kg) were performed preoperatively, or multiple morphine injectionsf (0.5 mg/kg intramuscularly [IM]) were given during the postoperative recovery period.
A modified Robert Jones bandage was applied to the first 50 cases for 72 hours postoperatively. For the other cases, a limited Robert Jones bandage was applied for 24 hours and removed at discharge. All owners received written instructions concerning postoperative care and follow-up examinations. Instructions for postoperative confinement and the 8-week rehabilitation period were based on TPLO course recommendations and modified according to surgeon preference.
All dogs were discharged with 2 weeks of nonsteroidal antiinflammatory therapy and butorphanol tartrateg (0.05 mg/kg orally q 8 hours). Eight weeks after surgery, each dog was evaluated clinically and radiographically.
Statistical Analysis
The relationship between the occurrence of a complication and the type of arthrotomy performed was assessed using the chi-square test of independence. All analyses were performed at the 5% level of significance. If the chi-square statistic was significant, contrasts among the types of arthrotomy were tested using Freeman-Tukey deviates. The resulting P values were adjusted for making multiple comparisons using the Bonferroni method.
Results
Signalment
Of the 696 TPLOs performed during the study period, 51% (n=355) were done on males and 49% (n=341) were done on females. The mean age was 6.2 years (range 9 to 13.2 years). The mean weight was 38.4 kg (range 5.0 to 91.8 kg). The five most commonly affected breeds were the Labrador retriever (29%; n=201), golden retriever (19%; n=131), rottweiler (11%; n=74), Newfoundland (7%; n=51), and mastiff (4%; n=28). The remaining dogs were mixed-breed dogs or a variety of small-breed dogs.
Clinical presentation ranged from acute CCL rupture of <1 week’s duration to chronic lameness for an extended period of time, with a mean of 9 weeks (range 3 days to 56 weeks). The mean preoperative tibial plateau angle was 26° (range 16° to 36°). The left stifle was affected in 57% (n=397) of the dogs, and the right stifle was involved in 43% (n=299). Bilateral TPLOs were performed in 17% (n=118) of the cases, with a mean interval of 6.8 months (range 1 to 15 months) between TPLOs.
Surgical Parameters
Surgery time (i.e., time elapsed from skin incision to closure) for the first 100 procedures averaged 105 minutes. By the end of the study, surgical time had decreased to an average of 45 minutes (range 40 to 65 minutes) for the last 100 cases. A medial arthrotomy was performed in 13% (n=90) of the cases and had a complication rate of 40% (n=36), whereas a limited medial arthrotomy was done in 81% (n=564) of the cases, with a complication rate of 15.6% (n=88). A medial meniscus-releasing incision was performed in 89% (n=619), and a partial meniscectomy was used in 5% (n=35) of the cases. These procedures had complication rates of 18.7% (n=116) and 22.8% (n=8), respectively. No joint invasion was undertaken in 6% (n=42) of the dogs (also precluding any meniscal surgery). In these cases, the complication rate was 16.6% (n=7). When compared, the arthrotomy group had significantly more complications than did the limited medial arthrotomy (P<0.001) or the group that did not undergo arthrotomy (P=0.024). The limited medial arthrotomy group and the group that did not undergo arthrotomy were not significantly different from each other (P=1.000). The mean postoperative tibial plateau angle was 6° (range 2° to 9°) for the 696 animals.
Postoperative Follow-up
All dogs were examined at least twice in the postoperative period, with one of the visits at 8 weeks. Most (93%) dogs were seen at staple removal (i.e., 10 days postoperatively) and for follow-up radiographs 8 weeks postoperatively. Some (7%) dogs were seen more frequently for various complications.
Complications
The overall complication rate (n=131) was 18.8%. There were 12 different types of complications reported [see Table]. Perioperative complications occurred in 1% (n=7) of the cases and included hemorrhage (n=4), chip fracture (n=1), implant failure (i.e., broken screw, n=1), and fibular head fracture (n=1). Short-term complications were identified in 9.3% (n=65) of the cases and most frequently involved swelling (i.e., edema or bruising) at the incision site (6%, n=43). Other short-term complications included premature removal of skin staples (2%, n=13), incision line inflammation (1%, n=7), tibial fracture (n=1) [Figure 1], and joint capsule swelling (n=1).
Long-term complications were recorded in 8.5% (n=60) of the dogs. Tibial tuberosity fractures [Figures 2, 3] occurred in 4% (n=28) of the cases at 2 to 6 weeks postoperatively. Patella tendon swelling occurred 5 to 8 weeks postoperatively in 3% (n=19) of dogs. Implant complications (e.g., screws loosening or backing out) [Figure 4] occurred in 1% (n=6) of the cases and were found at various times (most commonly at the 8-week recheck). Fibula (n=4) and tibia (n=2) fractures occurred in six cases.
Discussion
It is interesting to note that although the total number of cases in this report exceeded those of other retrospective studies of the TPLO procedure, the complication rate (18.8%) was lower.16–18 In a similar retrospective study, Pacchiana et al. reported complication rates of 28%.16 Because both surgeons in this report attended the TPLO coursec together and were able to closely collaborate on the first 150 cases, the learning curve for the procedure may have been shorter, thereby decreasing the number of complications. Based on the overall complication rate determined in this study, complications associated with TPLO may be no greater than complications with other invasive orthopedic procedures where aggressive bone-plate-screw application techniques are applied.19
Of the 696 cases in the current report, the distributions of breeds were similar to those reported in previous articles.16,18 In this study, mean weight of the operated dogs was 38.4 kg, which correlated well with other studies that found mean weights of 39.9 kg and 41.2 kg for dogs undergoing TPLO surgery.16,18 Although other studies have found a preponderance of spayed females with CCL rupture, the numbers of males and females were essentially equal in the present study.16,18,20,21
Bilateral surgeries were performed on 118 dogs in a staged fashion, and those dogs had no higher complication rates than dogs with unilateral surgeries. These results reflected a complication rate for bilateral TPLOs that was less than that from a report where bilateral, simultaneous TPLOs were associated with a complication rate of 40%.17 In this study, no dog underwent simultaneous TPLO. Waiting at least 4 weeks between procedures may have allowed some bone healing to occur and may have lowered the complication rate following bilateral TPLOs.
Slocum et al. recommended a postoperative TPLO tibial plateau angle of 6°, which was achieved in most cases in this study.1 More recent reports, however, indicate that tibial plateau angles may or may not have a bearing on the incidence of CCL rupture.20,21 The differences in correlating tibial plateau angles with CCL rupture may relate to variability in determining tibial plateau angle measurements. Although there is subjectivity in current procedures for obtaining both preoperative and postoperative tibial plateau measurements, it is still prudent to continue documenting these measurements.22
Although surgery time decreased from an average of 105 minutes in the first 100 cases to an average of 45 minutes in the last 100 cases, there were no significant differences in the number of dogs that developed complications in relationship to the length of surgery. This was interesting, since other reports have indicated increased risks of infection and glove perforations in surgical procedures lasting >60 minutes.23,24
In 90 (13%) cases, a medial arthrotomy was performed. After the first 50 cases, in an attempt to decrease surgical time and morbidity, an arthrotomy was no longer performed unless the surgeon determined that there was a need for thorough joint evaluation.
Since the complication rate dropped significantly when the standard medial arthrotomy was no longer performed, it may be prudent to avoid complete arthrotomy in favor of either a limited approach or none at all if there is no reason to suspect meniscal damage. When meniscal surgery was performed, the type (i.e., partial, meniscal-release, or none) was not significantly associated with whether complications developed, which was similar to other reports.16,18 In light of the current debate as to whether prophylactic meniscal release is necessary during TPLO repair, the results reported here seem to indicate that by using a partial arthrotomy, the opportunity is provided to perform this technique without necessarily increasing the rate of complications.
In the perioperative period, hemorrhage during TPLO procedures may result from elevation of muscle groups located around the proximal tibia. Recommendations for the TPLO procedure include instructions to pack the space between the caudal or lateral proximal tibia and overlying muscles. By ensuring adequate packing of the muscles overlying the proximal tibia, the popliteal artery and musculature are protected from the saw blade.25 All four cases of significant hemorrhage reported in this study occurred before packing the lateral surface of the proximal tibia. At no time in any of these cases was any type of ligation or stapling technique required. Pressure and packing were usually sufficient to control hemorrhage. When packing alone was not successful in controlling hemorrhage, a clotting agenth was applied (n=2).
Edema or bruising at the surgical site was the most common complication reported during the immediate postoperative period and the most common of all the complications encountered. Postoperative edema responded to conservative management in all cases. In a small number of cases, clinically relevant incision line inflammation occurred and was correlated to subcutaneous suture placement. This complication did not involve significant edema or bruising of the entire stifle, but instead resulted in a localized redness that was closely associated with the incision line. This soft-tissue inflammation resolved and did not require antibiotics or diagnostic procedures.
One dog experienced joint capsule swelling of the operated stifle. Synovial fluid effusion was confirmed via fine-needle aspiration and cytology. Culture and sensitivity testing performed on the fluid was negative for any infectious agents, and treatment (i.e., Penrose drain) resolved the swelling.
Tibial tuberosity fractures were the most common long-term complication recorded, occurring in 4% of the TPLOs performed. This rate compared favorably with other studies where tibial tuberosity fractures occurred in 1% to 4% of cases.16–18 In the study reported here, tibial tuberosity fractures generally occurred from 2 to 6 weeks postoperatively. Owners noticed the dogs became acutely lame at the time of the fracture. The acute lameness occurred after a time of initial recovery in which the dog was weight bearing. Tibial tuberosity fractures were visible on radiographs but were not significantly displaced. Tibial tuberosity fractures occurred below the temporary threaded stabilization pin. Without precise placement of the TPLO jig, stress may occur on the tibial tuberosity. It is recommended that the jig be placed so that the pin is as high as possible in the tuberosity, thereby reducing the chance of tibial tuberosity fracture. In this study, none of the tibial tuberosity fractures occurred through the osteotomy-stabilizing pin site, but all were located below the pin placement site.
In the 19 dogs with patellar tendon swelling, radiographic evidence of the swelling was found. These dogs also exhibited pain upon manipulation of the stifle and palpable swelling of the patellar tendon. This complication occurred 5 to 8 weeks postoperatively, as the dogs became more active. Radiographically, the position of the bone plate and screws was normal, and thickening of the patellar tendon was the only abnormality noted.
No cases of osteomyelitis were recorded, but implant complications (e.g., screws loosening or backing out) occurred in six cases. These complications were noted on postoperative radiographs taken at the 8-week recheck or upon palpation. If screw loosening was minimal and did not clinically impede the dog, no intervention was needed.
Besides tibial tuberosity fracture, there were six other fracture complications during the long-term follow-up period. Four fibular fractures occurred (one necessitated external fixation), and two tibial fractures required internal fixation.
Conclusion
Out of 696 TPLO procedures performed, the complication rate was 18.8%. Just as many complications occurred in the short-term postoperative period as in the long-term period. Edema or bruising at the surgical area was the most common complication encountered, followed by tibial tuberosity fractures. Based on the data from this study, a reasonably low rate of complications and good clinical outcomes can be expected following TPLO surgeries performed within 30 months of TPLO training.
US Patent No. 5,304,1880; Slocum Enterprises, Eugene, OR 97404
Slocum Enterprises, Eugene, OR 97404
Isoflurane; Attane, Minrad, Buffalo, NY 14203
Cephazolin; Faulding Pharmaceutical, Elizabeth, NJ 07207
Duramorph; Elkins-Sinn, Inc., Cherry Hill, NJ 08003
Morphine sulphate; Baxter Healthcare, Deerfield, IL 60015
Torbugesic; Fort Dodge Animal Health, Fort Dodge, IA 50501
HemaBlock; Abbott Laboratories, North Chicago, IL 60064
Acknowledgment
Special thanks to Dr. Carolyn Boyle of Mississippi State University, College of Veterinary Medicine, for her help with statistical questions.
Citation: Journal of the American Animal Hospital Association 42, 1; 10.5326/0420044
Citation: Journal of the American Animal Hospital Association 42, 1; 10.5326/0420044
Citation: Journal of the American Animal Hospital Association 42, 1; 10.5326/0420044
Citation: Journal of the American Animal Hospital Association 42, 1; 10.5326/0420044
Tibial fracture from the tibial plateau leveling osteotomy site through the distal screw holes of the bone plate that occurred 10 days postoperatively, after the 7-year-old, spayed female, terrier-mix dog fell down a flight of stairs. R=right.
Tibial tuberosity fracture (white arrow) that occurred 4 weeks postoperatively in a 4-year-old, spayed female golden retriever.
Tibial tuberosity fracture in a 4-year-old, 25-kg, spayed female chow chow-mix dog, illustrating an osteotomy (open arrow) that was placed/cut too far cranially, which may lead to tibial tuberosity fracture.
Radiograph from a 4.5-year-old, castrated male mastiff, showing loosening of the no. 1 and no. 3 screws and breakage of the no. 2 screw at the head/shaft interface. A fracture of the fibula is visible just distal to the fibular head (arrowhead).
