Short-Term Complications Associated With TPLO in Dogs Using 2.0 and 2.7 mm Plates
The purpose of this study was to evaluate short-term complications associated with the tibial plateau leveling osteotomy (TPLO) procedure in dogs of small and medium sizes. These study findings could then be compared with similar studies in dogs of all sizes. Ninety-eight procedures (82 dogs) using 2 or 2.7 mm TPLO plates were included spanning a 6 yr period with a minimum of an 8 wk follow-up period. The overall complication rate was 36%. Complications were grouped into three categories: intraoperative (2%), those occurring up to 2 wk postoperatively (11%), and those occurring >2 wk postoperatively (30%). Radiographic evidence of delayed healing of the TPLO osteotomy site was the most common complication in the dogs (9%), none of which had clinical signs attributed to delayed healing. Other common complications included patellar tendon thickening (8%) and tibial tuberosity fracture (6%). Seven percent of the complications were considered major (requiring additional surgical intervention). Fifty-two percent of complications did not require treatment. Short-term complications should not dissuade an experienced surgeon from performing the TPLO procedure in small- and medium-sized dogs.
Introduction
Cranial cruciate injury is the most common condition associated with the stifle joint in dogs.1–4 The cranial cruciate ligament functions to stabilize the joint by opposing cranial drawer motion, hyperextension, and internal rotation.2,5–9 Rupture of that ligament causes instability of the joint, leading to inflammation that causes degradation of the hyaline cartilage matrix, commonly leading to stifle osteoarthritis.3,10,11
Instability of the stifle after cranial cruciate ligament rupture is a result of cranial tibial thrust.8,9 Cranial tibial thrust is the cranial-directed shear force generated during the weight-bearing phase of ambulation and is a function of the tibial plateau angle (TPA), musculotendinous attachments, and tibial compression into the femoral condyles.8,9,12–14
Several techniques have been described to stabilize the stifle joint. The techniques are generally divided into intra- and extracapsular procedures. Several studies have been conducted to investigate the efficacy and the complication rates of such techniques.14–24
Tibial plateau leveling osteotomy (TPLO) was reported by Slocum and Slocum in 1993.13 The purpose of the extracapsular technique was to eliminate cranial tibial thrust without replacing the ligament or eliminating cranial drawer. The premise behind the procedure is that by decreasing the TPA, cranial tibial thrust will be neutralized during weight bearing, thereby preventing the instability normally present without a functioning cranial cruciate ligament.
Several reports of complications associated with the TPLO procedure have been published in the last 10 yr.20–24 Complications rates vary from 10 to 28%, with most complications being minor in nature.20–24 Previously reported complications include tibial fracture, intra-articular screw placement, hemorrhage, tissue swelling, bandage complications, seroma, dehiscence, incision inflammation and infection, patellar tendon thickening (desmitis), osteomyelitis, meniscal injuries, implant loosening, implant failure, fibular fracture, “pivot shift,” medial patellar luxation, and patellar fracture.20–29
In the study authors’ opinion, TPLO is avoided in small- and medium-sized dogs due to an anecdotally reported increased complication rate associated with the procedure. The purpose of this descriptive study is to report the short-term complications associated with the TPLO procedure in small- and medium-sized dogs using 2 and 2.7 mm TPLO plates. These findings can then be indirectly compared with similar studies of a variety of plates and dogs to assess differences in complication types and rates.
Materials and Methods
The medical records of all dogs that underwent a TPLO procedure using either a 2 or 2.7 mm TPLO plate performed at Metropolitan Veterinary Hospital were reviewed. Those plate sizes were chosen in an effort to delineate small- and medium-sized dogs from the general population. The diagnosis of cranial cruciate ligament rupture was confirmed by presence of cranial tibial thrust, cranial tibial drawer, or visualization at the time of surgical stifle exploration. Patients were included if they presented 10–14 days postoperatively for examination and again 8 wk postoperatively for examination and radiographic evaluation (i.e., for mediolateral and craniocaudal views of the stifle). All osteotomies needed to be stabilized using either a 2 or 2.7 mm TPLO plate to be included. Medical records were required to include signalment, body weight, surgical report, record of complications, and results of the radiographs of the affected stifle performed preoperatively, immediately postoperatively, and 8 wk postoperatively.
Information obtained from the medical records included age, sex, breed, body weight, body condition score, limb affected, physical exam findings, serial lameness scores, bone saw size, plate size, preoperative TPA, amount of rotation, presence of meniscal pathology, procedures performed on menisci, and radiographic findings at the 2 and 8 wk postoperative evaluations.
Complications associated with the procedure were defined as any unwanted outcome during the procedure or postoperative period confirmed by either physical exam or radiographs. Nature of the complication and date of diagnosis were recorded and grouped based on time at which they were noted relative to surgery. Type 1 complications included those occurring from the time of anesthetic induction to anesthetic recovery, type 2 complications were classified as acute postoperative complications developing from the time of anesthetic recovery up to and including 14 days after surgery, and type 3 complications were chronic and postoperative developing ≥15 days after surgery. Complications were further classified as major (those requiring a second surgical procedure as treatment) or minor (those requiring either medical therapy or no treatment). Surgical site infection was defined as one in which a purulent discharge formed within 14 days after surgery.30
Dogs were anesthetized for surgery using a variety of balanced anesthetic protocols. The affected limb was clipped and prepared for surgery. An epidural was administered using preservative-free morphine (0.1 mg/kg) and bupivacaine (0.55 mg/kg). All dogs received cefazolin (22 mg/kg IV) at the time of anesthetic induction and q 90 min thereafter until the completion of the procedure. Preoperative radiographs were performed, positioned according to Slocum’s recommendations, and TPA and saw blade size were determined using digital templatesa.13 Saw blade size was chosen with a goal to center the osteotomy over the tibial intercondylar eminences while leaving an adequate sized tibial tuberosity strut cranially (at the surgeon’s discretion). All surgeries were performed by one board-certified surgeon (R.M.D.). If the patient was of sufficient size, exploratory arthroscopy was performed prior to the TPLO procedure. Stifle fat pad and cranial cruciate ligament debridement were performed. The menisci were inspected by the arthroscope after joint retraction was performed. In the event of a complete cranial cruciate rupture, a medial meniscal release was performed if the meniscus was intact. If the meniscus was torn, a partial meniscectomy was performed. If a partial tear of the cranial cruciate was diagnosed (with >50% of the cranial cruciate ligament intact and uninjured) and the meniscus was intact, the medial meniscus was left intact. A skin incision was made on the medial aspect of the stifle. If arthroscopy had not been performed, a caudomedial arthrotomy (hence meniscal release) was performed via midbody meniscal transection. A partial meniscectomy was performed if a meniscal tear was present. If a caudomedial arthrotomy was performed, the cranial cruciate was neither visualized nor debrided. Caudomedial arthrotomy was performed when complete cranial cruciate ligament tears were diagnosed by physical examination. Any suspected partial cranial cruciate ligament ruptured was confirmed with arthroscopy or craniolateral arthrotomy. A TPLO jigb was applied to the medial aspect of the tibia followed by osteotomy using a TPLO saw bladec. An osteotomy was performed using the predetermined saw blade. The proximal tibial segment was then rotated the predetermined distance based on templated preoperative radiographs. The osteotomy was initially stabilized with an antirotational Kirschner wire placed just proximal to the level of the patellar tendon insertion and pointed reduction forceps. A 2 or 2.7 mm TPLO plated was then contoured and secured using screws. When a 2 mm TPLO plate was used, nonlocking cortical screws were used. When a 2.7 mm TPLO plate was used, locking cortical screws were used except for the most proximal screw, in which case a nonlocking cancellous screw was used. Following bone plate application, the antirotational Kirschner wire and pointed reduction forceps were removed. Postoperative radiographs were obtained (mediolateral and craniocaudal views). All radiographs were reviewed by the surgeon performing the procedures, and a modified Robert-Jones bandage was placed postoperatively and removed the morning following surgery.
Dogs were typically discharged the day following surgery with cephalexin (22 mg/kg per os [PO] q 8 hr) for 5 days, tramadol (4 mg/kg PO q 8 hr) for 10 days, and carprofen (4.4 mg/kg PO q 24 hr) for 14 days. Owners were instructed to confine the patient to a small room or kennel at all times except for leash walks outdoors for the purpose of eliminations. Physical exams were performed by a supervised surgical resident or R.M.D. 10–14 days postoperatively, 8 wk postoperatively, and radiographic evaluation 8 wk postoperatively (i.e., mediolateral and craniocaudal views). Radiographs were performed earlier if complications were suspected. Appropriate bone healing was defined as loss of the radiolucent osteotomy and evidence of bone bridging along caudal aspect of tibia to the proximal osteotomized segment on the mediolateral radiographic projection. If appropriate bone healing was determined at the 8 wk postoperative radiographic evaluation, slow reintroduction to activity was recommended over the ensuing 8 wk.
For the purpose of this study, radiographs performed 2 and 8 wk postoperatively were recovered from the hospital archives when possible. Digital or digitally scanned standard radiographs were used to calculate TPA at each time interval by a single author (R.M.D.) as described by Slocum.13
Statistical Analysis
Associations between variables were tested using χ2 tests. A two-sample t test was also performed when assessing variables associated with complications. For all analyses, a value of P < .05 was considered significant.
Results
Signalment
Ninety-eight procedures in 82 dogs fit the inclusion criteria. Forty-eight dogs (49%) were female, and median age at the time of surgery was 7 yr (range, 1–15 yr). Median body weight of the patients was 12.5 kg (range, 4.5–31.9 kg). Represented breeds were listed in Table 1.
Preoperative Radiographic Evaluation
Median and mean preoperative TPA was 30 and 31.1°, respectively (range, 16–47°).
Surgery
Fifty-eight surgeries (59%) were performed on the left limb. The most common blade size used was 18 mm (n = 33), followed by 12 mm (n = 24), 15 mm (n = 20), 21 mm (n = 19), 10 mm (n = 1), and 24 mm (n = 1). Median tibial tuberosity rotation was 7.5 mm (range, 3.5–11.75 mm). A 2 mm TPLO plate was used in 44 procedures (45%) and 2.7 mm TPLO plate in 54 procedures (55%). Arthroscopy was performed in 12 procedures (12%), with a partial meniscectomy performed in 2 and a meniscal release in 6. In the 86 procedures in which arthroscopy was not performed (88%), a partial meniscectomy was performed in 49 and a meniscal release in 37. In total, a meniscal tear was diagnosed in 53 of all procedures (54%).
Postoperative TPA
Radiographs associated with 91 procedures were available for TPA measurements. The median and mean TPA immediately postoperative was 6° and 6.5°, respectively (range, 1–20°). The 8 wk postoperative TPA median and mean was 7° and 7.8°, respectively (range, 1–24°). The median and mean change in TPA from the immediate postoperative radiographs to those performed 8 wk postoperatively was 1° and 1.3° (range, −2–17°), respectively.
Complications
Forty-two complications were noted in 35 out of 98 procedures (36%) performed. Multiple complications were related to six procedures (6%). Four of those procedures had two complications and one procedure had three separate complications. Seven procedures (7%) had a major complication requiring an additional surgical intervention, with one patient requiring two separate reparative surgical procedures.
Type 1 complications, accounting for 5% of total complications, occurred in two dogs. In the first dog, screw placement was noted to be within the joint space on postoperative radiographs (Figure 1). The screw was suspected to be within the femoropatellar fat pad and not in contact with any other articular surfaces; therefore, the screw was left in place. That same patient also subsequently developed delayed healing osteotomy and infected incision (group 3).
Citation: Journal of the American Animal Hospital Association 50, 6; 10.5326/JAAHA-MS-6074
The second type 1 complication was a fractured fibula, which occurred during rotation of the proximal tibial segment. The preoperative TPA was 23° and rotation was 3.75 mm. A type 2 external skeletal fixator (ESF) was placed on the tibia. Healing of the osteotomy and fracture was confirmed on radiographs 8 wk postoperatively, and no lameness was noted.
Type 2 complications were recorded following 11 procedures. Those complications accounted for 26% of all complications and included inflammation of the incision (n = 2), incisional dehiscence with surgical site infection (n = 2), seroma formation (n = 2), residual tibial thrust (n = 2), tibial tuberosity fracture (n = 2), and lameness of unknown etiology (n = 1).
Two dogs with inflammation of the incision consisting of mild swelling, erythema, and discomfort on palpation were noted. No incisional discharge was noted. Both patients were treated with cephalexin (22 mg/kg PO q 8 hr) for 7 days and owner applied warm compressing, resolving both complications.
Klebsiella spp. and Enterobacter spp. were cultured individually from two dehisced incisions in which purulent discharge was noted. Each was treated with antibiotics based on sensitivity results, local surgical debridement, and Penrose drain placement. One patient healed uneventfully while the other healed initially, later formed a draining tract leading to implant removal 5 mo postoperatively.
Seroma formation occurred in two patients. Both were diagnosed on physical examination findings and fine-needle aspirate of the swelling. Cytology of the aspirates revealed a hypocellular collection of degenerate red blood cells with infrequent neutrophils and no micro-organisms. Each dog was treated conservatively with owner-applied warm and cold compressing and application of an Elizabethan collar.
Appreciable cranial tibial thrust was present at the 2 wk postoperative evaluation in two patients. One dog had similar findings at the 8 wk postoperative evaluation; however, did not develop clinical signs attributable to that finding. The second dog had unremarkable radiographs performed at the 2 wk postoperative evaluation due to mild discomfort on palpation. However, worsening lameness the subsequent week prompted repeat radiographs that revealed backing out of the most proximal screw. This screw was surgically tightened and a type 1A lateral tibial ESF was placed and the patient went on to heal uneventfully at the 8 wk post operatively radiographic evaluation at which point the fixator was removed.
Two tibial tuberosity fractures were documented in the two weeks following surgery. Both were noted at the scheduled 2 wk postoperative examination and were associated with discomfort and palpable crepitus during range of motion of the stifle. The first was treated with tension band placement, type 1A lateral tibial ESF application and plate and screw replacement (due to unstable implants noted at the time of surgery). This patient had radiographic documentation of healing 10 wk following the initial procedure. The second was treated with tension band placement alone, but later developed a draining tract that was treated with antibiotics based on culture and sensitivity. Radiographic healing was confirmed in this patient 8 wk following the TPLO procedure.
In a single patient, persistent unexpected severity of lameness of unknown origin was present 2 wk postoperatively. The lameness improved, but persisted at the 8 wk postoperative evaluation. Radiographs at that time revealed stable implants with an appropriately healing osteotomy site. Physical therapy was recommended and performed by the owner at home. The patient presented 8 mo later for a contralateral cranial cruciate rupture and the lameness had resolved in the previously operated limb.
There were 29 type 3 complications, which accounted for 70% of all complications. Complications included delayed union of the osteotomy (n = 9), patellar tendon thickening (n = 8), tibial tuberosity fracture (n = 4), incision swelling (n = 2), infection (n = 2), implant failure (n = 2), painful periosteal proliferation (n = 1), and incision region lick granuloma (n = 1).
All procedures with delayed healing of the TPLO osteotomy were noted incidentally on routine radiographs at an 8 wk postoperative examination (Figure 2). Patients had no clinical signs attributable to delayed union. A prolonged rehabilitation period, including 12–16 wk of slow reintroduction to activity, was recommended, and all patients continued to do well clinically. Radiographs to confirm eventual healing were not routinely performed.
Citation: Journal of the American Animal Hospital Association 50, 6; 10.5326/JAAHA-MS-6074
Patellar tendon thickening was noted on radiographs, physical examination, or both in eight operated stifles (Figure 3). That was an incidental finding at the 8 wk postoperative evaluation, and clinical signs associated with this finding were absent. Routine 8 wk rehabilitation was recommended for those patients.
Citation: Journal of the American Animal Hospital Association 50, 6; 10.5326/JAAHA-MS-6074
Tibial tuberosity fracture was documented in four dogs at a point >14 days after surgery. Those were all documented as an incidental finding at the time of 8 wk postoperative radiographic evaluation. In three of those four patients, an extended rehabilitation plan was recommended, all three continued to do well clinically. After diagnosis, strict cage rest was recommended for 6 wk for the fourth patient with tibial tuberosity fracture. Radiographic evidence of bony union was documented 6 wk later, and the dog was slowly returned to normal activity.
Incision swelling was noted in two patients. The first was an incidental finding at the 8 wk postoperative examination and did not require treatment. The second case was noted 6 wk postoperatively. Owner-applied warm compresses were recommended, and swelling resolved by the 8 wk postoperative evaluation.
Infection accounted for two of the type 3 complications. Methicillin-resistant Staphylococcus aureus was cultured from a draining tract following surgical correction of a tibial tuberosity fracture in one dog. Antibiotic therapy was instituted, and the tract resolved without further treatment and was last assessed 9 wk following initial surgery. Implant removal was required following recurrent draining tracts despite several attempts of antibiotic therapy guided by culture and sensitivity in another dog. Culture at the time of implant removal revealed an Enterobacter spp. and was subsequently treated with 6 wk of antibiotics. No further drainage was noted as of 22 mo following surgery.
Implant failure including screw pullout and screw breakage was diagnosed in two cases, 3 and 4 wk postoperatively respectively. The first was treated with screw replacement and type 1A lateral tibial ESF application and was confirmed healed 11 wk postoperatively. The second was treated in the same manner and was successfully healed 12 wk postoperatively.
Radiographic evidence of excessive periosteal proliferation of unknown etiology in the region of the osteotomy was noted in one patient 3 wk following surgery after acute lameness was noted. Radiographs revealed the periosteal proliferation along with intramedullary sclerosis consistent with new bone formation. Cephalexin (22 mg/kg PO q 8 hr) for 21 days (to treat a possible underlying bacterial cause) and carprofen (4.4 mg/kg PO q 24 hr) for 10 days were prescribed. Radiographs performed 8 wk postoperatively revealed resolving periosteal proliferation, healing osteotomy, and the dog was no longer lame.
A lick granuloma was noted over the incision in one dog at the 8 wk postoperative evaluation. Cephalexin (22 mg/kg PO q 8 hr) for 2 wk was prescribed along with an Elizabethan collar. The granuloma resolved uneventfully.
The presence of short-term complications was not significantly associated with plate size (P = .95), presence of a meniscal tear (P = .15), age (P = .73), body weight (P = .30), body condition score (P = .67), saw blade size (P = .27), amount of rotation (P = .809), preoperative TPA (P = .27), immediate postoperative TPA (P = 1), 8 wk postoperative TPA (P = 1), or change in TPA from the immediate postoperatively to 8 wk postoperatively (P = .19).
Discussion
This study reviewed 98 TPLO procedures in 82 dogs utilizing 2 and 2.7 mm TPLO bone plates in small- and medium-sized dogs. The goal of this retrospective study was to review complications associated with the TPLO procedure in this patient population using 2 and 2.7 mm TPLO plates and compare the results to similar studies that include a variety of sized plates (and dogs) to assess differences in complication types and rates.
The median age of 7 yr with 49% male and 51% female are similar populations to that of previously reported studies.20–24 A meniscal tear was present in 54% of cases. That rate is similar to previous reports in which the meniscal injury at the time of diagnosis varies from 20 to 77%.25,31 The overall complication rate was 36%. The complication rate in other recent studies varied from 10 to 28% for the TPLO procedure.20–24 Other procedures commonly performed to treat cranial cruciate injury include tibial tuberosity advancement, lateral fabellotibial suture (LFS), and fibular head transposition. Complications rates for those procedures were 31–42%, 17.4%, and 16.5–27%, respectively.15–19
Body weight of the patients varied from 4.5 to nearly 32 kg. When looking at body weight alone, one may consider 32 kg to be a large-breed dog. However, all dogs weighing >25 kg had a body condition score of ≥4.5, alluding to the fact that their body weight was not proportionate to bone size. Three out of four of the heaviest patients were bulldogs with tibiae too small to allow the use of larger implants.
Major complications have been previously classified as any complication requiring an additional surgical procedure or lameness persisting >12 wk.20 Using that classification system, 17% of complications were major (7% of all procedures) in this study. This rate is consistent with previous major complications rates of 14–45%.20,21,24
No factors were identified in this study that influenced the occurrence of complications. Previously reported risk factors for complications following the TPLO procedure include breed, performance of arthrotomy, body weight, sex, and complete ligament tears.20–24
An intraoperative complication rate of 2% was noted in this study. Previous reported intraoperative complication rates vary from 1–1.5%, including the studies by Stauffer et al. (2006) and Pacchiana et al. (2001).21,23 Priddy et al. (2003) did not identify specific intraoperative complications; however, the perioperative complication rate (from the time of surgery until discharge) was 13%.22 A slightly higher intraoperative complication rate may indicate the procedure is technically more challenging in smaller patients, warranting the necessity of an extensive amount of experience with the procedure prior to attempting it in smaller dogs.
Incision complications were common in this study, accounting for over one-quarter of all complications in groups 2 and 3. However, a second surgical procedure to address the incision complications was only required in two cases.
Patellar tendon thickening was noted at the radiographic evaluation appointment after eight procedures (8%), accounting for 20% of all complications. However, clinical signs associated with this were absent in all patients. This rate is somewhat higher than previously reported values of up to 5%.20,21,23 Theories as to why the patellar tendon becomes thickened in patients include trauma caused by the saw blade at the time of osteotomy, placement of the antirotational pin though the distal patellar tendon at the time of proximal tibial segment rotation, early or excessive postoperative patient activity, or a change in the stifle joint biomechanics caused by movement of the intercondylar eminence.21,32 A previous study found a cranial osteotomy, a partially intact cranial cruciate ligament in conjunction with a cranial osteotomy and postoperative tibial tuberosity fracture to be risk factors for postoperative thickening of the patellar tendon.27 A possible explanation for the higher rate of patellar tendon thickening in this study compared with other studies is the limited amount of space between the saw blade and the patellar tendon when creating the osteotomy in smaller dogs. A temporary antirotational pin was used in all procedures; however, careful attention was paid to place the pin just proximal to the level of insertion on the tibia, making this cause less likely. Antirotational pins were also used in other similar studies.20–24 Recommendations given to owners regarding activity restriction postoperatively are similar to most others.20,21
The most common type 3 complication was delayed healing, accounting for one-fifth of all complications. Delayed healing was measured subjectively on assessment of radiographs performed 8 wk after surgery. No dog was clinically lame at the time of evaluation. This rate is substantially higher than previously reported 0.3%.20 Other similar studies did not identify delayed healing as a complication.21–24 It is not possible to determine if this was not considered a complication or whether it did not occur in those populations of dogs. If delayed healing was not considered a complication in this study, the overall complication rate is reduced from 36 to 27%.
When accounting for the significantly higher rate of delayed healing when compared with Fitzpatrick and Solano (2010), there are several potential explanations.20 The first is that healing was measured subjectively. Another possibility included true delayed healing. The most common causes of delayed healing of fractures and osteotomies were infection, lack of stability, and inadequate blood supply to the region.33 Infection was documented in only one of nine cases of delayed healing. Lack of stability of the osteotomy was a potential cause and a result of the owners’ inability to adequately restrict the activity of the dogs after surgery or the procedure itself. A 2 mm TPLO plate was used in seven of the nine cases in which delayed union was a complication, leading one to believe smaller bone plates and screws may provide less stability to the osteotomy, predisposing to delayed healing. The difference in plate size and its association on delayed healing approached significance (P = .053), and it was possible if a larger patient population was used, that difference would become significant. In the authors’ opinion, contouring was more difficult when using 2 mm TPLO plates versus larger plates. Small contouring discrepancies could have more profound effect on osteotomy compression in those small patients. Locking screws were commonly used in 2.7 mm TPLO plates, but not 2 mm TPLO plates. Locking screws have been shown to provide greater stability and improved healing of TPLO osteotomies when using 3.5 mm TPLO plates.34 Another potential cause of instability in those patients was the removal of the antirotation pin prior to wound closure. In the study by Fitzpatrick and Solano, the antirotational pin was left in place prompting the current study authors to investigate the effect of leaving the antirotation pin in place versus removal.
No postliminary meniscal injuries were documented in this study. To confirm the frequency of postliminary meniscal injuries, one would need to perform an arthrotomy, arthroscopy, or ultrasound to visualize each meniscus. Postliminary meniscal injuries have been documented in 0–10.5% of cases reported previously.20–24,35 Those tears were documented during either arthrotomy or arthroscopy performed to make the diagnosis and allow for treatment. In the current study, medial meniscal release was performed in 96% of stifles in which a meniscal tear was not present at the time of surgery. Those figures support the theory that medial meniscal release, though controversial, may be protective of future meniscal injury, as has been previously reported.35 However, the follow-up times in this study may not be sufficient to detect all postliminary meniscal injuries. Also, it is possible that the clinical signs of the single dog with lameness of unknown origin that resolved with physical therapy were a result of postliminary meniscal injury.
Tibial tuberosity fractures occurred postoperatively following 6% of procedures. That complication was more common than in previous studies, reported to occur in 1–4% of patients.21–23 Although an attempt was made during all procedures in this study to create a tibial tuberosity segment that was increasing in width distally when creating the osteotomy, the size of the segment was naturally smaller in that population of patients compared with larger dogs. It is possible that this played a role in the higher fracture rate. A temporary antirotational wire was placed slightly proximal to the insertion of the patellar tendon in all procedures in this study. That was similar to previous reports in which tibial tuberosity fractures were reported, making it less likely to impact the complication rate.21–23 The mean preoperative plateau angle in this study was 31.1°, compared with 26–27° in other studies.21–23 The greater angle and secondary greater rotation may have contributed to higher rate of fractures.
A lateral type 1A ESF was applied to treat three complications. The complications treated included two implant failures and one tibial tuberosity fracture with implant instability noted intraoperatively. Although not previously reported as treatment of those complications, the fixators were placed at the surgeon’s discretion in an attempt to augment the stability of the construct. By providing lateral support, it was thought to reduce the likelihood of further complications associated with instability while keeping skin penetrating fixator elements as far from the primary stabilization as possible to minimize pin tract facilitated bacterial seeding of bone plate and screws. Mild pin tract drainage was noted in association with all three dogs; however, no other complications were associated with the constructs. All three dogs healed uneventfully.
Subclinical complications were common in this study, accounting for 52% of all complications. Those complications were not noted by owners at home, did not contribute to lameness at the time of observation, and did not require medical or surgical therapy. If those complications are removed from the overall complication pool, the complication rate is 17%, appreciably lower than the overall complication rate of 36% in this study. A similar comparison was not reported in previous similar studies.20–24
LFS has been the treatment of choice in dogs with cranial cruciate ligament injury for many clinicians. There are few current studies evaluating the complication associated with and outcome of this procedure.19,36 Compared with the study of Casale and McCarthy (2009) regarding complications associated with LFS, several similar findings were noted. The reoperation, documented surgical site infection, and implanted related complication rates were found to be the same at 7, 4, and 3%, respectively. Although the overall complication rate was lower for LFS (17%), major complications accounted for over two-fifth of all complications, more than twice the rate of the current study.
The median immediate postoperative TPA of 6° is an acceptable postoperative outcome based on previous studies evaluating functional outcomes of varying angles following TPLO.14,37–38 A median difference in TPA between immediate and follow-up radiographs of 1° (mean, 1.3°) was consistent with a previous study evaluating 149 dogs following TPLO. In the current study, a median of 1.5° change was noted.40 Although subsidence or “rock back” of the proximal osteotomized segment was a certain concern, other factors such as intraobserver variability (previously reported medians of 1.5–1.7°), patient positioning for radiographs, radiographic beam center location, and tibial plateau remodeling have been shown to affect measurement of TPA.40–42 In a few patients, a net decrease in TPA was noted at the 8 wk radiographic evaluation compared with the immediate postoperative views. Intraobserver variability and radiographic artifact likely explain the small number of cases in which a “negative” change in TPA occurred.
A major limitation to this study is its retrospective nature. Consistent with any retrospective study, records may be incomplete and owners may have elected treatment elsewhere, unknown to the authors. There was also no standardization of procedures, including screw type used. Another limitation with this study is the subjective measurement of many parameters including lameness and radiographic healing. The follow-up time may be considered short in this study; however, the goal was to assess short-term complications associated with the procedure. It is possible that complications, particularly postoperative meniscal tears, occurred after the follow-up protocol requested by the clinicians and were not detected in this study. None of the reported patients voluntarily presented for recurrent lameness on the operated limbs with times of 1–7 yr following surgery. Those surgeries were performed by a single experienced TPLO surgeon and represent his results. TPLO in the small animal patient is arguably technically more challenging and individual results may vary with surgeon and experience level.
Conclusion
This study revealed an overall complication rate somewhat higher than many other similar studies that evaluated complications associated with the TPLO procedure. Despite the higher overall complication rate, over half of those encountered were minor and nonconsequential to clinical outcome 8 wk postoperatively. The reoperation rate was similar compared with previous studies. Based on the information obtained from this study, short-term complications occurring in small- and medium-sized dogs should not discourage an experienced surgeon from performing the TPLO procedure in this population of patients. A prospective study with standardized procedures and long term follow-up would be beneficial to determine long term functional outcome and overcome some of the short comings of a retrospective study.
Lateral (A) and caudocranial (B) radiographic views of the left stifle of a 5 yr old castrated male bichon frise immediately following tibial plateau leveling osteotomy procedure. The cranial screw in the proximal portion of the bone plate penetrates the joint surface but was believed to be cranial to the weight-bearing surface.
Lateral radiographic view of the left stifle of a 4 yr old spayed female mixed-breed dog recorded 8 wk following tibial plateau leveling osteotomy procedure. Delayed healing of the osteotomy is noted with proliferative callus formation along the caudal aspect of the tibia with minimal bone bridging of the osteotomy gap.
Preoperative (A) and 8 wk postoperative (B) lateral radiographic views of the left stifle of a 6 yr old castrated male mixed-breed dog. Marked patellar tendon thickening is noted on the postoperative radiographs, most notably in the distal half of the tendon. Caudocranial projection confirmed extra-articular location of the most proximal screw.
Contributor Notes
