Excellent long-term clinical outcome in a young athletic dog following surgical complications of a comminuted patellar fracture and patellar ligament rupture

Abstract

A 1-year-old athletic golden retriever dog was presented for a non-weight-bearing right pelvic limb lameness that occurred during a canicross race. Orthopedic examination revealed pain and inflammation of the right stifle joint. Radiographs showed a traumatic comminuted patellar fracture. Surgery was elected and a midsubstance patellar ligament rupture was also identified. Surgical management consisted of a combination of techniques for patellar fracture fixation and primary patellar ligament reconstruction. All orthopedic implants were removed secondary to implant failure. Postoperative 20-month radiographic follow-up revealed patella alta and non-union of the fracture. However, an excellent clinical outcome was achieved with a complete return to a high activity level.

Key clinical message:

Surgical stabilization of patellar fractures is usually a significant challenge. A return to apparent normal limb function was achieved in this dog suffering complex patellar fracture with concomitant patellar ligament rupture despite postoperative fracture non-union.

The patella is the largest sesamoid bone in the body and is located between the tendon of insertion of the quadriceps femoris muscle proximally and the patellar ligament distally (1). The patella is a key component of the stifle joint extensor mechanism and is prone to trauma because of its superficial and cranial location as well as the high tensile forces to which it is constantly subjected (1).

Patellar fractures are rare in dogs and are most often secondary to trauma but have also been reported as a complication following tibial plateau leveling osteotomy (TPLO) surgery (2–4). Simultaneous bilateral fractures of the patellae have been described and are more common in cats than in dogs (5–9). These fractures are thought to be associated with chronic repetitive trauma, but in some cases an underlying bone abnormality such as osteochondrosis or incomplete ossification of the patellar secondary ossification centers is suspected as described in human medicine (5–10). Displaced fractures are most often transverse, given the tensile strength of the stifle joint extensor mechanism on the patella, but they may also be longitudinal (2,11). In veterinary medicine, the prevalence of comminuted patellar fracture has not been reported, but it seems uncommon.

Open reduction and internal fixation are usually recommended for displaced and comminuted fractures especially if the patellar ligament is also damaged (12). The main goals of surgery are to restore the function of the stifle joint extensor mechanism, to maximize postoperative range of motion, and to provide an early return to function without complications (13). Several surgical techniques for patellar fracture fixation have been described in veterinary medicine (2,6,8,9). Although the pin and tension band wire technique is mostly used in dogs, the ideal fixation method should be based on the type of fracture and should provide anatomic fragment reduction and rigid construct stability (12,13). Conservative management may be recommended for non-displaced and minimally displaced patellar fractures with an intact stifle joint extensor mechanism as well as absent or mild clinical signs (4,8,12). Surgical stabilization of a patellar fracture is usually a significant challenge and prone to complications, given the small size of the patella in dogs and cats and the high tensile forces to which it is constantly subjected. Little is known about the postoperative long-term outcome following surgical treatment of traumatic comminuted patellar fractures, especially in dogs with a concomitant patellar ligament rupture. Most of the available information comes from clinical reports with short- to medium-term follow-ups (14,15).

Case description

A 1-year-old athletic male golden retriever dog weighing 27.5 kg was presented to the Centre Hospitalier Universitaire Vétérinaire (CHUV) of the University of Montreal for an acute onset of non-weight-bearing right pelvic limb lameness that occurred during a canicross race. As witnessed by the owner, the stifle joint was injured when the dog slipped on a rock while racing. The dog had no previous medical history. Upon presentation, the dog was non-weight-bearing lame on the right pelvic limb but was otherwise in good general health. All vital parameters were within normal limits. Clinical examination of the affected limb revealed pain, severe swelling, crepitus, and a puncture wound at the level of the stifle joint. No obvious joint instability was identified. An intravenous injection of hydromorphone (Hydromorphone; Sandoz, Boucherville, Quebec), 0.05 mg/kg body weight (BW), was administered to treat pain and, following thorough clinical assessment, the dog was sedated with dexmedetomidine (Dexdomitor; Zoetis, Parsippany, New Jersey, USA), 3 μg/kg BW, IV. Right pelvic limb radiographs revealed a comminuted patellar fracture with three displaced fragments in an inverted “Y” pattern (Figure 1). Stifle palpation and manipulation were repeated and, according to the Gustilo-Anderson classification, a Grade II open fracture was diagnosed (12). Open wound management was done, and a modified Robert-Jones bandage was placed. The dog was hospitalized overnight and received intravenous fluid therapy (Plasma-Lyte A; Baxter, Mississauga, Ontario) supplemented with 30 mEq/L of KCl at 60 mL/h as well as hydromorphone (Sandoz), 0.05 mg/kg BW, IV, q6h, meloxicam (Metacam; Boehringer Ingelheim, Burlington, Ontario), 0.1 mg/kg BW, IV, q24h, and cefazolin sodium (Cefazolin; Sandoz), 22 mg/kg BW, IV, q8h.

Figure 1.

Preoperative orthogonal radiographs of the right stifle joint showing a comminuted patellar fracture with 3 displaced fragments in inverted “Y” pattern. Cranio-caudal view (a) and medio-lateral view (b) of the right stifle joint.

The dog was transferred to the surgery service the following day and, given the athletic lifestyle of the dog as well as the patellar fracture type and pattern — traumatic and comminuted — open reduction and internal fixation was recommended. A complete blood (cell) count, serum biochemistry profile, and urinalysis were performed and were within normal limits. The dog was premedicated with acepromazine (Atravet; Boehringer Ingelheim), 0.02 mg/kg BW, IV, and dexmedetomidine (Zoetis), 1 μg/kg BW, IV, and induced with alfaxalone (Alfaxan; Jurox Animal Health, Rutherford, Australia), 1.5 mg/kg BW, IV. The dog was intubated with a 11.0-mm cuffed endotracheal tube and isoflurane 1 to 5% delivered in 100% oxygen was used for maintenance of anesthesia. Ultrasound-guided femoral and sciatic nerve blocks were then performed by a Board-certified anesthesiologist with ropivacaine hydrochloride injectible (Pfizer, Kirkland, Quebec), 1 mg/kg BW. Perioperative prophylactic antibiotic therapy with cefazolin sodium (Sandoz), 22 mg/kg BW, IV, was administered before initiating surgery and was continued every 90 min until the end of the procedure.

The right pelvic limb was clipped, prepared, and draped for aseptic surgery. A medial parapatellar skin incision was made, followed by a medial and lateral arthrotomy to isolate the patellar ligament and patellar fragments. No significant abnormality of the menisci, cruciate ligaments, and long digital extensor tendon was noted upon exploration of the stifle joint, but a complete mid-substance patellar ligament rupture was present. Moreover, no medial or lateral laxity was noted during intraoperative manipulation of the stifle joint. As identified radiographically 3 patellar fracture fragments — 1 proximal and 2 distal — were observed. The lateral distal fragment was excised being considered too small, less than 1/3 of the patella, for primary reconstruction. Reduction and internal fixation of the remaining patellar fragments was then performed using 2 Kirschner wires (1.4 and 1.1 mm, respectively) and an orthopedic wire (0.8 mm) placed in a figure-of-eight configuration around them to provide compression at the fracture site. The patellar ligament was repaired with a locking loop suture pattern and several additional simple interrupted epitendinous sutures using a non-absorbable monofilament suture material (Prolene 1 and 2-0; Ethicon, Somerville, New Jersey, USA). Prior to routine closure of the surgical site, a loop of orthopedic wire (1.2 mm) was inserted through the tendon of the quadriceps femoris muscle just proximal to the patella and through a hole pre-drilled in the tibial tuberosity to protect the repair by relieving the tension on the patellar ligament. The traumatic puncture wound was finally debrided by sharp dissection, thoroughly lavaged with warm physiologic sterile saline, and primarily closed in a routine fashion.

Postoperative right stifle joint radiographs were acquired. The Kirschner wires were left too long and the orthopedic wire looped around them lacked direct contact with the patella. Fracture realignment and reduction were difficult to fully assess given the superimposition by the implants (Figure 2). External coaptation with a modified Robert-Jones bandage incorporating a lateral splint was then placed on the right pelvic limb prior to recovery. No complications occurred during the procedure and the dog recovered uneventfully from general anesthesia. The dog was closely monitored with supportive care and pain management, namely intravenous fluid therapy, hydromorphone 0.05 mg/kg BW, IV, q4h, meloxicam 0.1 mg/kg BW, IV, q24h, and cefazolin sodium 22 mg/kg BW, IV, q8h. Trazodone hydrochloride (Mar-Trazodone; Marcan Pharmaceuticals, Ottawa, Ontario), 10 mg/kg BW, PO, q8h, was also initiated in hospital and continued for 4 wk to keep the dog calm and to minimize anxiety after surgery. The splinted modified Robert-Jones bandage was changed the following day before discharge and then at least once a week until the first radiographic postoperative assessment. Complete activity restriction was recommended until otherwise indicated. Meloxicam, 0.1 mg/kg BW, PO, q24h, and cephalexin (Cefaseptin; Vétoquinol, Lavaltrie, Quebec), 25 mg/kg BW, PO, q12h, were prescribed for 7 d to treat postoperative pain and inflammation as well as to prevent postoperative infection, respectively.

Figure 2.

Immediate postoperative orthogonal radiographs of the right stifle joint. The Kirschner wires are left too long and the orthopedic wires looped around them lack direct contact with the patella. Fracture realignment and reduction are difficult to fully assess given the superimposition by the implants. Thickening of the patellar ligament and emphysema extending from the proximal aspect of the tibial tuberosity to the proximal aspect of the patella secondary to surgery are observed mainly on the medio-lateral view. A small, mineralized fragment superimposed on the patellar ligament proximally, most consistent with a small residual patellar fragment from the fracture, is also noted but obscured by the orthopedic wire passing through the tendon of the quadriceps femoris muscle proximally and the tibial tuberosity. Cranio-caudal view (a) and medio-lateral view (b) of the right stifle joint.

Clinical and radiographic follow-ups were performed at 4, 8, and 12 wk after surgery, and then every 8 to 12 wk until the latest follow-up performed 88 wk after surgery. Radiographs were done under sedation and interpreted by the Board-certified radiologist on duty and the Board-certified surgeon who did the surgery. The same surgeon performed all follow-up clinical examinations. At each visit, lameness was subjectively assigned a score ranging from 0 to 5 according to a modified categorization scheme, with 0 = no lameness observable and 5 = non-weight-bearing lameness (16). The dog’s thigh muscle mass and stifle joint mobility were also subjectively measured and compared to the uninjured contralateral limb. At the first examination, 4 wk after surgery, the dog had a Grade III right pelvic limb lameness. The thigh muscle mass and range of motion of the stifle joint were significantly decreased compared to the contralateral limb. Implant failure identified on radiographs consisted of breakage of the orthopedic wire passing through the tendon of the quadriceps femoris muscle and the tibial tuberosity. Mild stretching of the figure-of-eight tension band wire, mild patellar fracture distraction, a subjective patella alta, and severe soft tissue thickening in the region of the patellar ligament, most compatible with patellar desmitis, were also noted.

Surgical revision was considered, but given the lack of significant loosening of the implant, the complexity of the patellar fracture, the risk for additional complications as well as the young age of the dog, conservative management was elected at that time. The external copatation was therefore replaced as before and changed once a week until recheck. Eight weeks after surgery, the dog’s gait analysis had improved to a Grade II but the thigh muscle mass and range of motion of the stifle joint had slightly worsened. No pain was elicited upon palpation and radiographs were comparable to those taken before. The external copatation was once more replaced and changed weekly until recheck. Twelve weeks after surgery, gait analysis still revealed a Grade II right pelvic limb lameness. The thigh muscle mass and range of motion of the stifle joint were similar to the previous evaluation and no pain was elicited during orthopedic examination. Further breakage of the orthopedic wire passing through the tendon of the quadriceps femoris muscle and the tibial tuberosity was documented radiographically. There was still no evidence of bone healing and the patellar fracture gap had increased, consistent with delayed union (Figure 3). Removal of the broken orthopedic wire was elected and performed surgically without complications. The external coaptation was removed and physiotherapy as well as a progressive and gradual return to normal activity level were initiated at that time.

Figure 3.

Postoperative orthogonal radiographs of the right stifle joint at 12 wk showing breakage in 3 areas of the orthopedic wire passing through the tendon of the quadriceps femoris muscle proximally and the tibial tuberosity. There is still no evidence of bone healing and the patellar fracture gap has slightly increased. Early signs of degenerative joint disease are observed, especially in the femoral trochlea. Cranio-caudal view (a) and medio-lateral view (b) of the right stifle joint.

After approximately 8 wk of rehabilitation, no lameness was observed on clinical examination and the dog had returned to normal activity. Twenty-six weeks after surgery, an acute onset of Grade II right pelvic limb lameness was reported without associated trauma. The thigh muscle mass and stifle joint range of motion were similar to that of the contralateral limb. Repeatable pain, however, was elicited when the joint was manipulated. Radiographic assessment revealed non-union of the patellar fracture and complete implant failure (Figure 4). Removal of all orthopedic implants was elected to decrease local inflammation and to avoid pain. The procedure was performed under general anesthesia and immediate postoperative radiographs confirmed complete implant removal. A modified Robert-Jones bandage placed on the right pelvic limb to protect the incision in the immediate postoperative period was removed 1 wk later. The dog’s lameness progressively resolved on follow-ups 4 and 8 wk after removal of the implants.

Figure 4.

Postoperative radiographs of the right stifle joint at 26 wk showing non-union of the patellar fracture and implant failure with migration of the Kirschner wires proximally and breakage of the orthopedic wire. There is moderate progression of degenerative joint disease. Cranio-caudal view (a) and mediolateral view (b) of the right stifle joint.

Eighty-eight weeks after surgery, no lameness was identified on gait evaluation and the thigh muscle mass as well as the range of motion of the stifle joint were the same as those of the contralateral limb. There was no pain, patellar luxation, or joint instability noted upon palpation. Non-union of the right patellar fracture, persistent mild-to-moderate joint effusion and thickening of the patellar ligament, moderate progression of degenerative joint disease, as well as a subjective static patella alta were noted radiographically (Figure 5). The dog returned to the pre-injury athletic activity level — competitive canicross race — without any apparent lameness or clinical complications reported.

Figure 5.

Postoperative orthogonal radiographs of the right stifle joint at 88 wk showing non-union of the patellar fracture, persistent mild-to-moderate joint effusion, thickening of the patellar ligament, progression of degenerative joint disease, as well as a subjective static patella alta. Cranio-caudal view (a) and medio-lateral view (b) of the right stifle joint.

Discussion

This report describes a traumatic comminuted patellar fracture and patellar ligament rupture in a young athletic dog treated surgically with an excellent 20-month outcome despite implant failure and non-union after surgery. Data on long-term outcomes are lacking and only few cases with these concomitant conditions are reported with short- to medium-term follow-ups (14,15). It is widely assumed that non-union of a patellar fracture will result in poor limb function but this is debatable, especially in cats (6,8). The dog reported herein returned to a high athletic activity level even if the distraction force on the patella was not sufficiently addressed and separation of the fragments resulted in non-union. Restoration of full function of the stifle joint extensor mechanism is therefore possible without complete patellar fracture healing even in large active dogs, although the outcome obtained in this dog may not be extrapolated to all cases. Objective clinical data regarding the dog’s gait analysis with the use of a force plate or a pressure sensitive mat would have provided a more accurate outcome regarding the load on the affected limb compared to the contralateral one.

Surgical correction in this dog was complex and consisted of a combination of techniques including tension band wire fixation of the patellar fracture, partial patellectomy, mattress suture through the tendon of the quadriceps femoris muscle and the tibial tuberosity with an orthopedic wire, and primary reconstruction of the patellar ligament. A bilateral arthrotomy was necessary in this dog to allow a good exposure of the patellar fracture and to adequately reconstruct the patellar ligament. The repair was protected with external coaptation using a modified Robert-Jones splinted bandage, but despite all efforts, implant failure occurred at various time points and removal of the implant was required. Surgical repair of patellar fractures is associated with a high complication rate especially when pins with a tension band wire are used (6,8). This may partly be explained by the fact that it is often difficult to ensure good contact between the orthopedic wire and the patella because of the surrounding soft tissue and the small size and curved shape of this bone (1,13). The stability of the construct is therefore compromised, causing distraction of the patellar fragments, implant failure, and fracture non-union over time as reported in this case.

In a human medicine study, a reinforced braided polyblend suture of 2 polyester strings and a polyethylene string (FiberWire; Arthrex, Naples, Florida, USA) was superior to stainless steel when used as a tension band for transverse patellar fractures (17). The use of such a suture could therefore be considered for future cases. In a recent biomechanical cadaveric model, combination of both figure-of-eight and circumferential wires was the most robust device for stabilizing a feline transverse patellar fracture (18). However, the cyclic loading implemented in this study to test the different devices was determined for cats between 4 and 5 kg and at strict rest. This combined surgical technique is interesting, and should lead to further investigation, especially since tension band wiring did not appear to be the most suitable method in dogs in another biomechanical cadaveric model (13). In that study (13), a locking plate specifically designed for the patella allowed maintenance of a more consistent, stable, and reliable fixation method throughout cyclic loading in a transverse canine patellar fracture model compared to a pin and tension band wire fixation technique. This plate is currently only a prototype and not yet available but could eventually be a suitable option for management of future cases.

In this dog, the use of a wire inserted through the tendon of the quadriceps femoris muscle and the tibial tuberosity combined with external coaptation was insufficient in allowing good postoperative immobilization of the stifle joint. A transarticular external fixator should therefore be considered for the management of future similar cases to better counteract the tensile forces exerted on the patella after surgery. Transarticular external fixation should be the gold standard for patellar fracture and patellar ligament repair even if fixator-associated complications are common in dogs (19). In a recent retrospective study evaluating complications associated with external skeletal fixators in dogs, superficial pin-tract infection was the most common complication identified and transarticular frames were more likely to develop a complication (19). Implant failure and long bone fracture are rare but can be encountered as described in a recent case report of a young adult dog treated for bilateral patellar fractures (9).

A partial patellectomy is recommended for comminuted and more complex patellar fractures when the size of a fragment is approximately less than 1/3 of the patella or irreducible for fixation (12). A partial patellectomy was performed in the dog herein for 1 of the smaller distal fragments. Previous reports of partial patellectomy in dogs including using an arthroscopic approach resulted in a good outcome without any postoperative complications (5,14,20). Total patellectomy has been associated with a poorer functional outcome, although there is a lack of information in the literature regarding this procedure and its follow-up (12). Nevertheless, with the uncertainty of function following total patellectomy, it should be considered as a final salvage procedure.

Due to implant failure and distraction of the patellar fragments, non-union occurred in the case presented in this report. This complication has already been documented in dogs after surgical repair of patellar fractures (3). In cats, whether treated surgically or conservatively, non-union is observed in most cases (6,8,21). In this dog, non-union likely resulted from interfragmentary movement exceeding the tolerable strain for callus formation (22). Given the static appearance of the fracture gap during radiographic follow-ups, the development of an elastic fibrous tissue allowing stabilization at the fracture site is suspected. The good clinical outcome observed in this dog and other cases suffering this complication supports this hypothesis (3,6,8).

A subjective patella alta was identified on the first radiographic assessment performed 4 wk after surgery despite surgical reduction and fixation of the fracture, patellar ligament reconstruction, external coaptation, and complete restriction of activity. It is difficult, however, to definitely diagnose a patella alta as pre-injury radiographs and radiographs from the contralateral pelvic limb were not available to make objective comparative measurements. Moreover, stifle joint positioning and patellar fracture distraction prevent adequate measurements of the patellar ligament length, the patellar length as well as the vertical distance between the proximal pole of the patella, and the transcondylar axis of the distal femur, needed to objectively identify a patella alta as described by Mostafa et al (23). Patella alta is a common radiographic finding following patellar ligament rupture and the main clinical consequence is medial patellar luxation (15,24). In this dog, the suspected patella alta on radiographs could have resulted from stretching of the patellar ligament and/or suture dehiscence. This may be caused by a ligament repair that was not strong enough biomechanically, but also from inadequate postoperative immobilization of the stifle joint. To repair the patellar ligament with greater strength, a 3-loop pulley suture pattern could have been used instead of the locking-loop suture, as a biomechanical study showed that this pattern is more resistant to gap formation during tensile loading (25). A fascia lata graft could also have been used to strengthen the repair and should be considered in cases with severe patellar ligament rupture (26). Ultrasound follow-ups could have been beneficial for this dog to ensure that the patellar ligament was healing without complications; this should be considered in the postoperative management of future similar cases. The subjective patella alta was, however, asymptomatic in this case and no patellar luxation was documented clinically or radiographically at any postoperative follow-up.

In conclusion, a return to apparent normal limb function was achieved in this dog suffering complex patellar fracture with concomitant patellar ligament rupture, despite postoperative fracture non-union. The excellent clinical outcome seen in this dog, however, cannot be extrapolated to all cases facing such complications. Optimal implant selection and placement leading to a robust surgical repair are key to counteract the high tensile forces exerted on the stifle extensor mechanism and to avoid postoperative complications. Novel methods of fixation using FiberWire or patellar locking plates may provide more robust fixation, lower complication rates, and better consistent clinical results in achieving healing of patellar fractures. Further studies, however, are required in this field before making definitive conclusions. CVJ