Quadriceps Tendon Disruption

The literature regarding quadriceps tendon rupture after TKA is scant, due to the rarity of this complication. Quadriceps tendon rupture has been reported to occur after 0.1% to 1.1% of primary TKAs. Dobbs et al. reported on the experience at the Mayo Clinic, identifying 24 (0.1%) tendon ruptures in a cohort of 23,800 primary TKAs. Lynch et al. found three quadriceps tendon ruptures after 281 primary TKAs. Even in the setting of revision TKA, the quadriceps tendon rupture is a rare entity. Yun et al. described three postoperative quadriceps tendon ruptures after revision when a quadriceps release or snip had been performed.

No clear underlying condition has been shown to increase the risk quadriceps tendon rupture after TKA, given the few reports of this complication. Rheumatoid arthritis, systemic and local steroid use, obesity, and diabetes have been suggested as comorbidities that put the quadriceps tendon at risk. Additionally, technical factors such as overresection of the patella and quadriceps release techniques have been identified as potential risk factors.

The treatment of quadriceps tendon rupture after TKA is most comprehensively discussed in the report by Dobbs et al., including 23 partial tears and 11 complete tears. Partial tears were defined as a tendon tear that did not cause a loss of active extension. Seven of the patients with a partial tear were treated nonoperatively with immobilization and had a satisfactory outcome, without re-rupture. Sixteen patients with a partial tear underwent direct suture repair, with five resulting in serious complications including re-rupture, knee instability, intraoperative popliteal artery laceration, fracture of the femoral condyle, and deep periprosthetic infection. The average postoperative range of motion revealed a worsened extensor lag. Ten of 11 patients with a complete tear had a surgical repair; two included mesh augmentation. Of the 10 patients who underwent primary repair of the tendon, four had re-rupture and two developed a periprosthetic infection.

The use of an extensor mechanism allograft including quadriceps tendon, patella, patellar tendon, and tibial tubercle has also been described for reconstruction of the extensor mechanism. When meticulous technique and tensioning are followed, a high rate of success may be expected from this procedure. Extensor mechanism allograft reconstruction may be used as a primary technique for the treatment of quadriceps tendon rupture after TKA, but it is especially attractive when direct repair or augmentation techniques are unsuccessful.

In summary, quadriceps tendon tears after TKA must be divided into two groups depending on the presence of active knee extension. Partial tears usually have some active extension and are treated by immobilization in extension for 4 to 6 weeks, followed by passive range of motion for 4 to 6 weeks and then strengthening. Complete tears have no active extension and should be treated by an augmented direct repair or by an extensor mechanism allograft. Operative treatment for complete tears aims to avoid the severe extension lags observed with nonoperative treatment.

Patellar Tendon Rupture

Patellar tendon rupture after primary TKA has been reported to occur at a rate of 0.17% to 1.4%. Rand et al. described 18 ruptures out of 8,288 TKAs at the Mayo Clinic over a 12-year period, accounting for a 0.17% rate of rupture. Boyd et al. reported a 0.56% incidence when reviewing 891 TKAs with and without resurfacing. Lynch at al. reported 4 patellar tendon ruptures (1.4%) in a consecutive group of 281 primary TKAs.

A variety of underlying factors have been found to potentially increase the likelihood of patellar tendon rupture after TKA. Systemic considerations such as diabetes, steroid use, and collagen vascular diseases are possible predisposing factors. Local factors such as implant type have also been implicated. The early generation of hinged prostheses such as the Stanmore, Walldius, and Guepar prostheses appear to have higher rates of patellar tendon rupture, possibly due the increased constraint. One study reported a 5.6% rate of patellar tendon rupture after use of the Walldius hinged prosthesis.

Patellar tendon ruptures are not specific to hinged designs, and numerous studies have reported rupture in more modern TKA designs, ranging from 0.17% to 1.4%. Resurfacing of the patella has been suggested to be a contributing factor to patellar tendon rupture, although this has never been clearly proved. An asymmetric or aggressive patellar resection could theoretically impair the patellar tendon attachment to the patella. In a comparison of resurfaced and unresurfaced TKAs, Boyd et al. found no statistically significant difference between resurfaced (3 of 396 knees) and unresurfaced (2 of 495 knees) TKAs.

The treatment of patellar tendon ruptures depends on the ability of the patient to actively extend the knee. Because of complications related to operative intervention and mixed results, the presence of active extension is best treated without surgery, including immobilization with a brace or cast. With loss of knee extension, surgical repair is necessary to avoid a severe extension lag.

Primary repair of patellar tendon ruptures is almost routinely unsuccessful, as reported by Rand et al. Of 16 operative repairs, nine included direct repair, four used staple fixation, two used xenograft, and one used autograft. Only four results were considered satisfactory, and direct repair was consistently unsuccessful. Lynch et al. described the direct repair of four patellar tendon ruptures, reporting one re-rupture, one deep infection, and two with extension lags of more than 15 degrees. Oglesby and Wilson reported on five direct repairs, with only one patient achieving full extension.

Cadambi and Engh reported on the use of semitendinosus autograft for the repair of seven patellar tendon ruptures with successful reconstitution of the extensor mechanism. The semitendinosus tendon is incised proximally, looped through a drill hole in the distal patella, and then sutured back down to its own insertion at the tibia. Tensioning is accomplished at 90 degrees to avoid patella Baja. Postoperatively, the patients are immobilized for 6 weeks in extension, followed by limited flexion for an additional 6 weeks. Although they did find an extension lag of 5 to 10 degrees in some patients and limited flexion in most patients, the result was acceptable in every patient and no complications were observed.

Medial or extended medial gastrocnemius flaps have also been used to repair a ruptured patellar tendon with success. Jaureguito et al. reported seven patients who underwent this technique. At an average of 2 years of follow-up, the average range of motion increased from 70 to 100 degrees and the average extensor lag improved from 53 to 24 degrees. Unfortunately, most patients still demonstrated a significant extensor lag. Two complications included a manipulation resulting in a 30-degree extensor lag and one patient with the need for a small skin graft.

The successful results of extensor mechanism allograft using quadriceps-patella-patellar tendon-tibial tubercle allograft have been well described at different centers as a treatment for complete quadriceps tendon or patellar tendon ruptures after TKA.

Patella Fractures

The consequences of patellar fracture after TKA are highly dependent on the resulting effect on extensor mechanism integrity. Some patellar fractures are benign, without effect on the patellar component or extensor mechanism. On the other hand, a patellar fracture can result in the need for significant surgical intervention aimed at reconstructing the continuity of the extensor mechanism. The patella has little inherent bone stock, and its resistance to fracture may be easily compromised during TKA, especially revision TKA.

The reported rate of patellar fracture after TKA has a wide variation in the literature. Patellar fracture after primary TKA has been reported in one large study to have a rate of 3.9%, with 0.4% of TKAs having disruption of more than 1 cm. Ortiguera and Berry reported 85 patellar fractures (0.68%) in a registry of 12,000 TKAs over a 12-year period. Grace and Sim found nine patellar fractures (0.12%) of 7,754 primary TKAs. The rate of patellar fracture may vary depending on approach, lateral release rate, posterior cruciate ligament retention, decision to resurface the patella, implant types, and inclusion of revisions.

The rate of patellar fractures has been suggested to be dependent on general patient factors such as inflammatory disease, osteoporosis, gender, or increased activity levels. Rheumatoid arthritis may be a predisposing factor if the disease results in a thin, eroded, or cystic patella, and some authors have suggested avoiding resurfacing when this is identified. However, there have been a few studies that found no difference between patients with rheumatoid or osteoarthritis, and there is little evidence that rheumatoid arthritis, in and of itself, significantly elevates the risk of patellar fracture after TKA.

The increased rate of patellar fracture with patellar resurfacing after TKA has been clearly described by a number of authors. Additionally, a number of biomechanical studies have demonstrated that strain in the remaining patellar bone increases as the thickness of the bone decreases. Both Reuben et al. and Lie et al. studied the anterior patellar strain after resection at varying depths. They found patellar strain to significantly increase when less than 15 to 16 mm of bone remained after resection. Although the rate of patellar fracture after TKA has not been specifically associated with a certain cutoff thickness of remaining patellar bone, most surgeons prefer to leave at least 12 mm of patellar bone stock during TKA, even when this results in an overall patellar thickness that is greater than preoperative values. During revision TKA, it is common to find a patellar thickness of less than 10 mm after component removal, especially with concomitant osteolysis. It is advisable to avoid simple resurfacing in these cases to avoid postoperative patellar fracture.

A variety of local insults have also been credited with the predisposition for patellar fracture. Devascularization of the patella and its tendon may theoretically occur with overzealous releases and fat pad excision. However, it remains unproved whether sacrificing the superior lateral geniculate vessels during surgery actually place the patella and its tendon at risk. Weber et al. compared 540 TKAs with a lateral release to 510 TKAs without. Although an increase in patellar fracture rate was observed (1.7% versus 0.8%), the difference was not statistically significant. Ritter and Campbell compared 84 primary TKAs with lateral release to 471 TKAs without. The rate of patellar fracture was found to be lower after lateral release (1.5%) compared with TKA without lateral release (3.6%). Data regarding patella fracture after TKA are difficult to interpret because of the multitude of factors involved in patellar fracture. Although a lateral release may reduce the blood supply to the patella, it may also reduce the forces on the patella in flexion. Kusuma et al. compared the short-term and long-term complication rates in over 1108 knees with or without lateral release. They found no difference between the two groups in regard to complications or Knee Society scores. With an increasing awareness of component rotation and position, the need for lateral release appears to have decreased ; however, most surgeons continue to advocate a lateral release if necessary during primary TKA.

Another factor that may contribute to the rate of patellar fracture is the type of implant used. The choice of patellar implant has been suggested as a factor contributing to patellar fracture after TKA. Inset patellar components, press-fit components, inappropriately sized implants, and fixation peg specifications have all been implicated in contributing to this complication.

The treatment of patellar fractures after TKA focuses on the integrity of the extensor mechanism as well as patellar component fixation. Ortiguera and Berry suggested a simple classification scheme to guide treatment of these fractures. Type I patellar fractures have an intact extensor mechanism and well-fixed patellar component. An abundance of literature suggests that conservative treatment with immobilization in extension is the optimal method of preserving function while avoiding the complications of a surgical procedure. Vertical fractures often do not disrupt the extensor mechanism or patellar button and may be treated conservatively. The preferred method is to immobilize in a cast or brace for 6 weeks in extension with weight bearing as tolerated. This is followed by 6 weeks of passive range of motion followed by strengthening.

Type II patellar fractures involve disruption of the extensor mechanism. A common fracture pattern in this class involves a transverse fracture in the pole of the patella, above or below the component, with a well-fixed component. These fractures are functionally similar to an avulsion of the patellar or quadriceps tendons from the patella and may be treated as described in the previous section. The smaller fragment of patellar bone may be preserved or excised depending on its facilitation or interference with repair. In most cases, a suture technique through vertical drill holes in the patella is more predictable than attempts to reduce and fix the patellar fragment of bone. The suture fixation may be further protected at the inferior pole by a tension band technique using a wire looped through the tibial tubercle. Furthermore, there may be cases when the quality of patellar or quadriceps tendon is not adequate for direct repair alone. This may be tested by observing the integrity of the repair when flexing the knee to 75 degrees. Similar to techniques described earlier, alternative tissue sources may be used to augment the repair in these cases. Unfortunately, complication rates are extremely high (50%) after surgical intervention for Type II fractures, even when the bone fragment is excised and the extensor mechanism is repaired with direct suture technique. Although nonoperative treatment can be considered for these fractures, large extensor lags would be expected.

Type III patellar fractures have an intact extensor mechanism and a loose patellar component. These cases may be treated with immobilization, with the hope that the patient will be asymptomatic after healing occurs, but most often, removal of the component is attempted. Type III fractures are subdivided into Type IIIa, with good bone stock, and Type IIIb, with poor bone stock. When the bone stock is good, removal of the patellar component may be followed by patelloplasty with component revision or resection arthroplasty. When the bone stock is poor, component revision should not be attempted. Instead, component removal should be followed by patelloplasty or patelectomy. The experience with operative fixation of patellar fractures with open reduction and internal fixation is poor, with a high complication rate. Often the fixation does not result in a union, and reoperation is necessary.

Other surgical options are also available for the treatment of patellar fractures. Nelson et al. described the use of hemispheric porous metal implants for primary fracture fixation after the removal of a loose patellar component. However, more investigation is needed before definitive conclusions can be made. When a disruption of the extensor mechanism exists, extensor mechanism allografting can also bypass the need for healing the patellar bone. This surgical technique is the author’s preferred treatment of the disrupted extensor mechanism when alternative methods are not reliable. The method and results of this technique follow.