Introduction
Patellar instability is a common cause of knee dysfunction in adolescents and young adults, with primary patellar dislocations occurring at an incidence of 42 per 100,000 person-years. Management after first-time dislocation can include conservative measures such as nonsteroidal antiinflammatory drugs, bracing, and physical therapy. In cases of recurrent patellar instability, surgical stabilization may be indicated. Determining the appropriate surgical procedure requires the assessment of soft tissue deficiency and bony malalignment. In cases of soft tissue deficiency, medial patellofemoral ligament reconstruction (MPFLR) may be performed. In cases of bony malalignment, tibial tuberosity osteotomy (TTO) and occasionally sulcus-deepening trochleoplasty (SDT) may be indicated. In this chapter, we will highlight the complications that have been reported with these surgeries and will discuss strategies for avoiding them.
Preoperative Planning
Identifying the Optimal Operation
Patella instability can be caused by a combination of anatomic and physiological factors. Identifying the primary contributing pathology is crucial in determining the appropriate procedure to stabilize the patella and minimize the risk of recurrence. Trochlear dysplasia, patella alta, a lateralized tibial tuberosity, and medial soft tissue incompetence are the most common contributing factors. Patients should be evaluated with a thorough history and physical examination, including an assessment of patellar stability. The glide test is used to quantify patellar translation, and the apprehension test indicates that the excessive laxity is symptomatic. It is important to rule out medial instability, which is seen most often in the setting of prior lateral retinacular release. Injury to other structures should be ruled out with a comprehensive ligamentous examination.
All patients under consideration for surgical stabilization require a radiographic workup. Lateral radiographs can be used to assess patellar height and trochlear dysplasia. Axial radiographs also demonstrate trochlear morphology, as well as persistent patellar subluxation. Anteroposterior and tunnel views are useful for identifying loose bodies. Full-length standing views show coronal alignment. Advanced imaging, including magnetic resonance imaging (MRI), computed tomography (CT), and dynamic studies, provides additional information which helps define the pathoanatomy and identify the optimum surgical solution.
MRI or CT axial imaging is used to measure the tibial tuberosity to trochlear groove (TT–TG) distance and to further characterize the patient’s osteology. For patients with TT–TG distance greater than 15 to 20 mm, TTO should be considered to correct the excessively lateralized tibial tuberosity. If the increased TT–TG distance is not addressed, the patient has a higher risk of recurrent instability. MPFLR alone has been shown to correct patellar tilt for knees with a TT–TG distance of up to 15 mm, but in knees with a TT–TG distance greater than 15 mm, MPFLR alone results in altered patellofemoral contact pressures and increased patellar translation. Combination MPFLR and TTO has been shown to be successful in this setting, with recurrent instability rates comparable to or lower than the isolated procedures. ,
SDT, although not commonly performed in the United States, may be a useful supplemental procedure for some patients because recurrent instability after MPFLR has been reported in cases of severe trochlear dysplasia. , , One study reported that MPFLR patients demonstrated lower functional scores if they had trochlear spurs greater than 5 mm in size. SDT for Dejour types B and D has been noted to have significantly better outcomes and less pain with sports when compared with types A and C, suggesting that trochleoplasty should be used primarily for the removal of a trochlear spur. Patients undergoing trochleoplasty experience a high reoperation rate of 14% to 17%, largely because of recurrent instability. , In some cases, this is owing to large TT–TG distances or medial patellofemoral ligament (MPFL) incompetence that goes unaddressed at the time of surgery. For patients with large trochlear spurs and an incompetent MPFL, MPFLR combined with SDT has been shown to produce good outcomes. Referral to a subspecialist with experience in trochleoplasty may be appropriate for these patients.
In the pediatric population, there are additional considerations regarding approach to the patient and choice of surgical stabilization procedure. Hemiepiphysiodesis may be necessary in patients with excessive valgus, and derotation osteotomy for those with excessive anteversion and/or tibial torsion. TTO and trochleoplasty are not options in skeletally immature patients because of the risk of physeal arrest and subsequent deformity. The distal femoral physis is at risk for injury with femoral graft placement during MPFLR.
As with all knee reconstruction surgery, prophylactic intravenous antibiotics should be administered within 60 minutes of making the incision. A second dose should be considered in the case of longer procedures which approach 1 to 2 half-lives of the antibiotic administered. Prophylactic antibiotics administered beyond 24 hours postoperatively are not warranted for elective orthopedic cases.
Intraoperative Complications
Medial Patellofemoral Ligament Reconstruction
Reconstruction of the MPFL provides the patient with a medial soft-tissue checkrein to resist excessive lateral patellar translation. The authors of a review of MPFLR complications reported a cumulative complication rate of 26.1%; another review reported a rate of 16.2%. In total, 47% of these complications were attributed to suboptimal surgical technique and were considered preventable. Preventable complications included cases of recurrent instability, patellofemoral arthrosis/pain, and patella fracture caused by improper femoral tunnel placement or patellar tunnel technique. Choosing the correct points for graft fixation and the correct graft length are of special significance so as to avoid overconstraint of the patellofemoral joint.
Femoral Tunnel Position
The femoral origin of the MPFL lies within a bony saddle formed by the medial epicondyle and the adductor tubercle. With a femoral tunnel that is too proximal or anterior, patellofemoral contact pressures increase with the knee flexion. Adequate exposure at the time of surgery can be helpful in properly identifying the bony landmarks. Intraoperative fluoroscopy is a useful tool to determining appropriate femoral tunnel position. Schöttle described a femoral starting point on a lateral radiograph by identifying a point 1.3 mm anterior to the posterior cortical line and 2.5 mm distal to the posterior origin of the medial femoral condyle ( Fig. 22.1 ). A perfect lateral on intraoperative fluoroscopy is essential in using fluoroscopic guidance, as 5 degrees of obliquity has been shown to result in a starting point that is 8 to 9 mm off-target ( Fig. 22.2 A-D ). Keeping the knee in a stable position, the image should be coned and centered, with distal and posterior femoral condyles perfectly overlapping. A Kirschner wire or 2.5-mm drill bit should be placed at the proposed starting point and used to confirm decreasing graft tension with knee flexion greater than 60 degrees before drilling the femoral tunnel.
Patella Fracture
Patella fractures are one of the most serious complications of MPFLR. When drilling a patellar tunnel, care must be taken not to violate either the anterior or posterior cortices of the patella. Violation of the anterior cortex can create a stress riser, which may predispose the patella to fracture ( Fig. 22.3 A, B ). , Violation of the subchondral bone and cartilage may lead to arthrosis. Fluoroscopic guidance is helpful in the intraoperative assessment of patellar tunnel placement, with the use of a drilling guide or cannulated drill bit to confirm a safe trajectory. Drilling smaller-diameter patellar tunnels, avoiding multiple tunnels, and using short blind tunnels, as with the docking technique, decreases fracture risk. Dissection on the superior patella should be limited so as to avoid disruption of the blood supply. To avoid fracture altogether, the graft can be fixed directly onto the medial edge of the distal quadriceps tendon. We prefer a technique where the graft is docked into a short 15-mm blind tunnel to reduce the risk of fracture.
Failure of Fixation
Numerous biomechanical studies have been performed to evaluate the strength of MPFLR graft fixation. Use of an intrapatellar bone bridge with a double-bundle graft has resulted in a 60% failure rate in cyclic testing at a lower load to failure than the native MPFL. In one study, suture anchor fixation in the patella was found to be significantly weaker than native MPFL, whereas suspensory cortical fixation and interference screw fixation was found to be stronger. A review of MPFLR graft fixation techniques demonstrated that suture fixation may permit excessive translation because one in four patients retained the feeling of hypermobility or instability. Tunnel fixation, by contrast, demonstrates hypermobility or patellar apprehension in 8.6% of cases and redislocation or resubluxation in 1.2% to 3.3% of cases. , In cases where initial attempts at patellar fixation with an implant fail, use of either the docking technique or cortical button suspensory fixation on the lateral side of the patella are both reasonable backup options. In situations where bony fixation is not adequate, suture fixation to the quadriceps tendon as described by Fulkerson can be performed as well.
Graft Length
Even with ideally positioned attachment points, inappropriate graft length may result in suboptimal outcomes after MPFLR. Grafts fixed with excessive tension result in overconstraint of the patellofemoral joint, whereas grafts fixed too loosely result in recurrent instability. In one biomechanical study, increased medial contact pressures and medial tilt of the patella were avoided by tensioning the graft with 2 N of force in 30 to 60 degrees of knee flexion. One cadaveric study revealed that strain patterns of the native MPFL were best replicated with fixation at 45 degrees of flexion. Authors of another paper reported that graft length variability from errant femoral starting points could be minimized when the graft was fixed at 30 degrees. Researchers using computational modeling showed that setting the graft length to allow 0.5 to 1 quadrant of lateral patellar translation at 30 degrees successfully prevented overconstraint of the medial patellofemoral joint. Direct arthroscopic visualization of the patellofemoral joint at the time of graft fixation has been shown not to change outcomes, patellar tilt, or patellar congruence angle. The preferred angle to fix the graft is at the greatest amount of graft tension, usually between 40 and 60 degrees of knee flexion. This prevents excessive tightness of the graft and overconstraint of the patellofemoral joint.
Painful Hardware
In a recent systematic review of MPFLR complications, reoperation rates were reported at 3.1%. Overall, 3.0% of patients reported painful hardware, which typically resulted from prominent staples or screws at the femoral attachment, although painful patella hardware has also been reported. , In total, 1.1% of patients have their hardware removed, a complication which may be minimized by ensuring that hardware is not left prominent at the time of surgery. , Wound complications are very rare, with a 0.1% combined risk of hematoma and dehiscence, and a wound infection rate below 0.01%.
Physeal Disruption
In skeletally immature patients, the femoral origin of the MPFL lies 10 mm distal to the physis and becomes slightly more distal with age. In one study, femoral tunnels that were drilled with less than 10 degrees of distal angulation violated the physis 41% of the time. Aiming the femoral tunnel approximately 15 to 20 degrees distal and 15 to 20 degrees anterior has been shown by computational modeling to be the safest path to avoid violation of the physis, the notch, and the cartilage ( Fig. 22.4 A, B ).
Tibial Tuberosity Osteotomy
TTO is customized to provide medialization, anteriorization, distalization, or a combination of the three. Medialization achieves the primary goal of extensor mechanism realignment in the coronal plane. Anteriorization decreases patellofemoral joint contact forces and unloads distal chondral lesions. Distalization can bring the patella more inferior so as to promote engagement in the trochlear groove at a lower flexion angle.
Wound Complications
The incision should be made just lateral of the tibial crest, thereby avoiding direct pressure on the incision with kneeling. Authors of a recent review article reported that wound complications such as hematoma and dehiscence occur at a rate of 0.8%. Superficial wound infections have been noted at a rate of 0.9%, whereas deep infections requiring hardware removal are rare at 0.1%. Wound complications can be minimized with cautious handling of the soft tissues and by avoiding excessive anterior displacement of the TTO shingle. Patients should expect decreased sensation lateral to the incision, as the infrapatellar branch of the saphenous nerve is routinely transected. Saphenous neuroma is a rare complication at 0.1% of cases.
Delayed Union and Nonunion
TTO is performed using either a saw or osteotomes. Tuberosity fracture is prevented by ensuring adequate shingle length of 5 to 6 cm and thickness of at least 5 mm. This ensures good bony compression and provides enough bone stock to prevent rare fracture of the tuberosity fragment when lag screws are placed. Thermal necrosis and periosteal stripping should be avoided as much as possible to maintain vascularity of the fragment. When making the transverse cut, the underside of the patellar tendon should be protected with a retractor. Whenever possible, the screws should be inserted perpendicular to the plane of osteotomy. Nonunion of the osteotomy occurs in 0.2% to 1.0% of cases. However, with complete detachment of the shingle, as is necessary with distalization, the risk of delayed union increases, and the risk of nonunion increases to 2.4% ( Fig. 22.5A‒D ).
