Issue: April 2022
Source: Alm L, et al. Knee. 2017;doi:10.1016/j.knee.2017.08.051.
Disclosures: Frumberg, Jabbouri, Pathak and Petit report no relevant financial disclosures. Fulkerson reports being a consultant for ConMed and president of the Patellofemoral Foundation.
April 20, 2022
8 min read
Save

A 7-year-old girl with progressive left neuromuscular patellar instability

Issue: April 2022
Source: Alm L, et al. Knee. 2017;doi:10.1016/j.knee.2017.08.051.
Disclosures: Frumberg, Jabbouri, Pathak and Petit report no relevant financial disclosures. Fulkerson reports being a consultant for ConMed and president of the Patellofemoral Foundation.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

A 7-year-old girl with asymmetric dystonic diplegic cerebral palsy, hydrocephalus with ventriculoperitoneal shunt and Williams syndrome had progressive left patellar instability. She had a gross motor function classification system level 1.

Steven D. Jones Jr.
Steven D. Jones Jr.
Donald (DJ) Scholten
Donald (DJ) Scholten

Her patella instability developed insidiously in 2 years, becoming severe enough to result in dislocation with any quadriceps activation. Examination revealed a dysplastic patella and lateral patellar tilt. Patella tracking demonstrated a J-sign, with the patella mobile four quadrants laterally at 0° and 30° of flexion (Figure 1). She had mild dystonia in the quadriceps and modified Ashworth scale 2 in the quadriceps and medial hamstrings. Passive range of motion was full. Standing alignment radiographs demonstrated normal mechanical axes. CT of the knee was obtained to characterize the bony dysplasia and, accordingly, demonstrated lateral patellar tracking and a shallow trochlear groove (Figures 2, 3). Tibial tubercle-trochlear groove distance was measured to be 14 mm. Surgical correction of the patient’s left patellofemoral instability was indicated given its interference with mobility, discomfort and risk for worsened dysplasia with continued growth.

left knee showing the patella easily dislocating with knee flexion
1. The left knee showing the patella easily dislocating with knee flexion is shown.

Source: David B. Frumberg, MD
Preoperative anteroposterior, lateral and sunrise views of the left knee demonstrating mild lateral patellar translation and a shallow appearing trochlear groove
2. Preoperative anteroposterior, lateral and sunrise views of the left knee demonstrating mild lateral patellar translation and a shallow appearing trochlear groove are shown.
CT axial view and 3D reconstruction of the patient's left knee demonstrating lateral patellar dislocation
3. CT axial view and 3D reconstruction of the patient's left knee demonstrating lateral patellar dislocation with a hypoplastic/shallow trochlear groove are shown.

What is the best next step in management of this patient?

See answer below.

Medial quadriceps tendon-femoral ligament reconstruction

A femoral nerve block was administered during general anesthesia. A diagnostic left knee arthroscopy was performed using inferomedial and inferolateral parapatellar portals.

The patellofemoral joint had no evidence of loose bodies. However, the equator of the patella and the lateral trochlea both had Outerbridge grade 1 chondral lesions. The patella’s station was four quadrants laterally translated in full extension and it jumped into the trochlear groove at 45° flexion by way of a sulcus in the lateral trochlea. The trochlear groove was dysplastic but present. Medial translation of the patella resulted in significant tilt and lateral tissue tautness. A lateral retinacular release was performed using a hooked cautery device (Figure 4). The tourniquet was deflated and the knee drained of fluid.

Knee arthroscopy image demonstrating the use of a hooked cautery device to release the lateral retinaculum
4. Knee arthroscopy image demonstrating the use of a hooked cautery device to release the lateral retinaculum is shown.

Graft preparation

A posterior tibialis allograft was prepared and ultimately trimmed to 5-mm diameter (Figure 5). One end was whip-stitched with #2 FiberLoop suture (Arthrex). Fluoroscopy was used to identify the distal femur physis and Schottle’s point.

Posterior tibialis allograft prepared and ultimately trimmed to 5-mm diameter with one end whip-stitched with a looped suture
5. Posterior tibialis allograft prepared and ultimately trimmed to 5-mm diameter with one end whip-stitched with a looped suture is shown.

A posteromedial incision was made at the planned location of graft origin. Careful dissection was performed to encounter the adductor magnus at the adductor tubercle. A location slightly proximal and anterior from the Schottle’s point was found fluoroscopically and a Beath pin was inserted (Figure 6). The perichondral ring was sufficiently far from this point and it was protected at all times to remain undisturbed. The anchor point was reamed. The graft was docked into the femoral tunnel with a 5.5-mm Swivelock biocomposite knotless anchor (Arthrex) (Figure 7). As the graft was cycled in tension, the anchor lost purchase; it was upsized to 6.25 mm with excellent purchase. The graft was aligned at the distal aspect of the anchor within the reamed tunnel.

Fluoroscopy demonstrating a Beath pin inserted slightly proximal and anterior from Schottle’s point
6. Fluoroscopy demonstrating a Beath pin inserted slightly proximal and anterior from Schottle’s point is shown.
Intraoperative images demonstrating docking of the femoral tunnel with a Swivelock 6.25-mm anchor
7. Intraoperative images demonstrating docking of the femoral tunnel with a Swivelock 6.25-mm anchor are shown.

Graft passage

A transverse incision was made over the quadriceps tendon insertion and the paratenon was divided. A vertical slit in the distal quadriceps tendon was made mid-substance. A second slit was made at the border of the vastus medialis obliquus (VMO) (Figure 8). The graft was brought up under the VMO through the medial incision. The tendon graft was looped around the medial quadriceps tendon via the central slit and delivered back out the medial slit (Figure 9). Graft tension was marked and cycled to confirm isometry with attention paid to not over constraining the patella. The graft was secured with absorbable and non-absorbable sutures. Excess graft was cut. Cycling confirmed satisfactory isometry and no overtightening; the patella tracked centrally.

Intraoperative images demonstrating the transverse incision over the quadriceps tendon insertion
8. Intraoperative images demonstrating the transverse incision over the quadriceps tendon insertion and the division of the paratenon are shown. A vertical slit can be seen in the distal quadriceps tendon, which was made mid-substance. A second one was made at the border of the VMO.
Intraoperative images demonstrate how the graft was brought up under the VMO through the medial incision
9. Intraoperative images demonstrate how the graft was brought up under the VMO through the medial incision then looped around the medial quadriceps tendon via the central slit and delivered back out the medial slit.

The arthroscope was again inserted into the knee without tourniquet, demonstrating restoration of normal tracking but without overt tension. Incisions were closed and a knee immobilizer was placed. Postoperatively, the patient was allowed to weight-bear as tolerated in a knee immobilizer and worked with physical therapy a few weeks later. At her 1-month follow-up, her incisions were well-healed with no signs of infection, and she had about 85° flexion with excellent patellar isometry (Figure 10). She had some difficulty with straight-leg raise. She was transitioned to a hinge-scoped T-knee brace 0° to 90°, which was locked in extension during weight-bearing.

The 1-month follow-up visit demonstrates well-healed incisions, 85° of knee flexion and a patella with excellent isometry
10. The 1-month follow-up visit demonstrates well-healed incisions, 85° of knee flexion and a patella with excellent isometry. A hinge-scoped T-knee brace is also shown.

Discussion

Patellofemoral instability is not uncommon in the pediatric population and can lead to significant pain, patellofemoral arthritis and disability. It usually presents in young patients with an average age of 10 to 16 years. However, patients with neuromuscular disease, including those with cerebral palsy, may present even earlier due to muscle imbalance, spasticity, weakness and knee joint dyskinesia that potentially increases their risk of patellar dislocation and instability.

Treatment varies from nonoperative to different surgical options depending on presentation, severity of instability, and surgeon preference. During the years, medial patellofemoral ligament (MPFL) reconstruction has been widely adopted and used to treat patellofemoral instability with favorable functional outcomes. However, MPFL reconstruction is known to have an increased risk of complications, including patellar fracture due to patellar drilling and various rates of recurrent instability. MPFL reconstruction in pediatric patients also carries the risk of distal femoral physeal injury given the close proximity of the physis to Schottle’s point. Furthermore, there is variability of the MPFL origin in pediatric patients which could be at, above or below the distal femoral physis. To avoid iatrogenic physeal injury, alternatives to graft anchoring, including the adductor sling and quadriceps tendon turndown, have been demonstrated. Postoperative failure after patellofemoral stabilizing surgery, in general, has ranged from 5% to 30% with no single procedure being clearly superior. Thus, advancements in surgical techniques are needed to improve outcomes and reduce the risk of postoperative complications, especially in patients with neuromuscular pathology who may present with patellofemoral instability at an earlier age.

Short-term reconstruction outcomes

Our case demonstrates a favorable short-term outcome of medial quadriceps tendon-femoral ligament (MQTFL) reconstruction in a 7-year-old patient with neuromuscular pathology which, to our knowledge, has not been documented in the literature. MQTFL reconstruction has been mainly reported in adults with promising outcomes and likely reduces the risk of patella fracture compared with standard MPFL reconstruction techniques. In addition, a biomechanical study that evaluated MQTFL and MPFL reconstruction in adult knee specimens found MQTFL reconstruction more closely restored native resistance to lateral translation. Robert C. Spang III, MD, and colleagues described a combined MPFL-MQTFL reconstruction procedure in children and adolescent patients with an average age of 15 years with avoidance of patellar drilling and favorable short-term outcomes. It is unclear if combined reconstruction of the MPFL and MQTFL is necessary or if isolated MQTFL reconstruction is sufficient for long-term patellofemoral stability. Future study with long-term follow-up is needed to compare these techniques.

Anatomic knowledge of the MQTFL relative to the physis, adductor tubercle and medial femoral epicondyle is instrumental when performing MQTFL reconstruction in the skeletally immature patient as poor isometry is associated with worse functional outcomes. One study found the MQTFL femoral origin in adult patients to originate, on average, 1.8 mm distal to the adductor tubercle. Sheeba M. Joseph, MD, and colleagues describe inserting the MQTFL graft at the distal extent of the adductor tubercle, but above the medial femoral epicondyle in adults. However, in the skeletally immature patient, the exact location of the adductor tubercle in relation to the physis has not been well defined and is likely variable given the differences of MPFL origin discussed in many pediatric studies. Recently, a cadaveric study was done on pediatric specimens to evaluate the anatomy and relationship of the MQTFL attachment to the quadriceps tendon and proximal pole of the patella. However, this study did not mention the location of the physis or comment on the specific origin of the MQTFL. Graft fixation in the metaphysis runs the risk of proximal migration with patient growth. This case utilized a relatively small graft size of 5 mm, potentially enabling it to attenuate with growth.

Neuromuscular etiologies of patellofemoral instability are heterogeneous. This patient’s quadriceps dystonia and underlying bony dysplasia were the major factors for instability. Her dystonia my also impact soft tissue surgical outcomes and a future revision surgery is not out of the question. We believe early realignment will be beneficial to improve function in these children. Several years of follow-up is needed to examine whether growth with this corrected alignment has improved bony alignment. By stabilizing the patella at a young age in a manner that does not over constrain the knee, this procedure has the potential to have a lasting impact for many years to come.