An Unusual Failure of TKA

An Unusual Failure of TKA: Surgical Considerations and Pain Management in the Revision Setting

Alexander Sah, MD

History


A pleasant 61-year-old female presents with a 2 year history of left knee pain. She had previously undergone uncomplicated left total knee replacement at an outside institution. She reports that she was doing well until she fell and broke her left hip in 2012, which was treated locally to her. She feels she has not walked well since that injury, and that her left knee symptoms worsened two years prior to presentation. She does not recall specific trauma. Her orthopaedic surgeon evaluated her, but no diagnosis was made. Overall, the pain has worsened in the past 3 months.

On a pain scale, the knee is rated an 8/10 and is constant. The pain is particularly bad with standing, walking, going up or down stairs, and twisting. The pain is located on the inside and front of the knee. The pain is described as sharp and she also complains of swelling, instability, and limping. She requires Tramadol for pain and other conservative treatment has included activity modification and knee bracing. She uses a walker for ambulation and is mostly housebound. She denies fevers, chills, or signs of infection.

Her past medical history is significant for depression, fibromyalgia, atrial fibrillation, kidney disease, and coronary artery disease. She takes eliquis for anticoagulation.

Preop


Patient is 5’3” and 235lbs, BMI of 41.6, in no acute distress. She is in a wheelchair and is unable to walk without a walker. She has a well-healed midline knee incision. The leg alignment is valgus, with moderate laxity to valgus/varus stress. There is a trace effusion with tenderness to palpation at the medial and lateral joint lines. Her passive knee range of motion is from 0-110 degrees, but there is a 15 degree active extension lag. Due to the size of the leg, it is difficult to assess anterior and posterior knee stability. Her motor, sensory, and vascular exams are intact distally.

Radiographs reveal an absence of space between the components on the AP view.

Full length standing AP film.
Full length standing AP film.
Preoperative AP film shows loss of space between the femoral and tibial components.
Preoperative AP film shows loss of space between the femoral and tibial components.

Her patella is tracking centrally.

Figure 3. Preoperative merchant view.
Figure 3. Preoperative merchant view.

A lateral knee X-ray shows a femoral component in an extended position with the anterior flange underneath the anterior femoral cortex.

Figure 4. Preoperative lateral film shows disassociation of the liner, with the femoral component in extension and notching the anterior femoral cortex.
Figure 4. Preoperative lateral film shows disassociation of the liner, with the femoral component in extension and notching the anterior femoral cortex.
Figure 5. Close-up image of the liner dislocated anterior to the tibial component.
Figure 5. Close-up image of the liner dislocated anterior to the tibial component.

There is more cement present adjacent to the femoral prosthesis than expected, suggesting a prior bone defect. The liner is seen disassociated from the tibial tray and dislocated anteriorly.  The need for revision surgery was discussed with the patient. She understands that the dislocated liner requires revision. We discussed that while it is tempting to replace the liner only, there is a possibility that it may fail again if nothing else is revised. In addition, the malpositioned femoral component places her at risk for continued knee failure, or possible periprosthetic fracture. Furthermore, the suspected bone defects filled with prior cement may require additional stems or augments at time of revision. We also discussed the challenge of managing her postoperative pain given her history of fibromyalgia, pain medication use, and limitations due to existing kidney disease.

Surgery


Preoperative workup was negative for infection by ESR and CRP, and intraoperative synovial fluid analysis, frozen section, and cultures were also negative. Upon exposure of the knee joint, the liner was protruding anteriorly. The liner was severely worn posteriorly. The suprapatellar pouch and medial and lateral gutters were re-established and debrided of inflamed synovium from poly debris. The femoral component was then exposed and its placement in extension was confirmed. The component was removed with care to preserve as much host bone as possible, but there was a large defect anteriorly from the notching of the index procedure and after removal of the prior cement. The tibia was then exposed and removed. There was significant lysis and loss of metaphyseal bone, which was more than expected based on the preop films. The cortex of the tibia was mostly intact and reasonably supportive.

The tibia was prepared for an intramedullary stem with hand reaming. A proximal tibial resection was then performed to establish proximal bone support. Because of the extent of metaphyseal bone loss, the tibia was prepared for a trabecular metal coupled tibial cone augment.

 

Figure 6. Trabecular metal coupled tibial cone fixed to tibial baseplate with stem extension.
Figure 6. Trabecular metal coupled tibial cone fixed to tibial baseplate with stem extension.
Figure 7. Close-up of tibial cone attached to tibial baseplate.
Figure 7. Close-up of tibial cone attached to tibial baseplate.

This provided improved proximal support of the component, maximizing contact to host bone. The femur was then prepared for a stemmed constrained component to bypass the anterior bone defect. Trial reductions were performed and there was appropriate knee motion and stability. A hybrid technique was used that cemented the components into place and a press-fit was used for the stems.

Figure 8. Postoperative AP film showing components with stems and coupled tibial augment.
Figure 8. Postoperative AP film showing components with stems and coupled tibial augment.
Figure 9. Postoperative lateral film.
Figure 9. Postoperative lateral film.

The tibial cone can be placed with or without cement, but was cemented in this case because of the quality of the host bone.

For pain management, intravenous Tylenol was used preoperatively. No NSAIDs were used because of her kidney and anticoagulation contraindications. A periarticular injection of Marcaine with epinephrine was performed for short-acting pain relief, and to possibly minimize bleeding. Liposomal bupivacaine was first injected throughout the posterior capsule, medial and lateral gutters, and periosteum. Because of the size of the patient and the revision exposure, 100ml of diluted injection was used; a large volume allows optimization to create as large of a field block as possible. A 22-guage needle was used to maximize infiltration so that the injection stays within the tissue. Finally, the remaining injection was used in the soft tissues and skin prior to closure.

Postop


The patient tolerated the surgery well. Her pain was well controlled and she was able to ambulate the day of surgery. She had an unremarkable hospital course. She was discharged home the second day after surgery. At her two week appointment her incision was healing well, her pain was much improved compared to the pain she felt pre-surgery, and she was ambulating with a cane. At her 6 week appointment she was taking no pain medications and was walking without an assist device.

Treatment of Severe Genu Valgus Deformity

Treatment of Severe Genu Valgus Deformity with Total Knee Arthroplasty and Peri-articular injection in a patient with previous history of narcotic induced delirium

Ryan M. Nunley, MD

 

HISTORY


A very pleasant 65-year-old female presents to clinic with longstanding bilateral knee pain. Her right knee pain is worse than the left and she notes that she has always had a “knock-knee” appearance to her legs since she was a child. She had a deep brain stimulator placed six months prior to presentation as part of her treatment for Parkinson’s disease. Since that time, she has developed progressive exacerbation of her right knee pain. She has been managed by her primary care physician with physical therapy, oral medications, and she did receive an intra-articular steroid injection that only provided 1 week of relief. She never had any surgeries on her knees. She does mention that she did receive narcotic pain medication following her deep brain stimulator placement and this led to severe narcotic induced delirium requiring a prolonged hospital stay with close observation. She had a 24-hour sitter in her room for 2-3 days before her mental status improved. She comes in seeking potential management with surgery to improve the pain and knee function as she has been told by her physical therapist that her knee dysfunction is limiting her recovery following her deep brain stimulator for her Parkinson’s disease.

 

PRE-OP


Her pre-operative X-rays are shown below.

Figure 1: Pre-op AP image standing short film
Figure 1: Pre-op AP image standing short film
Figure 2: Pre-op lateral image standing
Figure 2: Pre-op lateral image standing
Figure 3: Pre-op Merchant view
Figure 3: Pre-op Merchant view
Figure 4: Pre-op hip-knee-ankle long standing view
Figure 4: Pre-op hip-knee-ankle long standing view

The patient is 5 foot 5 inches, 205 lbs. with a body mass index of 34. On physical examination, the patient is alert and oriented but does have some slowing of her cognitive function as well as involuntary movements consistent with Parkinson’s disease. She also has spasticity and increased muscle tone in her bilateral lower extremities. She has difficulty walking on examination today and requires the use of a wheelchair for long distances and a walker for short distances inside her house and at the grocery store. Her right knee has significant valgus alignment of 20-25 degrees and is only partially correctable. Her right knee range of motion is 10-90 degrees of flexion. Distally she does have sensation intact to light touch through the L2-S1 nerve distributions, 5/5 strength in the quadriceps, hamstrings, tibialis anterior, extensor hallucis longus, and gastrocs. She has palpable pulses in bilateral legs and good capillary refill.

We had a long conversation with the patient and her family about the goals of surgery, the complexity of her knee deformity, and the potential for peroneal nerve dysfunction post-operatively given her severe valgus deformity and knee flexion contracture. We also had a long conversation about the options for her type of anesthesia (general vs. spinal) and how to best treat her post-operative pain to reduce or avoid narcotic medications given her previous history of narcotic induced delirium.

 

SURGERY


On the day of surgery, our anesthesia team was able to successfully place a long acting spinal block and the patient received the routine pre-operative antibiotics. The patient was positioned supine on the operative table and the decision was made to perform her total knee arthroplasty procedure without the use of a tourniquet to attempt to reduce her post-operative thigh pain and need for narcotic pain medications.

During surgery we used minimal retraction on the soft tissue structures throughout the case to reduce surgical induced trauma and inflammation. We used a traditional anterior mid-line surgical incision with a medial peri-patellar arthrotomy to access the knee joint. After adequate exposure we made a distal femoral resection with 3 degrees of valgus and with 2 additional millimeters of bone resection (11 mm total) to ensure that we were contacting bone on the lateral femoral condyle with our distal femoral bone cut.

Given the patient’s relative hypoplasia of her lateral femoral condyle, we used Whiteside’s line and the transverse epicondylar axis to ensure proper rotation of the femoral sizing instrument prior to placing the 4-in-1 femoral cutting block and completing our bone preparation on the femoral side. We then turned our attention to the tibial side and used the standard extra-medullary guides to cut the tibia perpendicular to the mechanical axis. At this point we stopped to assess our flexion and extension gaps and, as expected, the lateral side was tight in both flexion and extension.

Next, we placed a lamina spreader between the femur and tibia and palpated the tight structures. We determined the Iliotibial band (IT Band) was tight so we performed a selective inside-out pie crusting release of the IT band and then reassessed our flexion-extension gaps. The lateral side was still too tight so we replaced the lamina spreader and palpated the tight arcuate fibers in the posterior-lateral corner of the knee. Using a bovie electrocautery device we carefully released these tight fibers, which helped improve the lateral side in extension. We were still too tight in flexion so we performed a selective lengthening of the popliteus which allowed the knee to be balanced. We then resurfaced the patella and then placed the trial implants to assess the range of motion, ligament stability, and patellar tracking. Once we were satisfied with the knee range of motion, soft tissue balance, and kinematics, we removed the trail implants.

Next, we carefully injected the peri-articular tissues with long-acting local anesthetic. The key areas to inject of the peri-articular tissues include the skin edges, quadriceps muscle fibers adjacent to the arthrotomy, periosteum of the distal femur medially and laterally, the residual meniscal beds, and the posterior capsule. Take special precautions when injecting the posterior capsule on the lateral side of the knee to avoid injecting into the neurovascular structures. Our preferred injection technique is to use a large volume (60-100 mL) of long acting anesthetic and to inject approximately 50-80 different locations with each site only receiving a small volume of injections (0.5-1.0 mL). This will help to ensure that a greater number of nerve fibers are covered with the injection. It’s highly recommended to avoid injecting a large amount of the local anesthetic into only a few areas as the local anesthetic does not have the ability to diffuse very far from the site of injection. The size of the injection needle is also quite crucial so to avoid having the injected fluid extravasate out of the tissues once the needle is removed. We recommend using a 22 gauge or smaller needle for the peri-articular injection.

Since this patient has a history of narcotic induced delirium in the past and we know that the knee is generally painful after knee replacement surgery, we hoped to reduce or eliminate the need for narcotics post-operative. To help increase our chances for minimizing post-operative pain and the need for narcotic breakthrough medications, we asked our anesthesia team to augment the patient’s spinal block and intra-operative peri-articular injection with a saphenous nerve block (also known as an adductor canal block). This block is administered in the recovery room prior to the patient being transferred to the inpatient floor.

 

POST-OP


Post-operatively this patient did very well and her pain was managed with oral Tylenol, intra-venous Toradol, and only 2 doses of Dilaudid for breakthrough pain during her physical therapy sessions. The patient was successfully discharged home on post-op day 3 with only Tylenol and Celebrex for pain. She returned to clinic at 4 weeks and was already doing remarkably well from her knee replacement and had not experienced any post-operative complications nor episodes of delirium. She was so excited about her new right knee that she was already inquiring when she could have her left knee replaced. Her four week post-operative radiographs are shown below.

 Figure 5: Post-op AP image standing short film
Figure 5: Post-op AP image standing short film
Figure 6: Post-op lateral image standing
Figure 6: Post-op lateral image standing
Figure 7: Post-op Merchant view
Figure 7: Post-op Merchant view
Figure 8: Post-op hip-knee-ankle long standing view
Figure 8: Post-op hip-knee-ankle long standing view