Knee Replacement Technique

Early in my orthopedic residency training at University of Arkansas for Medical Sciences in the late 1990’s we were trained by one of the top joint surgeons in the country, Dr. Carl Nelson. He taught us about posterior stabilized TKA, modern cementing technique, laminar flow and soft balancing for a total knee arthroplasty. Each knee was cut with a 6 degree distal femoral valgus cut and the femoral external rotation was set to 3 degrees in a varus knee and 5 degrees in a severe valgus knee. We then proceeded to do extensive soft tissue releases using Insall’s techniques which often led to the use of thick polyethylene constrained spacers and we often had to do a lateral retinacular release for patellar balancing. It always bothered me if the knee wasn’t perfectly balanced. 

After completing residency , I worked part-time at the VA staffing joint replacement surgeries for a few years as I was building my local orthopedic practice. I was able to work with the UAMS orthopedic residents during their VA rotations and fine tune my surgery techniques. It bothered me even more if a knee wasn’t perfectly balanced when I was the staff surgeon in charge of the outcome. I was always a believer in correcting knee deformity more with bone cuts to minimize soft tissue releases. I began slightly varying the the distal femoral transverse cut angles to match the deformity. For an extreme varus knee (anatomic axis < 0), we would cut the distal femur at 4 degrees of valgus. For a valgus knee (anatomic axis > 5), we would cut the distal femur at 6 degrees of valgus. Anything in between was cut at 5 degrees of valgus. This decreased the soft tissue balancing needed for the extension gap. I felt this was ok to do because one of my prior mentors, Dr. David Collins routinely cut his distal femur’s at 6 degrees of valgus. We also began using the epicondylar axis as a guide for external rotation as I had worked with Dr. Walter Selokovich prior to his passing and he routinely used this technique to better balance the flexion gap at 90 degrees of flexion. I felt that if we could balance the knee at 0 degrees flexion and then at 90 degrees of flexion, then everything in between would be balanced as well. We also developed a concept called rotational matching which matched the tibial component rotation to the femoral rotation. We would trial the femoral and tibial components with a constrained spacer and a freely rotating tibial trial to find the midpoint of where the tibial component was best aligned and mark that point on the tibia prior to cementing. 

After completing my time at the VA , I had a rapidly growing joint replacement practice and continued to evolve my customization technique with new technology that was developing in the field of TKA customization and gap balancing. In the mid 2000’s, many companies were trying to customize the TKA to fit the individual patient. Otis Med used some extreme CT guided techniques to completely match the bony cuts to a patients deformity. The Biomet Signature knee used CT scans to make custom cutting guides for implantation of the implants. One company, Zimmer, developed a gender specific knee replacement that accommodates for the variability in male and female knee anatomy. The kinematic knee technique was also evolving at this time to make custom cuts on both the tibia and femur to match a patients pre-existing deformity. 

At this point in time, I felt justified in striving to customize each total knee using conventional implants. I began using a device called Synvasive to customize the flexion gap (external rotation varied 0-9 degrees). I also developed an algorithm to cut the distal femur (extension gap) from 4-7 degrees depending on the deformity and that later evolved to 3-8 degrees. The goal was to create a balanced rectangular flexion and extension gap. The algorithm doesn’t always perfectly balance the extension gap. Occasionally, I will add another degree to the distal femoral cut if it is too tight in one direction or another or if it is close just some mild soft tissue balancing. By 2005 , I began working closely with a team at Biomet (Warsaw, Indiana) to make some of my own devices to be used for measuring femoral external rotation and making a custom cut. The goal here was to create a balanced rectangular flexion gap. We obtained a joint patent in 2008 for a device that I call the “Spider”. The Spider was used not only to set the custom femoral external rotation for the flexion gap, but also had a built in sizer for the Biomet Vanguard femoral component. I actually prefer the prototype devices which are left and right specific over the patented universal device. 

During this time I was using the Biomet cemented Vanguard Posterior Stabilized Knee which had a universal tibia component. I combined the extension gap algorithms (balancing with the knee extended), with the Spider (balancing with the knee flexed to 90 degrees) and the rotational matching component. I also used an algorithm for knee flexion contractures and recurvatum deformity. I cut 1 mm more from the distal femur for each 5 degrees of flexion contracture and 1 mm less for each 5 degrees of recurvatum deformity. I began calling it the Patient Specific Knee (PST). We saw a dramatic decrease in collateral ligament instability, patellar component maltracking (fewer lateral releases), extent of soft tissue releases, post-op anemia and transfusions, and varus/valgus laxity. The patients rehabilitated better and were happier than the patients from the earlier years with extensive soft tissue releases. We weren’t seeing many wobbly knees. 

Now in 2025, I am still using the Patient Specific Knee technique, but with a different implant. I now use the Zimmer-Biomet Persona cemented knee which has a Medial Congruent (MC) polyethylene component and we no longer have to use the rotational matching process as these components are left / right specific. PST can also be used for a non-cemented TKA which is often done by my partner at Arkansas Bone & Joint , Dr. BJ Bailey. I use the technique manually and then check it with the Zimmer Biomet ROSA Robot (first generation robot). We use the robot as a way to assess final knee range of motion, collateral ligament stability, mechanical axis and to help pick the best spacer size. I still do the custom intramedullary femoral cuts, neutral extramedullary tibial cuts, and custom femoral external rotation cuts with the Spider. We manually check the gaps with blocks and adjust as needed. There are a few extra steps to incorporate PST, but I enjoy using emerging technologies. We call it TKA with PST and ROSA Robotic Assistance. Also patients enjoy emerging technologies. 

I think the first generation Robot is a nice measuring tool for any technique, but I prefer my present methods to those of doing the entire surgery with the robot. I do think any orthopedic surgeon using any technique can benefit from the feedback given by a robot to see how well their TKA’s are balanced and get confirmation or motivation to fine tune their existing technique. I also believe those surgeons using the Robot completely, should have a way to manually customize the procedure when the Robot isn’t working or available.