Abstract
Introduction
Surgical planning for Patient Specific Instrumentation (PSI) in total knee arthroplasty (TKA) is based on static non-functional imaging (CT or MRI). Component alignment is determined prior to any assessment of clinical soft tissue laxity. This leads to surgical planning where assumptions of correctability of preoperative deformity are false and a need for intraoperative variation or abandonment of the PSI blocks occurs. The aim of this study is to determine whether functional radiology complements pre-surgical planning by identifying non-predictable patient variation in laxity.
Method
Pre-operative CT's, standing radiographs and functional radiographs assessing coronal laxity at 20° flexion were collected for 20 patients. Varus/valgus laxity was assessed using the TELOS stress device (TELOS GmbH, Marburg, Germany, see Figure 1). The varus/valgus load was incrementally increased to either a maximum load of 150N or until the patient could not tolerate the discomfort. Radiographs were taken whilst the knee was held in the stressed position.
CT scans were segmented and anatomical points landmarked. 2D–3D pose estimations were performed using the femur and tibia against the radiographs to determine knee alignment with each functional radiograph and so characterise the varus/valgus laxity
Results
The mean coronal alignment on CT and standing radiographs were 3.8° varus (SD, 5.6°) and 4.3° varus (SD, 6.7°) respectively. Of these, 5 of the knees were valgus aligned and 15 varus aligned in both standing and CT positions. The varus group had a mean of 5.9° in CT and 6.9° varus standing, while the valgus group had means of 4.4° valgus and 5.4° valgus in standing, indicating a collapse into further coronal malalignment while weightbearing.
Each knee in the group had a laxity envelope calculated from the varus and valgus stressed radiographs. In the varus knees, the envelope ranged from 11.0° to 1.0° degree, with a mean of 5.1° (SD, 2.4°). In the valgus knees, the envelope ranged from 10.0° to 5.0° degrees, with a mean of 6.6° (SD, 2.3°), though this difference did not reach statistical significance. Using ±3° of neutral alignment as an indicator of correctable deformity, 7 of the 15 varus knees did not have a correctable deformity, while all of the valgus did.
As determined by laxity limits, the CT and standing alignments were not well centered within their functional radiology groups. Specifically, for the valgus knees, 2 were near the valgus limit (lower quartile) of their laxity envelope, while for the varus knees, 9 were near their varus limit (upper quartile) and 2 at the valgus limit. In total, 65% of the knees did not have their standing alignment well centered on their functional laxity imits.
Conclusions
Varus/valgus laxity in TKA appears to be subject specific and divorced from static radiological parameters. Surgical planning without reproducible clinical assessments of coronal laxity may not be sufficient to obtain a balanced TKA while avoiding ligament releases. Functional radiographs may be a viable method to individualise and refine the surgical plan in TKA on a per patient basis, incorporating objective information normally only available during the surgery itself.
For any figures or tables, please contact authors directly (see Info & Metrics tab above).
