We acknowledge correct performance of a TKR requires more than the accurate bony cuts delivered by navigation. Soft tissue balancing is paramount in the delivery of a satisfied patient and a symptom-free arthroplasty. Traditionally, surgeons manually assess soft tissue balance based upon their experience and their tactile sense that a ligament that is too tight or too loose. Whilst this technique is easily taught, it is neither measurable nor precisely reproducible from surgeon to surgeon. Several devices have been designed to measure and evaluate soft tissue tension and balancing – these include tensors, spacers and a recently introduced pressure-sensitive micro-electric instrument [3, 13, 26]. We prefer to measure laxity and balancing with navigation data and have utilised data from computer navigation for the past 10 years for real-time information on extension, flexion and joint play with gratifying results. We acknowledge that whilst computer navigation provides a reproducible and objective measurement of soft tissue balancing, the desirable and acceptable gaps in the medial and lateral compartments along the flexion arc remain yet to be determined and may be variable between patients with different degrees of ligamentous laxity.

In a case series of 90 TKRs, Saragaglia et al. [27] evaluated the role of computer navigation to predict whether soft tissue release would be required in TKR. Fully correctable deformities as shown with navigation were identified, and releases that may have been routinely performed were avoided. The incidence of medial release for genu varum was 17.6% compared with Engh’s estimated incidence of greater than 50% [28].

Matsumoto et al. compared computer navigation with a ligament tensor for soft tissue balancing in 30 TKRs analysing at 0° and 90° of flexion and confirmed the accuracy of computer navigation for soft tissue assessment when compared to tensors, without a statistically significant difference between the two methods [29]. In a randomised controlled trial, Joseph et al. [30] demonstrated that computer navigation is more accurate than conventional techniques for soft tissue balancing of the mediolateral extension space, utilising navigation software as a gold standard, but not the flexion space nor between the flexion and extension spaces.

Song et al. [31] reported a comparative study of medial and lateral laxity utilising stress radiographs on 86 conventional and navigated TKRs. They found no significant difference between the groups with an average of 3.5° laxity to valgus force and 4.4° for varus at full extension in the navigated group and, respectively, 4.0° and 4.2° in the conventional group. Similarly there was no significant difference in HSS scores and final range of motion between conventional and navigated knees.

Pang et al. [32] demonstrated more precise soft tissue balancing with computer-assisted surgery than conventional balancing. They demonstrated improved function scores in the navigated group at 6 months and 2 years, lower rates of residual flexion contracture and less malalignment outliers.

The role of computer navigation in TKR is well established in the literature and is the most accurate means of obtaining desired alignment in the sagittal and coronal planes. Its role in soft tissue balancing is more recent with the use of navigation data allowing the surgeon to recognise imbalance including mid flexion instability, tightness in flexion, failure to resolve flexion contractures and persisting pathological recurvatum.

Navigation-derived information allows the surgeon a real-time assessment across the entire flexion arc, and, importantly, objective measurements enable the surgeon to then address abnormalities revealed by navigation that may otherwise compromise the patient’s outcome and confirm resolution before the patient leaves the operating room.