Flexion and extension gaps created by the navigation-assisted gap technique show small acceptable mismatches and...

The aim of this study was to investigate the mechanism underlying the development of gap differences in total knee arthroplasty using the navigation-assisted gap technique and to assess whether these gap differences have statistical significance.

Ninety-two patients (105 knees) implanted with cruciate-retaining prostheses using the navigation-assisted gap balancing technique were prospectively analysed. Medial extension and flexion gaps and lateral extension and flexion gaps were measured at full extension and at 90° of flexion. Repeated measures analysis of variance was used to compare the mean values of these four gaps. The correlation coefficient between each pair of gaps was assessed using Pearson's correlation analysis.

Mean intra-operative medial and lateral extension gaps were 20.6 ± 2.1 and 21.7 ± 2.2 mm, respectively, and mean intra-operative medial and lateral flexion gaps were 21.6 ± 2.7 and 22.1 ± 2.5 mm, respectively. The pairs of gaps differed significantly (P < 0.05 each), except for the difference between the medial flexion and lateral extension gaps (n.s.). All four gaps were significantly correlated with each other, with the highest correlation between the medial and lateral flexion gaps (r = 0.890, P < 0.001) and the lowest between the medial flexion and lateral extension gaps (r = 0.701, P < 0.001).

Medial and lateral flexion and extension gaps created using the navigation-assisted gap technique differed significantly, although the differences between them were <2 mm, and the gaps were closely correlated.

These narrow ranges of statistically acceptable gap differences and the strong correlations between gaps should be considered by surgeons, as should the risks of soft tissue over-release or unintentional increases in extension or flexion gap after preparation of the other gap.