# 3D-printed shoe soles with structural properties towards reducing knee adduction moments during walking

## TL;DR

This study shows that 3D-printed shoe soles can reduce knee stress better than traditional insoles, which could help manage knee osteoarthritis.

## Contribution

Custom 3D-printed shoe soles with variable stiffness and soft heel designs are introduced as a novel alternative to conventional lateral wedged insoles.

## Key findings

- SH shoes reduced first and second peak knee adduction moments by 10.5% and 8.6%, respectively.
- VS and SH soles reduced knee adduction angular impulse more effectively than 5° lateral wedged insoles.
- GRF–knee lever arm and stance time were key contributors to the effectiveness of the 3D-printed soles.

## Abstract

Lateral wedged insole (LWI) is a representative foot orthosis for the conservative management of knee osteoarthritis. However, recent research indicates the heterogeneity of LWIs’ biomechanical effectiveness, indicating limitations in conventional LWI design and the need to develop new orthoses with improved effects. This study evaluated the efficacy of custom-designed 3D-printed shoe soles in reducing knee adduction moment (KAM), compared with conventional 5° LWIs.

Shoe soles were 3D-printed with a gyroid infill structure to allow adjustable stiffness, generating a variable stiffness (VS) sole with lateral-medial stiffness differentiation, and a soft heel (SH) sole incorporating the same differentiation plus a soft heel. Twenty-one healthy adults performed gait assessments wearing control shoes with neutral insoles, control shoes with 5° LWIs, VS, or SH shoes. Kinematic data and ground reaction forces (GRFs) were captured. Gait parameters, joint kinematics and kinetics were analysed. Repeated measures ANOVA assessed footwear effects, and multiple linear regression identified key contributors to KAM variation. Statistical significance was set at P < 0.05.

KAM varied significantly across footwear. SH shoes reduced 1st and 2nd peak KAMs by 10.5% (P = 0.001) and 8.6% (P = 0.032); VS shoes reduced the 2nd peak by 9.1% (P = 0.014). Both VS and SH reduced knee adduction angular impulse (KAAI) by 9.7% and 11.8% (P < 0.001). 5° LWIs only reduced 1st peak KAM by 7.9% (P = 0.009). GRF–knee lever arm was the main contributor to 1st peak KAM changes in VS shoes, while SH shoes’ effects involved both GRF and lever arm. For 2nd peak KAM, the lever arm was dominant contributor; for KAAI, lever arm and stance time were main contributors.

3D-printed soles reduced KAM variables more effectively than 5° LWIs, supporting their potential for biomechanically optimised footwear upon further assessments on individuals with knee osteoarthritis.