An Instrumented Hammer to Detect the Bone Transitions During an High Tibial Osteotomy: An Animal Study

TL;DR

This study introduces an instrumented hammer with an algorithm to accurately detect bone transitions during osteotomy, potentially improving surgical safety by preventing hinge rupture.

Contribution

The paper presents a novel instrumented hammer and detection algorithm that accurately identifies bone transitions during osteotomy, outperforming surgeon proprioception.

Findings

Algorithm detects bone transitions within 1-4 impacts of video measurement.

Detection always occurs before sample rupture, indicating safety margin.

Method shows promise for real-time surgical guidance.

Abstract

High tibial osteotomy is a common procedure for knee osteoarthritis during which the surgeon partially opens the tibia and must stop impacting when cortical bone is reached by the osteotome. Surgeons rely on their proprioception and fluoroscopy to conduct the surgery. Our group has developed an instrumented hammer to assess the mechanical properties of the material surrounding the osteotome tip. The aim of this ex vivo study is to determine whether this hammer can be used to detect the transition from cortical to trabecular bone and vice versa. Osteotomies were performed until rupture in pig tibia using the instrumented hammer. An algorithm was developed to detect both transitions based on the relative variation of an indicator derived from the time variation of the force. The detection by the algorithm of both transitions was compared with the position of the osteotome measured with a video camera and with surgeon proprioception. The difference between the detection of the video and the algorithm (respectively, the video and the surgeon; the surgeon and the algorithm) is 1.0$\pm$1.5 impacts (respectively, 0.5$\pm$0.6 impacts; 1.4$\pm$1.8 impacts), for the detection of the transition from the cortical to trabecular bone. For the transition from the trabecular to cortical bone, the difference is 3.6$\pm$2.6 impacts (respectively, 3.9$\pm$2.4 impacts; 0.8$\pm$0.9 impacts), and the detection by the algorithm was always done before the sample rupture. This ex vivo study demonstrates that this method could prevent impacts leading to hinge rupture.