Special Poster Session 51st International Society for the Study of the Lumbar Spine Annual Meeting 2025

Feasibility and effectiveness of real-time biofeedback for maintaining lumbar lordotic curvature using inertial measurement system (116164)

Sun Gun Chung 1 , Juchan Kim 1 2 , Keewon Kim 1
  1. Seoul National University and Hospital, Jongno-Gu, SEOUL, South Korea
  2. Department of Rehabilitation Medicine, , National Traffic Injury Rehabilitation Hospital, Yangpyeong, South Korea

Introduction:

Recent studies have shown that decreased lumbar lordotic curvature (LLC) is closely related to low back pain (LBP), leading to increased disc pressure, posterior migration of the nucleus pulposus, and eventually causing damage to the posterior annulus. By these mechanisms, activities that increase the frequency, duration, or magnitude of lumbar flexion, such as prolonged sitting or repetitive lifting, are known to elevate the risk of LBP. Therefore, focusing on the LLCpreservation in daily activities is crucial in the prevention and management of LBP. However, managing LLC in daily life has been challenging due to limited applicable methods, with most interventions restricted to postural education. Recent advancements in inertial sensor have enabled the real-time monitoring of LLC, offering a promising solution to these limitations. This study aims to evaluate the feasibility and effectiveness of a real-time biofeedback system for preserving LLC using an inertial measurement system, comparing it with postural education and their combined effects.

 

Methods:

Ten male participants with LBP wore clothing equipped with two inertial sensors at T6 and S2 and performed four daily tasks, including sitting at a computer, mopping the floor, picking socks up from the floor, squat for exercise, across four sessions undertaken in two separate days, with varying combinations of auditory feedback and postural education. The relative angles from the sensors for the mediolateral axis were monitored in real-time as lumbar lordotic angle (LLA), with auditory alarms triggered when LLA dropped below set thresholds. Postural education was delivered using instructional leaflets. The average deviation in LLA (LLAdev) was extracted for each session and task, and regression analysis was used to analyze the impact of education and feedback on increasing LLAdev, which indicates preservation of LLC.

 

Results:

As shown in Figure 1, for sitting and mopping tasks, both feedback and education demonstrated significant effects with similar impacts. In the sitting task, auditory feedback and postural education increased an additional 10.2˚ and 10.1˚ of LLAdev, respectively (p<0.001 for both). In the mopping task, postural education had a slightly greater effect, increasing LLAdev by 13.7˚ compared to the feedback’s 10.5˚ (p<0.001 for both). For picking up and squatting tasks, education had a greater impact. In picking up tasks, education significantly increased LLAdev by 11.7˚ compared to feedback’s increase of 5.8˚ (p<0.001 for both). In squatting, only education was effective, increasing LLAdev by 3.5˚ (p<0.001).

 

Discussion:

The real-time biofeedback system employing two inertial sensors was found feasible and effective for LBP patients in preserving LLC across various tasks, excluding the squatting task. In continuous tasks such as sitting and mopping, the feedback effects were comparable to those from postural education. While the impact of feedback was more modest in the lifting task, it still significantly contributed to preserving LLC. Importantly, the best preservation of LLC was observed when feedback was combined with postural education. These results highlight the potential benefits of integrating real-time feedback and education with traditional treatments.

674aa7ad20354-Figure+1+LLA+dynamic.png