Gait signatures of endurance runners with low back pain: A case controlled cross sectional study

Low back pain (LBP) is an understudied condition among runners, and it is unclear what biomechanical features could be targeted for gait retraining to mitigate pain.

How do running biomechanics differ between healthy individuals and those with running-related LBP?

This was a case-controlled, comparative study design of community runners: running-related LBP (n=52) and healthy controls (n=52). All runners completed running history forms and performed a 3-dimensional gait analysis. Kinematic data were collected using a motion capture system and normalized to a gait cycle, while participants ran on a level grade at self-selected speed on an instrumented treadmill. Current running volume, temporal-spatial, kinetic and kinematic features were compared between groups.

The LBP group had 39.5 % lower weekly distance and 15.4 % fewer were currently training for a race (all p<.05). Runners with LBP demonstrated lower cadence (166±10 step/min vs. 171±9 step/min; p=.05), greater center of gravity lateral displacement (1.4±0.5 cm vs. 1.2 ±.3 cm; p=.044) and greater stride width variability (1.3±0.4 cm versus 1.0 ± 0.04 cm; p=.008). Runners with LBP had a greater Vertical Average Loading Rate ([VALR] 67.7±22.2 bodyweights [BW]/s vs. 62.2±21.5 BW/s; p=.022), and higher joint moments (N*m/(kg*m)) at the knee in the sagittal plane (2.13±0.50 vs. 1.87±0.56; p <.001), frontal plane (1.44±0.39 vs. 1.29±0.29; p=.013), and at the hip in the frontal plane (2.04±0.51 vs. 1.84±0.41; p=.024). No differences were found between groups in the pelvis, hip, knee, and ankle joint excursions in any plane of motion during a typical gait cycle.

These collective motion signature may reflect challenges with control of motion and VALR in the presence of back pain. Cadence training to increase step rate, coupled with core/hip muscle activation, may be an important strategy to reduce motion variability, impact loading rate and pain symptoms while running.