Effects of unilateral and bilateral experimental low back pain on trunk muscle activity during stair walking in healthy and recurrent low back pain patients.

Aim To explore the trunk muscle activity in healthy and recurrent low back pain (R-LBP) patients with no present pain during stair ascent and descent before and after unilateral and bilateral experimental low back pain (LBP). Methods Twenty-five healthy controls and 25 pain-free R-LBP patients with recurrent mild to moderate LBP were included. All participants completed questionnaires on personal and functional status and Oswestry disability index scoring (ODI). The participants performed maximal voluntary contraction (MVC) and series of 10 stair ascent and descent motor tasks, initiated by the dominant side foot. Bilateral footswitch data and electromyography (EMG) from the dominant side m. iliocostalis, m. multifidus, m. longissimus, m. rectus abdominis and externus and internus obliquus muscles were recorded before and during experimental unilateral and bilateral LBP were induced in randomized order with a break between the sessions of at least 10 minutes or until the pain was absent. The pain intensity was scored on a numeric rank scale (NRS) and the perceived pain areas were recorded by drawings immediately after the sessions. The motor tasks were separated in 3 phases (toe-off, initial contact and double stance) detected by the footswitch signals for both dominant and non-dominant leg in ascent and descent stair climbing. The raw EMG data were bandpass filtered (10-500 Hz), full-wave rectified, smoothed and normalized to MVC values. The mean Root-mean-Square (RMS) EMG of the 3 ascent and 3 descent phases were extracted and mean differences between baseline and pain sessions were calculated (delta-RMS-EMG) and compared between groups and sessions. Results Mean ODI in the control group was 20.5% (20-23%, minimal disability) and in the R-LBP group it was 33.0% (26-60%, moderate to severe disability). Mean NRS was higher in the R-LBP group during both pain conditions compared with controls (P<0.001) and after bilateral compared with unilateral pain in the control (P<0.02) and R-LBP group (P<0.01). Pain areas were larger after bilateral compared with unilateral pain in the control (P<0.001) and R-LBP group (P<0.001) with no difference between the groups. The baseline RMS-EMG activity was higher in the R-LBP group compared with the control group in the back (P<0.03) and abdominal muscles (P<0.04). In the R-LBP group delta-RMS-EMG decreased in the 3 back muscles and increased in m. rectus abdominis during all phases, with larger decrease during bilateral compared with unilateral pain (P<0.01). In the control group, delta-RMS-EMG increased in all muscles during all phases with larger increase during bilateral compared with unilateral pain in the back muscles (P<0.04). Conclusions The impact of unilateral and bilateral experimental LBP on trunk muscle activity was different between healthy participants and R-LBP patients. Pain resulted in increased trunk muscle activity in healthy, while R-LBP patients decreased the back and increased the abdominal muscle activity. However, the baseline activity in the back and abdominal muscles in the R-LBP patients was higher and the changes during pain probably can be considered as a strategy to diminish the impact of the local back muscle pain. The clinical implication is that trunk muscle activity during pain must be addressed in the physical examination, but it remains unknown if the observed changes are appropriate strategies in relation to the pain condition. Acknowledgement The study was supported by CNAP, Aalborg University and UCN Department of physiotherapy, Denmark.