Objective To investigate associations between deep abdominal muscle activation and long-term pain outcome in chronic non-specific low back pain (LBP).

Methods Recruitment of transversus abdominis and obliquus internus abdominis during the abdominal drawing-in manoeuvre was recorded by B-mode ultrasound and anticipatory onset of deep abdominal muscle activity with M-mode ultrasound. Recordings were done before and after 8 weeks with guided exercises for 109 patients with chronic non-specific LBP. Pain was assessed with a numeric rating scale (0–10) before and 1 year after intervention. Associations between muscle activation and long-term pain were examined by multiple linear and logistic regression methods.

Results Participants with a combination of low baseline lateral slide in transversus abdominis and increased slide after intervention had better odds for long-term clinically important pain reduction (≥2 points on the numeric rating scale) compared with participants with small baseline slide and no improvement in slide (OR 14.70, 95% CI 2.41 to 89.56). There were no associations between contraction thickness ratios in transversus abdominis or obliquus internus abdominis and pain at 1-year follow-up. Transversus abdominis lateral slide before intervention was marginally associated with a lower OR for clinically important improvement in pain at 1-year follow-up (OR 0.76, 95% CI 0.62 to 0.93). Delayed onset of the abdominal muscles after the intervention period was weakly associated with higher long-term pain.

Conclusion Improved transversus abdominis lateral slide among participants with low baseline slide was associated with clinically important long-term pain reduction. High baseline slide and delayed onset of abdominal muscles after the intervention period were weakly associated with higher pain at 1-year follow-up.

Clinical Trial Registration number The study was preregistered in ClinicalTrials.gov with identifier NCT00201513.