Background/Purpose: The mechanisms underlying the transition from acute to chronic low back pain (cLBP) are poorly understood. Physiological and psychological factors are implicated. Although significant associations have been found between neuromuscular control of the lumbar spine and the level of fear of pain, it is still unknown whether acute exposure to fear of pain alters trunk motor control. The objective of this study was to determine if experimentally induced pain expectations modulate trunk neuromuscular responses differently in subjects with and without cLBP.

Methods: This cross-sectional study included 22 patients with cLBP and 22 healthy participants. They performed 6 trunk flexion-extension tasks under three experimental conditions: innocuous heat, noxious stimulation with low pain expectation and noxious stimulation with high pain expectation. Noxious stimulation was generated by thermal cutaneous heat stimulations in the lumbar region (L4-L5), whereas low or high pain expectations were generated by verbal and visual instructions (see Fig. 1). After each task, experimental pain was evaluated using a numerical rating scale (NRS). Surface electromyography (sEMG) of erector spinae at L2-L3 and L4-L5 as well as lumbopelvic kinematic variables were collected during the tasks. Pain ratings, sEMG and kinematic variables were compared between groups and conditions using two-way mixed ANOVAs. Pearson’s correlation coefficients were calculated in cLBP patients to determine whether the effects of expectations were associated with disability, pain catastrophizing, state and trait anxiety and fear avoidance beliefs.

Fig. 1 Study design

Results: Pain ratings were significantly different between high and low pain expectations conditions (P<0.001). This difference was similar between patients with cLBP (15.2 ±13.4) and control participants (13.3 ± 10.2). In patients with cLBP, the increase in sEMG activity in full flexion caused by expectation was related to higher pain catastrophizing, but not to disability, anxiety and fear-avoidance beliefs. Two-way mixed ANOVA yielded a significant “group x condition” interaction for sEMG in Full flexion (P<0.05). Planned comparisons revealed a stronger effect of pain expectation in healthy participants than in patients with cLBP. Lumbopelvic rhythm was significantly different between groups (P<0.05), but similarly affected by pain expectation.

Conclusion: As anticipated, the increase in sEMG activity caused by expectations was related to higher pain catastrophizing in patients with cLBP. Nevertheless, expectations of high pain resulted in neuromuscular adaptations that were weaker in patients with cLBP than in healthy participants. In conclusion, chronic pain appears to generate rigid and less variable movement patterns in patients with cLBP, which attenuate their response to acute fear of pain exposure.