Association of lumbar spine stiffness and flexion-relaxation phenomenon with patient-reported outcomes in adults with chronic low back pain – a single-arm clinical trial investigating the effects of…

Ting Xia, Cynthia R. Long, Robert D. Vining, Maruti R. Gudavalli, James W. DeVocht, Gregory N. Kawchuk, David G. Wilder, Christine M. Goertz

Spinal manipulation (SM) is used commonly for treating low back pain (LBP). Spinal stiffness is routinely assessed by clinicians performing SM. Flexion-relaxation ratio (FRR) was shown to distinguish between LBP and healthy populations. The primary objective of this study was to examine the association of these two physiological variables with patient-reported pain intensity and disability in adults with chronic LBP (>12 weeks) receiving SM. A single-arm trial provided 12 sessions of side-lying thrust SM in the lumbosacral region over 6 weeks. Inclusion criteria included 21-65 years old, Roland-Morris Disability Questionnaire (RMDQ) score ≥ 6 and numerical pain rating score ≥ 2. Spinal stiffness and FRR were assessed pre-treatment at baseline, after 2 weeks and after 6 weeks of treatment. Lumbar spine global stiffness (GS) were calculated from the force-displacement curves obtained using i) hand palpation, ii) a hand-held device, and iii) an automated indenter device. Lumbar FRR was assessed during trunk flexion-extension using surface electromyography. The primary outcomes were RMDQ and pain intensity measured by visual analog scale (VAS). Mixed-effects regression models were used to analyze the data. The mean age of the 82 participants was 45 years; 48% were female; and 84% reported LBP >1 year. The mean (standard deviation) baseline pain intensity and RMDQ were 46.1 (18.1) and 9.5 (4.3), respectively. The mean reduction (95% confidence interval) after 6 weeks in pain intensity and RMDQ were 20.1 mm (14.1 to 26.1) and 4.8 (3.7 to 5.8). There was a small change over time in the palpatory GS but not in the hand-held or automated GS, nor in FRR. The addition of each physiologic variable did not affect the model-estimated changes in VAS or RMDQ over time. There was no association seen between physiological variables and LBP intensity. Higher levels of hand-held GS at L3 and automated GS were significantly associated with higher levels of RMDQ (p = 0.02 and 0.03, respectively) and lower levels of flexion and extension FRR were significantly associated with higher levels of RMDQ (p = 0.02 and 0.008, respectively) across the 3 assessment time points. Improvement in pain and disability observed in study participants with chronic LBP was not associated with the measured GS or FRR. NCT01670292 on clinicaltrials.gov, August 2, 2012.