Preliminary investigation of the mechanisms underlying the effects of manipulation: exploration of a multivariate model including spinal stiffness, multifidus recruitment, and clinical findings

Abstract

Study design: Prospective case series.

Objective: To examine spinal stiffness in patients with low back pain (LBP) receiving spinal manipulative therapy (SMT), evaluate associations between stiffness characteristics and clinical outcome, and explore a multivariate model of SMT mechanisms as related to effects on stiffness, lumbar multifidus (LM) recruitment, and status on a clinical prediction rule (CPR) for SMT outcomes.

Summary of background data: Mechanisms underlying the clinical effects of SMT are poorly understood. Many explanations have been proposed, but few studies have related potential mechanisms to clinical outcomes or considered multiple mechanisms concurrently.

Methods: Patients with LBP were treated with two SMT sessions over 1 week. CPR status was assessed at baseline. Clinical outcome was based on the Oswestry disability index (ODI). Mechanized indentation measures of spinal stiffness and ultrasonic measures of LM recruitment were taken before and after each SMT, and after 1 week. Global and terminal stiffness were calculated. Multivariate regression was used to evaluate the relationship between stiffness variables and percentage ODI improvement. Zero-order correlations among stiffness variables, LM recruitment changes, CPR status, and clinical outcome were examined. A path analysis was used to evaluate a multivariate model of SMT effects.

Results: Forty-eight patients (54% women) had complete stiffness data. Significant immediate decreases in global and terminal stiffness occurred post-SMT regardless of outcome. ODI improvement was related to greater immediate decrease in global stiffness (P = 0.025), and less initial terminal stiffness (P = 0.01). Zero-order correlations and path analysis supported a multivariate model suggesting that clinical outcome of SMT is mediated by improvements in LM recruitment and immediate decrease in global stiffness. Initial terminal stiffness and CPR status may relate to outcome though their relationship with LM recruitment.

Conclusion: The underlying mechanisms explaining the benefits of SMT appear to be multifactorial. Both spinal stiffness characteristics and LM recruitment changes appear to play a role.

Figures

Figure 1

Time line for sessions and measurements in the study.

Figure 2

Mechanized indentation instrument used for spinal stiffness assessments.

Figure 3

Ultrasound image of the lumbar multifidus (LM) during a submaximal contraction using a contralateral arm raise. Contracted thickness was measured between the posterior-most portion of the L4/5 facet joint and the plane between the muscle and subcutaneous tissue.

Figure 4

Spinal manipulation technique. Each side of the subject's pelvis was manipulated at each session.

Figure 5

Path analysis and output from initial theoretical model. Direct standardized regression coefficients between variables are shown with each arrow.(* p <0.05, ** p <0.001) The R2 values represent the explained variance accounted for by the variables linked in the model. (χ2 / df = 1.6, p = 0.11, RMSEA = 0.12, CFI = 0.89, AGFI = 0.76)

Figure 6

Path analysis and output from the trimmed model. Direct standardized regression coefficients between variables are shown with each arrow.(* p <0.05, ** p <0.001) The R2 values represent the explained variance accounted for by the variables linked in the model. (χ2 / df = 0.86, p = 0.55, RMSEA = 0.0, CFI = 1.0, AGFI = 0.88)