Motor Control Stabilisation Exercise for Patients with Non-Specific Low Back Pain: A Prospective Meta-Analysis with Multilevel Meta-Regressions on Intervention Effects

Abstract

Low-to-moderate quality meta-analytic evidence shows that motor control stabilisation exercise (MCE) is an effective treatment of non-specific low back pain. A possible approach to overcome the weaknesses of traditional meta-analyses would be that of a prospective meta-analyses. The aim of the present analysis was to generate high-quality evidence to support the view that motor control stabilisation exercises (MCE) lead to a reduction in pain intensity and disability in non-specific low back pain patients when compared to a control group. In this prospective meta-analysis and sensitivity multilevel meta-regression within the MiSpEx-Network, 18 randomized controlled study arms were included. Participants with non-specific low back pain were allocated to an intervention (individualized MCE, 12 weeks) or a control group (no additive exercise intervention). From each study site/arm, outcomes at baseline, 3 weeks, 12 weeks, and 6 months were pooled. The outcomes were current pain (NRS or VAS, 11 points scale), characteristic pain intensity, and subjective disability. A random effects meta-analysis model for continuous outcomes to display standardized mean differences between intervention and control was performed, followed by sensitivity multilevel meta-regressions. Overall, 2391 patients were randomized; 1976 (3 weeks, short-term), 1740 (12 weeks, intermediate), and 1560 (6 months, sustainability) participants were included in the meta-analyses. In the short-term, intermediate and sustainability, moderate-to-high quality evidence indicated that MCE has a larger effect on current pain (SMD = -0.15, -0.15, -0.19), pain intensity (SMD = -0.19, -0.26, -0.26) and disability (SMD = -0.15, -0.27, -0.25) compared with no exercise intervention. Low-quality evidence suggested that those patients with comparably intermediate current pain and older patients may profit the most from MCE. Motor control stabilisation exercise is an effective treatment for non-specific low back pain. Sub-clinical intermediate pain and middle-aged patients may profit the most from this intervention.

Keywords: LBP; chronic low back pain; exercise; lumbago; lumbalgia; meta-analysis; motor control exercise; nonspecific; sensorimotor; stabilization; unspecific low back pain.

Figure 1

Participants’ flow.

Figure 2

Data and Forest plots for the pooled outcome estimates for the short-term effects of motor control stabilisation exercise vs. no additional exercise. (A): current pain intensity; (B): characteristic pain intensity; (C): disability; MCS, multicenter study; SCS; single-center study; A–F are the single study sites; a stands for MCE, b for MCE + behavioral. SD, standard deviation; P, p-value; IV, inverse variance; CI, confidence intervals; experimental, motor control stabilisation group.

Figure 3

Data and Forest plots for the pooled outcome estimates for the mid-term effects of motor control stabilisation exercise vs. no additional exercise. (A): current pain intensity; (B): characteristic pain intensity; (C): disability; MCS equals multicenter study; SCS equals single-center study, A–F are the single study sites; a stands for MCE, b for MCE + behavioral. SD, standard deviation; P, p-value; IV, inverse variance; CI, confidence intervals; experimental, motor control stabilisation group.

Figure 4

Data and Forest plots for the pooled outcome estimates for the long-term/sustainability effects of motor control stabilisation exercise vs. no additional exercise. (A): current pain intensity; (B): characteristic pain intensity; (C): disability. MCS equals multicenter study; SCS equals single-center study, A–F are the single study sites; a stands for MCE, b for MCE + behavioral. SD, standard deviation; P, p-value; IV, inverse variance; CI, confidence intervals; experimental, motor control stabilisation group.

Figure 5

Bubble plots of the meta-regressions (single predictor). (A) current pain intensity at baseline and (B) participants mean age at baseline on the effect size estimator of the mid-term effect (12 weeks) on disability. The size of the bubble illustrates the weighting of the respective study arm by inverse variance.