Brain plasticity following lumbar disc herniation treatment with spinal manipulation therapy based on resting-state functional magnetic resonance imaging

Abstract

As a prevalent spine disorder, Lumbar disc herniation (LDH) has been affecting more than 2 % of the worldwide population and is characterised by uncertain causes and recurring episodes. Studying the brain activity of patients could potentially provide insights into its pathogenesis and significantly enhance therapy. Therefore, we here examined brain function in patients under Spinal Manipulative Therapy (SMT). By analysing regional homogeneity (ReHo) at different frequency bands, we identified the discrepancies in brain activity between LDH patients and healthy people, highlighting the frequency dependence of spontaneous low-frequency oscillations among patients with LDH. Choosing seeds based on the peak ReHo differences helped to elucidate the functional connectivity alterations in the brain regions of LDH. Overall, this study showed that SMT significantly reduced pain, improved dysfunction, and partially rectified aberrant local consistency and functional connection in patients with LDH, not only offering insights into the pathophysiology of LDH from a neurological standpoint, but also providing inspiration for the development of new therapies based on neurobiology.

Keywords: Brain plasticity; Functional connectivity; Lumbar disc herniation; Regional homogeneity; Resting-state functional magnetic resonance imaging; Spinal manipulative therapy.

Fig. 1

Spinal Manipulation Therapy (SMT) performed by rolling (A), kneading (B), pushing (C), and the pulling and rotating (D) to relax the muscles in the lower back area for the purpose of alleviating pain and improving lumbar function.

Fig. 2

Significantly different regions in ReHo between LDH-pre and HCs using two-sample t-tests. A. Result in the conventional band. LDH-pre patients showed decreased ReHo values in the left orbital part middle frontal gyrus (LO-MFG, MNI coordinate: −30 48 -6) compared with HCs. B. Result in the slow-5 band. LDH-pre patients showed decreased ReHo values in the left orbital part middle frontal gyrus (LO-MFG, MNI coordinate: −33 51 -6) compared with HCs. C. Result in the slow-4 band. LDH-pre patients showed increased ReHo values in the left Cerebelum_4_5 (MNI coordinate: 9, −30, −36) compared with HCs. (GRF correction, voxel p < 0.05 and cluster p < 0.05). Color bar indicates the t score.

Fig. 3

Brain regions showing group differences between LDH_pre and HCs in static FC value. A.the ROI 1(Frontal_Mid_Orb_L, MNI: 6, 33, 3) as seed. B. the ROI 2 (Frontal_Mid_Orb_L,MNI: 30,39,-6) as seed. C. the ROI 3(Cerebelum_4_5_L, MNI: 9, −30, −36) as seed. ORBmid.L = Frontal_Mid_Orb_L; ORBinf.R = Frontal_Inf_Orb_R; ORBinf.L = Frontal_Inf_Orb_L. CRBL45.L = Cerebelum_4_5_L; PCUN.R = Precuneus_R (voxel p < 0.05, cluster p < 0.05, GRF correction).

Fig. 4

Bar graph shows comparisons (LDH_pre vs HCs, LDH_post vs HCs and LDH_post vs LDH_pre) of average ReHo in regions that showed significant differences. A. significant differences in Frontal_Mid_Orb_L in 0.01–0.08 HZ. B. significant differences in Frontal_Mid_Orb_L in 0.01–0.027 HZ. C. significant differences in Cerebelum_4_5_L in 0.027–0.073 HZ. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 5

Group differences of zFC values from brain regions in conventional frequency band. A1. The zsFC values of signal extracted from Frontal_Inf_Orb_R (6, 33, 3) of each participant. A2. The zsFC values of signal extracted from Frontal_Inf_Orb_L(-30, 39, −6) of each participant. B1. The zsFC values of signal extracted from Frontal_Inf_Orb_L (−27, −9, −9) of each participant. B2. The zsFC values of signal extracted from Frontal_Inf_Orb_R (6,33, 3) of each participant.C1. The zsFC values of signal extracted from Cerebelum_4_5_L (9, −30, −39) of each participant.C2. The zsFC values of signal extracted from Precuneus_R(18,-60,27) of each participant. The LDH patients and HCs were compared using two-sample t-tests. *P < 0.05, **P < 0.01, ***P < 0.001.