Interaction of acupuncture treatment and manipulation laterality modulated by the default mode network

Abstract

Appropriate selection of ipsilateral or contralateral electroacupuncture (corresponding to the pain site) plays an important role in reaching its better curative effect; however, the involving brain mechanism still remains unclear. Compared with the heat pain model generally established in previous study, capsaicin pain model induces reversible cutaneous allodynia and is proved to be better simulating aspects of clinical nociceptive and neuropathic pain. In the current study, 24 subjects were randomly divided into two groups with a 2 × 2 factorial design: laterality (ipsi- or contralateral side, inter-subject) × treatment with counter-balanced at an interval of one week (verum and placebo electroacupuncture, within-subject). We observed subjective pain intensity and brain activations changes induced by capsaicin allodynia pain stimuli before and after electroacupuncture treatment at acupoint LI4 for 30 min. Analysis of variance results indicated that ipsilateral electroacupuncture treatment produced significant pain relief and wide brain signal suppressions in pain-related brain areas compared with contralateral electroacupuncture. We also found that verum electroacupuncture at either ipsi- or contralateral side to the pain site exhibited comparable significant magnitudes of analgesic effect. By contrast, placebo electroacupuncture elicited significant pain reductions only on the ipsilateral rather than contralateral side. It was inferred that placebo analgesia maybe attenuated on the region of the body (opposite to pain site) where attention was less focused, suggesting that analgesic effect of placebo electroacupuncture mainly rely on the motivation of its spatial-specific placebo responses via attention mechanism. This inference can be further supported by the evidence that the significant interaction effect of manipulation laterality and treatment was exclusively located within the default mode network, including the bilateral superior parietal lobule, inferior parietal lobule, precuneus, and left posterior cingulate cortex. It is also proved that disruptions of the default mode network may account for the cognitive and behavioral impairments in chronic pain patients. Our findings further suggested that default mode network participates in the modulation of spatial-oriented attention on placebo analgesia as a mechanism underlying the degree to which treatment side corresponding to the pain.

Keywords: Interaction effect; capsaicin pain model; default mode network; electroacupuncture; laterality; placebo analgesia.

Figure 1.

Experimental design. A total of 32 healthy subjects were recruited in this experiment. Three subjects did not meet the inclusion criteria of pre-selected examination and one was allergic to capsaicin pain. One in ipsilateral electroaupuncture group could not tolerate electroacupuncture treatment. Three subjects were also excluded due to poor quality (head movement exceeded 2.5 mm), one of them was in ipsilateral group, and the others were in contralateral group. Finally, 24 subjects completed the randomized, controlled, crossover study and were used for data analysis, 12 subjects for ipsilateral electroacupuncture group, and 12 subjects for contralateral electroacupuncture group. VA = verum electroacupuncture, PA = placebo electroacupuncture.

Figure 2.

Schematic illustration of the fMRI experimental sessions. Capsaicin pain was established at the medial aspect of left forearm by using capsaicin cream within region A. After 30 min, the cream was removed. Then, an eight-block paradigm of mechanical stimulation testing with MRI scanning was performed in the secondary hyperalgesia shown as region B by 20.9 g von Frey hair to provoke allodynia. Each stimulation lasted 30 s interrupted by a baseline of 30 s.

Figure 3.

Subjective pain rating changes of pre- and post-acupuncture treatment. Each bar represents the mean of pain ratings scores, and error bars denote standard devation. Significant group differences between pre- and post- acupuncture treatment shown by *P < 0.05, **P < 0.01, and ***P < 0.005.

Figure 4.

Subjective pain ratings changes (pre- minus post-treatment, mean ± SD) between the ipsilateral and contralateral groups for verum and placebo acupuncture. Each bar represents the mean of pain ratings difference, and error bars denote standard devation. *P < 0.05.

Figure 5.

Brain activations in response to all pre-treatment capsaicin allodynia pain stimulation. Significant activations occurred in the bilateral insular, thalamus, ACC/MPFC, SI, SII, and DLPFC (voxel-wise P < 0.005 corrected with 10 contiguous voxels). ACC/MPFC = medial prefrontal cortex, SI = primary somatosensory cortex, SII = secondary somatosensory cortex, DLPFC = dorsolateral prefrontal cortex.

Figure 6.

Interaction effect of laterality manipulation and acupuncture modality by ANOVA analysis. Significant brain signal changes were mainly located within the DMN, inclucding the right inferior parietal lobule (a), left inferior parietal lobule (b), right superior parietal lobule (c), right precuneus (d), left precuneus (e), respectively (voxel-wise P < 0.005 corrected with 15contiguous voxels). ANOVA = analysis of variance, DMN = default mode network.