Waiguan Stimulation May Kindle Anticorrelated Brain Networks: Functional Magnetic Resonance Imaging Data Revisited

Abstract

Subtraction of functional magnetic resonance imaging activity data results in a loss of information regarding possible general patterns of brain activation under experimental conditions. We, hence, reanalyzed previous Waiguan acupuncture data to qualitatively elucidate patterns of cerebral correlations to acupuncture and placebo conditions. Healthy individuals (n = 24) were randomly allocated to true and sham Waiguan acupuncture and to true and sham needling of a nonacupuncture point (nonacupoint), and functional magnetic resonance imaging scans were performed during stimulation. Statistical parametric mapping group comparisons revealed clearly different patterns of activation between Waiguan stimulation and the corresponding stimulation of a nonacupoint. The former condition produced less neocortical activation than the nonacupoint stimulation. Cerebellar activation was typically seen only during true Waiguan acupuncture. The reduced neocortical activity during both true and sham Waiguan acupuncture may indicate that this point activates anticorrelated networks, with possible intrinsic healing properties. Cerebellar activation during true Waiguan acupuncture implies the region’s influence on healing networks.

1. Introduction

Functional magnetic resonance imaging discloses patterns of increased or decreased brain activity, which can uncover pathological conditions or be used to compare one state of activity in the brain with another. In most experimental studies, a resting state of the brain is compared with one ensuing from a specific activity or stimulation by subtracting the activity data for the two conditions. A fundamental problem with analyzing subtraction images is the loss of intrinsic qualities of the two situations, i.e., the likelihoods that not only the specific condition, but also a diversity of basic conditions change during the experimental state.

In a previous study of patients with a specific phobia, we reasoned that a default-mode network typical of the resting brain was replaced by an innate state of fight or flight behavior [1–3]. We likewise suggested that a state of secondary pain, which was also influenced by the default-mode network, was replaced by a state of primary pain in fibromyalgia patients exposed to painful stimulation [4]. Brain activities during the states of both fight and flight and the primary pain reaction were characterized by clearly reduced neocortical activity in favor of the primary and secondary sensor areas.

Assuming that in our previous study, information on the general activation patterns of the brain was lost when a state of resting wakefulness was subtracted from a state of Waiguan acupuncture (SJ 5), we reanalyze those data here [5]. The previous study showed that true SJ 5 acupuncture activated Brodmann area 2, the left cerebellum, and the right inferior semilunar lobule [5].

2. Materials and methods

The experimental conditions are described in detail in our previous study (trial registration number: ChiCTRNRC00000255) [5], so they are only briefly outlined here. The present study included 12 male and 12 female right-handed volunteers with a mean age of 23.1 ±2.2 years (range, 21–28 years), who were randomly divided into true and sham SJ 5 acupuncture, and true and sham needling of a nonacupoint groups. All participants signed a written informed consent [5].

The SJ 5 acupoint is located on the forearm, 2 cun above the transverse crease of the dorsum of the wrist, between the radius and the ulna. The location of the nonacupoint was on the same level as SJ 5, but to the right of the midline between the Sanjiao Meridian of Hand-Shaoyang and the Small Intestine Meridian of Hand-Taiyang. The procedure of needling was described in our previous study, as was the device for brain imaging, a 3.0-T whole-body magnetic resonance imaging scanner with a standard head coil [5].

The Statistical Parametric Mapping (SPM2, http://www.fil.ion.ac.uk) software with the Matlab 6.1 platform was used for functional magnetic resonance imaging data processing. Slight movements of the participant’s head between scans were realigned to the first image of the time series. Images were then normalized to the Montreal Neurological Institute space, spatially smoothed with a 5 × 5 × 5 Gaussian kernel, and analyzed voxel by voxel according to a linear model.

The time course of the functional images was analyzed on the magnetic resonance imaging work station. An ideal wave form based on the time course of each voxel’s signal intensity was set before the voxel’s signal—activity curve was compared with that of the reference. Voxels with an uncorrected p < 0.0001 and a minimum cluster size of 10 voxels (K > 10) were regarded as being related to the experimental task and were used to produce images of significant changes in the functional activity [6].

3. Results

Waiguan acupuncture and sham needling of SJ 5 activated a total of 10 brain regions, whereas the matching non-acupoint stimulation activated a total of 35 brain regions, which were all of neocortical origin (Table 1 and Fig. 1). Both true and sham needling of SJ 5 activated the right inferior parietal lobule and the right insula. SJ 5 acupuncture additionally activated the left cerebellum. In contrast to the subtle activation pattern caused by SJ 5 stimulation, both true and sham needling of the nonacupoint produced a robust general activation of neocortical associative fields (Table 2 and Fig. 1).

Table 1 Brain areas activated by true and sham SJ5 acupuncture.

Figure 1

Cerebral activation patterns during true and sham SJ 5 acupuncture, and during true and sham nonacupoint needling.

Table 2 Brain areas activated by true and sham needling of the nonacupoint.

4. Discussion

True Waiguan acupuncture resulted in a discrete pattern of activation that was clearly different from the pattern of activation resulting from stimulation of a nearby sham acupoint. As in our previous study, we also see a typical activation of the left cerebellum, which agrees with the acupoint’s relevance in acupuncture and the initiation of anticorrelated networks [5,7].

According to the theory of traditional Chinese medicine, stimulation of a nonacupoint should not produce healing effects. We here see that sham SJ 5 stimulation, which can be compared with a touching the skin, resulted in the activation of four brain regions as compared to 20 for stimulation of the nonacupoint (Fig. 1). Extrapolating from previous studies of the fight or flight behavior and the primary pain reaction, this may result from a capacity of the acupoint to downgrade cognitive processing in favor of innate brain networks with a healing capacity [1,4].

A possible confounding factor in our study may be the deqi response that is routinely elicited during acupuncture. As a senior acupuncturist performed the needling, we assume that deqi was obtained during true SJ acupuncture. However, we cannot exclude the possibility that a different response, including, for example, pain, may have been elicited during nonacupoint needling. In conclusion, our data indicate that Waiguan acupuncture may trigger anticorrelated networks with possible importance in healing processes.