Objective Acupuncture is beneficial in treating stroke neuropsychiatric symptoms. The present study aimed to identify functional brain response to active acupuncture in patients with unilateral ischaemic stroke using functional MRI (fMRI).
Methods A total of 10 patients aged 47–65 years with left hemispheric ischaemic stroke received single-session manual acupuncture at the TE5 point of the affected (right) forearm. A 6-min tactile control procedure in which an acupuncture needle tip was alternately touched and removed from the skin at the acupuncture point for 30 s each was performed first, followed by active acupuncture in a blocking paradigm consisting of six 30-s twist blocks of rotation interspersed between six 30-s blocks of rest. A whole brain scan was simultaneously conducted on a 3.0-T imager. Activated and deactivated brain regions during tactile stimulation and active acupuncture relative to rest were obtained via group analysis.
Results Compared to tactile stimulation, needling with twist manipulation modulated many more widespread brain areas. All the brain areas activated and deactivated by active acupuncture relative to tactile stimulation were distributed in the primary sensorimotor and medial frontal cortex of the unaffected, but not the affected hemisphere.
Conclusions Active acupuncture results in lateralisation of functional cerebral response to the contralateral unaffected hemisphere in patients with unilateral stroke. This lateralisation may represent an effect of acupuncture in enhancing a compensatory process by redistributing functions into the intact cortex, particularly in the unaffected hemisphere.
Statistics from Altmetric.com
Acupuncture treatment has become an important option in the management of stroke neuropsychiatric sequelae.1 ,2 This is because numerous studies have demonstrated the therapeutic benefits of acupuncture in reducing stroke motor and neurological symptoms,3–6 urinary incontinence,7 insomnia8 and depression.9 Acupuncture is also capable of normalising neuroimaging, neurochemical and behavioural abnormalities in patients who have experienced stroke and animal models.5 ,10 ,11 However, it is well documented that stroke recovery and rehabilitation are associated with a compensatory process by redistributing functions of the damaged regions to surviving brain regions, particularly in the unaffected hemisphere.12–14 This has led to the assumption that the therapeutic benefits of acupuncture for post-stroke complications may be related to its enhancement of the compensatory process after stroke.
Over the past two decades, neuroimaging approaches, mainly functional MRI (fMRI) and positron emission topography (PET), have been widely introduced into acupuncture research.15 A salient finding from neuroimaging studies is that acupuncture at traditional acupuncture points produces more widespread and intense cerebral responses than control and non-traditional acupuncture points in healthy volunteers.15 However, little information is available for functional neuroimaging brain responses to acupuncture stimulation in pathological conditions.
Recently, we have determined the brain response to active acupuncture at TE5 (Wai-Guan) in healthy volunteers using a PET approach.16 ,17 TE5 is an important traditional acupuncture point often used for the treatment of stroke-related motor, neurological and autonomic nerve problems in clinical practice.18 The purpose of the present study was to further examine the functional brain regional response to active acupuncture at the TE5 point in patients with unilateral ischaemic stroke.
This study was conducted in the First Affiliated Hospital of Guangzhou University of Chinese Medicine and Nanfang Hospital of Southern Medical University at Guangzhou between October 2008 and August 2010. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Guangzhou University of Chinese Medicine and registered at http://www.chictr.org (registration no.: ChiCTR-ONRC-08000255). All participants gave voluntary, written and informed consent before entering the study.
A total of 10 patients aged 47–65 years who met the following criteria participated in the study: (1) had a neuroimaging-confirmed ischaemic stroke in the left hemisphere that occurred at least 1 month ago but no more than 12 months ago, with significant right hemiplegia but no severe dysphasia, as shown by a score of ≥4 (moderate motor inability) on either upper or lower limb or both, but a score of ≤3 (patients could communicate simply) on speech ability on the Chinese Stroke Scale19; (2) the patient's current condition was stable under conventional medical management; (3) the patient was right-handed; and (4) the patient had never had acupuncture treatment or had not had acupuncture treatment for at least 4 weeks. Patients had no comorbid neuropsychiatric illnesses, substance addiction, coagulation disorders or other serious medical conditions. The basic characteristics of the patients are summarised in table 1.
Patients were assigned for manual acupuncture at the TE5 point of the right affected forearm. The point is located on the dorsal aspect of the forearm at midpoint of the interosseous space between the radius and the ulna, 2 cun proximal to the dorsal wrist crease. Manual manipulation, consisting of a tactile control procedure and active acupuncture, was performed by an experienced acupuncturist (J-QC) with a sterile silver needle (0.30 mm in diameter and 40 mm in length). Patients first received a 6-min tactile control procedure in which a needle tip was alternately touched and removed from the skin at the acupuncture point for 30 s each, followed by 6-min active acupuncture in a blocking paradigm consisting of six 30-s twist blocks of rotation interspersed between six 30-s blocks of rest, with a 5-min interval of rest between the control procedure and active acupuncture (figure 1). The 5-min interval was considered sufficient to restore sensitivity of cutaneous sensory organs as sensory adaptation only lasts for few seconds. Tactile stimulation has been widely used as a non-invasive control for acupuncture neuroimaging studies.20 For active acupuncture, the needle was inserted perpendicularly into the skin for a depth of about 15 mm and then twisted evenly at a frequency of 60 circles per min. Needling sensation (de qi) was achieved generally within 15–20 s of rotational manipulation and recognised by the subjects’ feelings, such as soreness, numbness, heaviness and distension, or by the acupuncturist's fingers, which felt increased resistance to further movement of the inserted needle.15 During experimental stimulation, subjects were required to keep quiet and remain calm, without speaking.
The fMRI brain scan was conducted on a 3.0-T whole-body scanner (GE Signa) with a standard head coil. Patients were first prevented from experiencing auditory and visual activities via earplugs and eyeshades, respectively. High-resolution 3D structural images for anatomical localisation were collected for each subject by using a T1-weighted gradient echo sequence at a voxel size of 1 mm3, parallel to the anterior commissure–posterior commissure (AC–PC) line and covering the whole brain. The blood oxygenation level-dependent functional imaging signals were acquired using a T2*-weighted single-shot, gradient-recalled echo-planar imaging (GR-EPI) sequence (TR=3.0 s, TE=20 ms, field of view (FOV)=256×256 mm2, matrix size=24×24, slice thickness=6.0 mm, interslice gap=1.0 mm, flip angle=90°).
Imaging data were processed using statistical parametric mapping software (http://www.fil.ion.ucl.ac.uk/spm/). Spatial and temporal preprocessing was conducted to correct bias caused by head motion by adjusting within-scan acquisition time differences between slices and then realigning to the first scan. Those whose gross motion exceeded 2 mm in any axis were removed from the analysis (in fact, no subjects had head motion exceeding 1 mm on any axis). The images were then normalised based on the standard brain template of the Montreal Neurological Institute (MNI), and spatially smoothed with a 5-mm full-width-at-half-maximum Gaussian kernel. For each individual subject, the brain areas activated or deactivated during tactile control procedure and active acupuncture relative to rest condition (ie, no stimulation) were obtained using fixed effects analysis with the general linear model on a voxel-by-voxel basis. Group analysis was performed to further identify the cerebral areas exhibiting significant differences in fMRI signal intensity induced by active acupuncture in comparison with tactile control procedure using random effects analysis with a two-sample t test. Statistical significance was thresholded at an uncorrected level of p<0.001 and a minimum cluster size of 30 voxels. The resulting coordinates were transformed into Talairach space with Brodmann area (BA) labels.
Relative to rest condition, tactile stimulation with needle tip on the skin significantly deactivated the bilateral parietal precuneus (BA7); but no brain areas were activated (data not shown). Needling with rotational manipulation produced significant activations of the ipsilateral precuneus (BA7, BA19), supramarginal gyrus (BA40), inferior parietal lobule (BA40), middle temporal gyrus (BA39), occipital cuneus (BA17) and posterior cingulate (BA23), and the contralateral inferior frontal gyrus (BA10), occipital cuneus (BA18) and middle temporal gyrus (BA21). Needling with rotational manipulation also significantly deactivated the ipsilateral postcentral gyrus (BA3, BA5), precentral gyrus (BA4) and medial frontal gyrus (BA10), and the contralateral postcentral gyrus (BA3), medial frontal gyrus (BA6, BA10) and middle temporal gyrus (BA21) (table 2, figure 2).
Relative to the tactile control procedure, all areas activated and deactivated by active acupuncture were distributed in the unaffected (right) hemisphere. The activated areas were the postcentral gyrus (BA2, BA3), precentral gyrus (BA4, BA6) and medial frontal gyrus (BA6); the deactivated areas were the postcentral (BA3) and precentral gyrus (BA6) (table 3, figure 3).
The principal finding of the present study is that, relative to the tactile control procedure, needling with twist manipulation at the right TE5 point significantly activated and deactivated fMRI-visualised activity of the unaffected, but not affected, primary sensorimotor and medial frontal cortex in patients with unilateral stroke. This finding indicates that active acupuncture results in lateralisation of cerebral functional response to the unaffected hemisphere. Similar lateralised cerebral responses were also observed in patients with chronic hemiparetic motor stroke, who received active acupuncture at body and scalp acupuncture points of bilateral sides.21 The lateralised response pattern of patients who have experienced stroke is apparently different from that of healthy subjects examined via a PET approach, demonstrating that the brain areas modulated by active acupuncture at the right TE5 point were distributed in the contralateral cerebrum and cerebellum as expected neuroanatomically.16 The difference in fMRI-visualised cerebral response was also present between healthy and stroke subjects with electrical stimulation at LI4 (He-Gu) and LI11 (Qu-Chi) on the affected side of the body.22 It is well documented that stroke recovery and rehabilitation are associated with a compensatory mechanism, in which the spared peri-infarct areas, distant sites and even regions in the unaffected hemisphere take over functions from regions lost to stroke.12–14 Therefore, the active acupuncture-induced lateralisation of cerebral response observed in the present study most likely represents an enhancement effect of acupuncture in redistributing functions of the lesion regions to the intact cortex, particularly in the unaffected hemisphere.
The present study and our previous PET study in healthy subjects16 consistently showed that many more cerebral areas were activated and deactivated during rotational manipulation at TE5 compared to tactile control stimulation. The greater response was also observed in deep versus shallow and rotating versus non-rotating manipulation in healthy subjects.15 The more widespread cerebral response may represent a consequence of more efficient acupuncture stimulation at acupuncture points, beneath which more sensory organs and afferent nerve fibres are excited by certain manual techniques (eg, rotation), resulting in the release of more nerve mediators, in particular from non-neuronal cells (eg, mast and lymph cells) via biochemical and biophysical reactions in the neural acupuncture unit, which is defined as a collection of acupuncture-activated neural and neuroactive components distributed in the skin, muscle and connective tissues surrounding the inserted needle.15
The present study revealed that most brain areas activated and deactivated during rotational manipulation at TE5 were distributed in the primary sensory and motor cortex of the ipsilateral unaffected hemisphere. Previous studies have also found that electrical stimulation at the LI4 and LI11 points of the affected side of the body resulted in greater activation in the somatosensory cortex of patients who have experienced stroke than healthy controls.22 Moreover, clinical improvement in motor function of the affected upper limbs was significantly correlated with changes in activation of the ipsilesional motor cortex in patients who have experienced stroke treated with active acupuncture at body and scalp acupuncture points of bilateral sides.21 Considering the fact that acupuncture is particularly beneficial in alleviating stroke-induced motor impairment, including limb paralysis, swallowing dysfunction and motor aphasia,3 ,6 the acupuncture treatment benefit on stroke motor symptoms should be, at least in part, attributed to the specific modulation of the primary sensory and motor cortex function.
The current study also found that active acupuncture activated the medial frontal cortex of the ipsilateral unaffected hemisphere. The medial frontal cortex mainly consists of the medial prefrontal cortex and anterior cingulate cortex.23 It is well known that the latter two brain regions are extensively involved in the processing of acupuncture signals associated with emotion, pain and autonomic nervous functions.15 ,24 Numerous clinical studies also have suggested that acupuncture treatment is remarkably helpful in alleviating post-stroke depression and visceral symptoms.7 ,9 It seems that the alleviation is likely associated with acupuncture modulation of the medial frontal cortex functions as observed in the present study. Unexpectedly, the present study showed that the cerebellum was unaffected during active acupuncture in patients who have experienced stroke, while our PET study showed several cerebellar areas activated during active acupuncture at the TE5 point in healthy subjects,16 suggesting that the cerebellum may play a minor role in the therapeutic response to active acupuncture in the stroke condition.
Several limitations of the present study should be noticed. First, the present study only included 10 subjects with varied severity and duration of the illness and gender domination. The inclusion of more subjects should heighten the statistical power and ensure the accuracy of the conclusion drawn. In addition, gender differences have been found in pathological characteristics of stroke and treatment outcomes.25 ,26 Whether there is a gender difference in functional neuroimaging brain response to acupuncture needs to be further investigated. Second, the stroke clinical symptoms were only evaluated at baseline, but not post treatment. The study only conducted a single session of acupuncture treatment and brain imaging scans. This made it difficult to correlate changes in neuroimaging variables and treatment effects in improving stroke symptoms. Despite this, the present study provides a piece of evidence favouring the assumption that acupuncture could enhance post-stroke compensatory process for the lesion sides and regions of the brain, and is consistent with the findings of previous studies showing that acupuncture stimulation selectively activates the brain on the lesion side in patients who have experienced stroke with aphasia and dysphagia.5 ,6 Finally, the present study used non-invasive tactile stimulation at the same acupuncture point as a control procedure, rather than needling at a non-point, which is generally defined as being at a certain distance (usually 1–3 cm) from the point examined.15 This is because previous studies have demonstrated similarity in neuroimaging response patterns between meridian-based points and adjacent control points.15
In summary, active acupuncture at the TE5 point of the affected forearm results in lateralisation of fMRI-visualised cerebral response in patients with unilateral stroke, with activation and deactivation of the ipsilateral unaffected primary sensorimotor cortex and medial frontal cortex. The lateralised cerebral response of patients who have experienced stroke is apparently different from that of healthy subjects and may represent an effect of acupuncture in enhancing a compensatory process by redistributing functions to the intact cortex. The present study provides useful information for developing more efficient acupuncture modalities for stroke treatment.
Patients with stroke in the left hemisphere were given acupuncture on the (affected) right arm.
MRI showed an acupuncture response only in the right, unaffected hemisphere, compared with control.
This response may be associated with a role of acupuncture in enhancing compensatory processes.
This study was supported by National 973 Program of China (no.: 2006CB504505, X-SL), National 973 Program of China (no.: 2012CB518504, X-SL), Third Key Subject of the ‘211 Project’ of Guangdong Province (YH) and General Research Fund (GRF) of Hong Kong Research Grant Council (RGC) (786611, Z-JZ).
YH and J-QC contributed equally to this article.
Contributors YH, X-SL and Z-JZ were involved in conception and design of the study. Z-JZ, J-QC and YH conducted final data analyses and drafted the manuscript. J-QC, C-ZT, J-JY, HC, J-XW, H-LX S-SQ and Y-DY performed experimental and acupuncture manipulation.
Funding National 973 Program of China (no.: 2006CB504505, X-SL), National 973 Program of China (no.: 2012CB518504, X-SL), Third Key Subject of the ‘211 Project’ of Guangdong Province (YH), General Research Fund (GRF) of Hong Kong Research Grant Council (RGC) (no.: 786611, Z-JZ) and HKU intramural funds (No.: 10400876, Z-JZ).
Competing interests None.
Patient consent Obtained.
Ethics approval The Ethics Committee of the First Affiliated Hospital of Guangzhou University of Chinese Medicine.
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.