Electroacupuncture alleviates pain after total knee arthroplasty through regulating neuroplasticity: A resting‐state functional magnetic resonance imaging study

Abstract Introduction We aimed to evaluate the efficacy of electroacupuncture in relieving acute pain after total knee arthroplasty (TKA) and related mechanism. Methods In this randomized, single‐blind, and sham‐acupuncture controlled study. Forty patients with postoperative acute pain were recruited and randomly divided electroacupuncture group (n = 20) and sham‐acupuncture group (n = 20) from November 2020 to October 2021. All patients received electroacupuncture or sham‐acupuncture for 5 days after TKA. Their brain regions were scanned with resting‐state functional magnetic resonance imaging before and after intervention. Pain was scaled. Another 40 matched healthy controls underwent scanning once. The amplitude of low‐frequency fluctuation (ALFF) values was compared. Pearson's correlation analysis was utilized to explore the correlation of ALFF with clinical variables in patients after intervention. Results Compared with the HCs, patients with acute pain following TKA had significantly decreased ALFF value in right middle frontal gyrus, right supplementary motor area, bilateral precuneus, right calcarine fissure and surrounding cortex, and left triangular part of inferior frontal gyrus (false discovery rate corrected p < .05). Patients had higher ALFF value in bilateral precuneus, right cuneus, right angular gyrus, bilateral middle occipital gyrus, and left middle temporal gyrus after electroacupuncture (AlphaSim corrected p < .01). Correlation analysis revealed that the change (postoperative day 7 to postoperative day 3) of ALFF in bilateral precuneus were negatively correlated with the change of NRS scores (r = −0.706; p = .002; 95% CI = −0.890 to −0.323) in EA group. Conclusions The functional activities of related brain regions decreased in patients with acute pain after TKA. The enhancement of the functional activity of precuneus may be the neurobiological mechanism of electroacupuncture in treating pain following TKA.


Conclusions:
The functional activities of related brain regions decreased in patients with acute pain after TKA. The enhancement of the functional activity of precuneus may be the neurobiological mechanism of electroacupuncture in treating pain following TKA.

K E Y W O R D S
acute pain, electroacupuncture, resting-state functional magnetic resonance imaging, total knee arthroplasty

INTRODUCTION
Total knee arthroplasty (TKA) can effectively treat end-stage knee osteoarthritis (KOA) by reducing pain and helping patients to resume daily life activities. Acute postoperative pain after TKA prolongs rehabilitation duration, weakens the therapeutic effect, and thereby decreases the patient's quality of life. Intense postoperative pain is associated with chronic pain, and it may also prolong rehabilitation duration, weaken the therapeutic effect of TKA, and thereby decrease the patient's quality of life (Coppes et al., 2020;Hsia et al., 2018). The American Society of Anesthesiologists has recommended that multimodal pain management (administration of two or more drugs via the same route or different routes) should be implemented whenever possible to maximize the analgesia effect while minimize the potential adverse effects and reduce the consumption of opioids (American Management, 2012). But the incidence of moderate to severe pain following TKA is still as high as 58%, indicating that postoperative pain management is far from satisfactory (Summers et al., 2020). Conventional analgesia medicine could increase the risk of nausea, vomiting and other digestive system side effects, abnormal liver/ renal functions, and inhibit bone formation and healing (Zhao & Davis, 2019). Thus, effective and safe nonpharmacologic interventions for analgesia are necessary.
Acupuncture, a methodology of treating human diseases in China with a history of more than 3000 years, has been applied to various pain disorders, such as migraine (Tu et al., 2020), KOA (Kong et al., 2018), low back pain (Yu et al., 2020), fibromyalgia pain (Mawla et al., 2021), neck pain , and cancer pain Yang et al., 2021). Increasing evidence has proved the safety and effectiveness of acupuncture as an analgesic treatment and its advantage in decreasing the need for opioids (Michaelides & Zis, 2019). Neuroimaging studies have shown that acupuncture can achieve analgesic effect by restoring the pain processing, regulating pain perception, improving abnormal structure, and functional activities of patients (Tian et al., 2021;Wen et al., 2021). Electroacupuncture (EA), which combines the traditional acupuncture theory with electrical stimulation, has standardized intensity, frequency, duration and other parameters, making it more suitable for scientific studies than traditional manual acupuncture. EA has been found to relieve pain and reduce opioid consumption following TKA (Tedesco et al., 2017). The EA's analgesic effect on neuropathic pain may rely on the activation the brain functional connectivity between bilateral hemispheres and the sensorimotor cortex (Hou et al., 2020).
The descending pain modulation systems, including the anterior cingulated cortex, the periaqueductal gray, and the rostral ventromedial medulla, play an important role in the analgesic effect of EA . Although the central integration and plasticity play a critical role in the analgesic mechanism of acupuncture (Xiao et al., 2018), the underlying mechanism of EA in regulating the brain central system to relieve acute pain following TKA is largely unknown.
Resting-state functional magnetic resonance imaging (rs-fMRI), which gauges fluctuations in the blood oxygen level-dependent signal, is widely used to explore the neural mechanisms, assess the efficacy of acupuncture treatment, and it is also commonly used for pain research.
The amplitude of low-frequency fluctuation (ALFF), which depicts the intensity of regional spontaneous neuronal activities, strikes a good balance between test-retest reliability and replicability (Chen et al., 2018). ALFF reflects the blood oxygenation level-dependent (BOLD) signal fluctuations within the gray matter and the local properties of spontaneous neuronal activity. The enhancement of ALFF shows that the excitability of brain area is activated, and the BOLD signal deviated from the baseline. The weakening of ALFF indicates that neurons are inhibited and their activities are decreased. ALFF can be represented as the square root of the power spectrum in low-frequency range (0.01-0.08 Hz), which measures evaluate the brain's pathophysiological state by computing the regional intensity of spontaneous fluctuation in BOLD signal at rest (Zang et al., 2007). Studies no various diseases such as heroin addicts , trigeminal neuralgia (Ge et al., 2022), migraine , cervical discogenic pain , chronic low back pain (Zhang et al., 2019), fibromyalgia (Katherine et al., 2019), suggest that ALFF, as a reliable indicator of regional spontaneous neural activity in resting-state, can be widely used in pain disease studies.
In this study, we used rs-fMRI to explore the brain central mechanism of EA in treating acute pain following TKA. To our knowledge, few neuroimaging research has focused on this area. We speculated that (1) the ALFF patterns in patients with acute postoperative pain might be abnormal, and (2) EA could treat acute pain after TKA by regulating the functional activities of specific brain regions.

Standard protocol approvals, registration, and consents
The study took place at Shanghai Guanghua Hospital of Integrated Tra-

Study participants
Enrolled were KOA patients having undergone primary unilateral TKA.
The inclusion criteria included (1) aged 60-80 years; (2) right-handed; (3) a pain intensity score of 5 or higher on a 0-10 Numerical Rating any contraindications to rs-fMRI scanning (such as defibrillator, cardiac pacemaker, metal stents or electronic implant, intraocular mental foreign body, claustrophobia, and hyperpyrexia); (4) unwillingness to sign the consent form; (5) having received acupuncture/EA in the past three months.
The sex-and age-matched healthy controls (HCs) without any illness-induced pain sensation or psychological diseases were also recruited.

Experimental design and sample size
A participant-blinded, randomized and sham-acupuncture (SA) controlled clinical trial was applied. Previous studies have shown that 15 participants should be included in each group to ensure stable statistical effect for brain fMRI analysis (Qiu et al., 2016;Szucs & Ioannidis, 2020). In this study, rs-fMRI was conducted to explore the cerebral mechanism rather than observe the clinical efficacy. The participants were randomly assigned to either EA or SA group (n = 20 each) for rs-fMRI scanning.
The study lasted for 5 days: from postoperative day (POD) 3 to POD 7. Patients were instructed to completed postoperative pain diary by documenting the onset time, pain intensity (measure by NRS score), rescue medication use, and rs-fMRI scanning, which collected at POD 3 and POD 7.

Masking and intervention
The group allocation information was sealed an envelope and given to the acupuncturist. The surgeons, principal investigators, study staff, data analysts, and the participants were blinded to grouping. Using the sham electroacupuncture design in the previous literature (Liu et al., 2017), adhesive pads were applied to both groups, thus the blunttipped placebo needles with a similar appearance to conventional needles provided participant-blinding effects but no skin penetration.
Besides, the SA included a connecting cord with a broken inner wire with no actual current output. Participants in the EA group received acupuncture. Acupoint locations were selected based on the standardized acupuncture protocol for TKA (Zhong et al., 2019). In SA group, noninserted sham needles were applied to the same acupoints as in EA group. The electrodes were attached to these needles with, the same treatment setting as in the EA group. The EA device was turned on, but the electrodes were not inserted into an active port on the device, and no skin penetration or needle manipulation was achieved for "De qi." Intervention was not performed on HCs. All the patients included in the final analysis completed 5 treatment sessions during the 5-day treatment.

Clinical assessment
Nonsteroidal anti-inflammatory drug was used for analgesia, and supplementary dosage could be administered if needed. No additional analgesic medication was asked for in both EA and SA groups throughout the study. All patients underwent rs-fMRI scan twice.
The demographic information and clinical scale data (dependent variables) at POD 3 (preintervention) and POD 7 (postintervention) were analyzed. The primary outcome was mean reduction in pain intensity represented by the NRS score of patients. To obtain the NRS score, the patients were asked to circle a number ranging from 0 (no pain) to 10 (the most intense pain imaginable) that best fit their current level of pain. The reduction in Zung Self-Rating Depression Scale (SDS) score was calculated as the secondary outcome. SDS was used to evaluate the subjectively reported depression by the patients. The scale consisted of 20 items and each item was divided into 4 levels according the frequency of symptoms, which is suitable for adults with depression. For the SDS score, patients scored on the scale according to their emotional state.

Rs-fMRI data acquisition
The rs-fMRI data were acquired using a Clinical 1.

Baseline characteristics
Totally, 38 patients (20 in EA group, 18 in SA group) completed the two rs-fMRI scans (baseline and after 5-day treatment). Forty HCs completed one rs-fMRI scan. Due to excessive head movement (>3 mm) in scan, 7 patients and 8 HCs were excluded (Figure 1). No significant differences in age, gender, and body mass index were observed between patients and HCs. Table 1 shows that there was no difference between in the EA and SA group in pain intensity and SDS at baseline (age, gender, and body mass index).   Figure 3). In SA group, no significant differences in ALFF were observed between POD 3 and POD 7. Coordinates (X, Y, Z) refer to the peak MNI coordinates of brain regions with peak intensity. The resulting statistical map uses false discovery rates correction for multiple comparisons analysis (voxel-p < .05, cluster-p < .05, cluster size > 63). ALFF, amplitude of low-frequency fluctuation; MNI, Montreal Neurological Institute; POD3, postoperative day 3; HCs, healthy controls; L, left; R, right.

F I G U R E 2
Comparisons of ALFF between patients in postoperative day 3 and healthy controls. Blue represents the area with decreased ALFF value. ALFF, amplitude of low-frequency fluctuation; FMG, middle frontal gyrus; IFGtriang, triangular part of inferior frontal gyrus; SMA, supplementary motor area; PCUN, precuneus; CLA, calcarine fissure and surrounding cortex; L, left; R, right.

Correlation between ALFF and pain intensity after intervention
The EA-stimulation difference (EA_ POD 7 vs. POD 3) and group dif-

Complications and adverse events
All incisions were healed by the first intention. No skin ulcer, hematoma, infection, liver or kidney injury, or other adverse events was observed.

DISCUSSION
The transmission of pain in the central nervous system is highly complicated and involves multiple brain regions (Kim et al., 2021;Qi et al., 2018). In this study, we investigated the underlying neurobiology of postoperative acute pain and the modulatory effect of EA on acute pain after TKA. Coordinates (X, Y, Z) refer to the peak MNI coordinates of brain regions with peak intensity. The resulting statistical map uses AlphaSim correction for multiple comparisons analysis (voxel-p < .01, cluster-p < .01, cluster size > 107). ALFF, amplitude of low-frequency fluctuation; MNI, Montreal Neurological Institute; POD3, postoperative day 3; POD7, postoperative day 7; L, left; R, right.
MFG and IFGtriang in the prefrontal cortex can modulate pain perception (Morton et al., 2016). The MFG activity is implicated in pain stimulation and modulation, perception of negative emotions, and cognitive evaluation (Boissoneault et al., 2020). Deactivating the left IFGtriang is associated with response inhibition, as well cognitive and emotional dysregulation . Precuneus, a cortical element in the medial surface, is involved in attention, spatial integration, self-awareness, psychological activities, social cognition, and emotional processing (Pereira-Pedro & Bruner, 2016). CAL, surrounded by the primary visual cortex, is related to cognitive function . Although all participants in this study showed normal cognitive function, the decreased functional activities in the above brain regions may lead to the experience of intense pain in patients, further reducing their brain activities and weakening cognitive function (Huang et al., 2015). We speculate that patients with pain sensitivity after TKA may have a higher risk of pain-induced cognitive impairment, which is associated with the changed functional brain activities. The SMA in the frontal cortex can delimit the motor from the prefrontal cortical regions and link cognition to action in normal behaviors, involving movement preparation, short-term retention of pain dynamics, and other cognitive and motor-related processes (Khoshnejad et al., 2017).
The decreased functional activities of SMA may be related to the knee motor dysfunction. In KOA patients, the pathological changes influenced the spatial patterns of intrinsic brain activity, leading to altered brain function and structure (Kang et al., 2022). The deactivation of these regions involved in pain regulation may indicate inadequate inhibition or increased facilitation of sensory information, which is associated with functional impairment in pain processing, thus contributing to acute pain after TKA.
This preliminary clinical trial has shown that EA reduced pain intensity and depression tendency after TKA. The analgesia effect of EA may be achieved through regulating the plasticity of brain functional activities, while SA cannot, exert such an effect. And this explains why the patients in EA group had higher ALFF in right cuneus, right angular gyrus, bilateral MOG, right MTG, and bilateral precuneus at POD 7 than at POD 3.
The cuneus integrates somatosensory input with other sensory stimuli and cognitive processes, involving the emotion dimension of pain (Price, 2000). The angular gyrus is implicated in cognitive functions, including attention and spatial cognition, default mode network, and social cognition (Seghier, 2013). Factors associated with cognition are likely to modulate pain perception. The surgical injury triggers a myriad of responses in the pain matrix, from sensitization of central pain pathways to anxiety and depression (Small & Laycock, 2020). Emotion affects pain perception, and patients experiencing severe acute pain are more likely to develop a negative emotion, which in turn increases pain perception, thus forming a vicious circle (Michaelides & Zis, 2019). Decreased ALFF of left MOG is observed in depressive patients (Teng et al., 2018). Temporal cortex participates in pain perception and modulation, as confirmed in human neuropathic states, and it is also closely related to emotional control and sensory process modulation, and pain sensation duration (Khoshnejad et al., 2017). The SDS score in EA group showed the patients receiving EA had a lower tendency to develop depression, which can be explained by the reduced pain severity and negative emotions after EA, and the resultant increase of the functional activities in the relevant brain regions. Another explanation is EA stimulation directly activates the relevant brain areas to regulate the adverse emotions and reduce the perception of acute pain. Pain relief could be evidenced by the reduction of NRS score in the EA group, compared with the SA group, although both received a conventional analgesic agent during the whole study. Changes of these brain functional activities indicate that EA relieves the pain by increasing the activities of the brain regions associated with the pain modulation, and thus changing the pain perception from nociceptive receptors to the pain matrix.
The EA-stimulation (EA_POD 7 vs. POD 3) and group (HC vs. POD 3) difference as shown by ALFF overlapped in precuneus. The nucleus cuneiformist is an inhibitory region in the descending pathway, and the hypoactivity in this region indicates reduced inhibition of the pain response in patients, which produces a higher pain score (Schwedt et al., 2015). Precuneus, a key region in the neuronal network for continuous information gathering and assessment of self-relevant sensations, mediates intrinsic activities throughout the default model network (Goffaux et al., 2014). The ALFF in precuneus decreased when acute pain is evoked in chronic low back pain patients (Zhang et al., 2019). When receiving pain stimuli, precuneus presents deactivation, and pain sensitivity is negatively correlated with the activity in the precuneus (Goffaux et al., 2014). The correlation analysis between the reduction of NRS score and the ALFF change of bilateral precuneus in the present study, we found that the activated changes in precuneus may play a crucial role in modulating postoperative pain perception.
And we speculate that EA increases the functional activity of precuneus to reduce the postoperative pain sensitivity and perception, which may be a central mechanism of EA in alleviating acute pain after TKA.
The pain intensity of patients in the SA group was also gradually eased over time after TKA. This may be attributed to the fact, the patients tended to limit their limb movement on the operative side to reduce the pain sensation due to the poor analgesic effect of SA. No changes (POD 7 vs. POD 3) in brain regions related to pain regulation were found on rs-fMRI in SA group. The different changes of brain regions in the two groups indicated the central mechanisms of EA and SA were different. Part of the therapeutic effect of SA may be produced by the physiological effects on the skin during the stimulation.
It should be noted that the AlphaSim correction was applied to explore the mechanism of EA analgesia in our study. Although the precuneus functional activities in EA group were closer to those HCs, the finding did not survive the more stringent FDR correction methods, and the AlphaSim correction method is relatively loose; therefore, further studies are needed. it should be emphasized that the exclusion of the images exerts no impact on treatment response. (4) Due to some practical difficulties and the limitations of experimental conditions, we failed to enroll an untreated group to investigate the effect of elapsed time on pain management after TKA. Hopefully, we will overcome the disadvantages in the subsequent work.

CONCLUSION
In conclusion, the functional activities in the right MFG, right SMA, bilateral precuneus, right CAL, and left IFGtriang in patients with acute pain after TKA decreased. EA on four acupoints (Futu, Zusanli, Yinglingquan, and Yanglingquan) can increase the functional activities of right cuneus, right angular gyrus, MOG, left MTG, and precuneus.
The functional activity of the precuneus is a biomarker of pain after TKA. Enhancement of functional activity of precuneus may be the neurobiological mechanism of EA in treating acute pain following TKA.

AUTHOR CONTRIBUTIONS
BXK, CZ, and JM conceived the study; BXK drafted the study; HQW, XLG, HX, SZ, CXG, and XRX recruited the participants. XYA and JX collected clinical data. MMD and CZ were responsible for statistical analyses and tables. LBX and JS have primary responsibility for the final content. All authors contributed to writing and revising the paper and agreed to submission.

ACKNOWLEDGMENTS
We are grateful to the participants who volunteered to participate in this research.

CONFLICT OF INTEREST STATEMENT
The authors declare that they have no conflict of interest.

DATA AVAILABILITY STATEMENT
The data sets used and analyzed during the current study are available from the corresponding author on reasonable request.

ETHICS STATEMENT
This study was approved by the Ethics Committee of Shanghai

Guanghua Hospital of Integrated Traditional Chinese and Western
Medicine and was in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Written informed consent to participate was obtained from all of the individual participants included in the study.

TRIAL REGISTRATION
The trial was registered in Chinese Clinical Trial Registry (ChiCTR 2000033778).