Effect of acupoint with moxa cone on 1H-MRS of hippocampus and posterior cingulate gyrus in patients with mild cognitive impairment

Background: Mild cognitive impairment (MCI) is often the prodromal stage to Alzheimer's disease (AD). Most patients with MCI harbor the pathologic changes of AD and experience progressive brain metabolites change. We investigated the changes of metabolites in bilateral hippocampus (HIP) and posterior cingulate gyrus (PCG) in MCI patients, and observed the effects of moxa cone moxibustion on the metabolites of bilateral hippocampus and posterior cingulate gyrus in patients with MCI. Methods: Sixty-nine patients with MCI and sixty-seven cases of age-matched normal controls (NC) were enrolled in this study.MCI patients were randomly divided into drug group(Drug, n=24),sham acupoint group(Sham, n=20) and acupoint group(Acupoint, n=25).All subjects received proton magnetic resonance spectroscopy ( 1 H-MRS) for measurement of NAA/tCr and Cho/tCr ratios in bilateral hippocampus and posterior cingulate gyrus before and after treatment. Mini-Mental Status Examination(MMSE) and Montreal Cognitive Assessment(MoCA) were used to evaluate cognitive function. Results: MCI patients were characterized by lower NAA/tCr in bilateral hippocampus and posterior cingulate gyrus and Cho/tCr in right hippocampus compared to those in NC group(P < 0.05). After two months treatment, NAA/tCr in bilateral hippocampus and posterior cingulate gyrus and Cho/tCr in right hippocampus were increased in the three treatment groups, and there was no signicant difference compared to NC group(P > 0.05).NAA/tCr in right hippocampus and Cho/tCr of right posterior cingulate gyrus of acupoint group was signicantly higher than those in sham acupoint group(P < 0.05) and NC group(P < 0.05).The scores of MMSE and MOCA in the acupoint group and sham acupoint group were better than those in the drug group, especially in the long-term ecacy. Conclusions: Moxa objective imaging basis for and of

Bo'an hospital, Southern medical university [11], exceed those to treat individuals with cancer and cardiovascular disease combined, placing an immense burden on Medicare and Medicaid [12]. AD is the most common cause of dementia and a growing health problem. It affects more than 35.6 million people living with dementia worldwide, with approximately 6.5 million AD patients in China [13], while the global rates are expected to more than triple by 2050 [14]. At present, the pathogenesis of AD is not clear, and there is no effective way to prevent and treat it [15]. Therefore, early detection and prevention of AD is particularly important.
Neuroimaging is becoming increasingly important in the study and clinical diagnosis of AD [16][17].
Hydrogen 1 ( 1 H) magnetic resonance (MR) spectroscopy is a quantitative biochemical imaging technique and follows structural MR imaging as one of the most extensively investigated MR modalities used to assess AD [18]. 1 H-MRS is a non-invasive technique that allows detection and quanti cation of certain biochemical compounds in brain tissue, such as N-acetylaspartate(NAA), creatine (Cr), choline(Cho), and lipids [19,20]. NAA is found at relatively high concentrations in the human central nervous system and is particularly localized within neurons and related to neuronal processes [19,21]. A decrease in NAA concentration is routinely considered as an indicator of neuronal loss or dysfunction [22,23] and has been observed in different brain regions in various neurodegenerative disorders [20,[24][25][26][27][28]. NAA decrease is mostly observed at the moment when the disease is in progression. Cr, which is known to play an important role in energy metabolism, has been reported to be constant throughout the brain and resistant to change in several degenerative brain diseases [19,29,30]. Cr is often used as an internal reference to measure the content of other metabolites in many spectral studies [31,32]. Occasionally Cr can't be used as a reference value [33], because of abnormal energy metabolism in some diseases,such as high malignant tumor. Cho is considered a marker for cell turnover [19], which can re ect the damaged cholinergic neurons in AD. 1 H-MRS technique show tissue damage that may precede the evidence of atrophy on morphological imaging and the accuracy of this technique was demonstrated by the correlation of metabolic MR biomarkers with clinical and neuropathological severity in neurodegenerative disorders such as AD [34]. In addition, the MRS Consensus Group have indicated the 1 H-MRS as a complementary tool to conventional MRI for diagnosing and monitoring disease progression and treatment response in neurodegenerative disorders [35]. 1 H-MRS has been used for early diagnosis and differential diagnosis of AD [36]. Prior studies have demonstrated that atrophy of the medial temporal lobe, including the hippocampal area, is a common structural feature of AD [37]. The hippocampus is involved in learning and memory and implicated early in the pathophysiology of AD [38]. Ishiwata et al reported that the hippocampal atrophy was correlated with the severity of cognitive impairment in AD [39]. Since the hippocampus is involved early in the AD process, it is particularly useful to understand the metabolic change of the hippocampus for early diagnosis and evaluation of intervention [40]. Some structural MRI studies have demonstrated that AD-associated brain atrophy may involve not only the medial temporal lobe but also the posterior cingulate gyrus (PCP), and parietotemporal area particularly in early onset AD [41,42]. Patients with AD and MCI were reported to exhibit tissue degeneration in the posterior cingulate gyrus [43].
Currently, no drug has proven effective in treatment of MCI. Cholinesterase inhibitors have not been shown to decrease risk of progression from MCI to dementia at 1 and 3 years. In addition, cholinesterase inhibitors have limited or no signi cant effects on cognitive function over the short term and may substantially increase adverse effects, based on a meta-analysis of 4 trials (1960 participants)and another meta-analysis of 9 trials (5149 participants).Consequently, cholinesterase inhibitors and memantine are not recommended for MCI treatment and there are currently no FDA-approved medications for MCI [44,45]. We have known a series of studies in men and animals that acupuncture and moxibustion can improve memory on AD or MCI [46][47][48]. Moxibustion could help prevent cognitive impairment, and prevention is currently the only suitable management technique for this disorder [49]. Moxibustion could improve clinical symptoms, regulated blood lipid levels and neuropeptides associated with learning and memory in dementia patients [50,51]. Although studies have evaluated the cognitive effect of moxibustion, no one has tried to prevent mild cognitive impairment by moxibustion at acupoints. The effect of moxibustion on neuroimaging in patients with MCI is still unclear. In this study, the effect of moxibustion on cognitive impairment was evaluated from the perspective of prevention for the rst time, and the effect of moxibustion on brain metabolites in patients with MCI was evaluated by 1 H-MRS.

Subjects
Sixty-nine patients with MCI and sixty-seven cases of age-matched normal controls(NC group) were recruited from January 2014 to December 2017 at The First A liated Hospital of Guangxi University of Chinese Medicine. MCI patients were randomly divided into acupoint group with twenty-ve cases, sham point group with twenty cases, drug group with twenty-four cases.
The inclusion criteria were: (1) Presenium and elderly people aged 55-82, educational objects, (2) MCI group should meet the MCI diagnostic criteria of 2006 edition of Chinese guidelines for the diagnosis and treatment of dementia [52], with the main complaint of memory impairment and abnormal MoCA score [53][54][55], All MCI patients included in this study belong to aMCI. (bilateral). Manipulation: the patient was in supine position. The above acupoints were selected according to "acupuncture and acupoint" edited by Shi Xuemin [57]. Bilateral acupoints were taken each time. Thin vaseline was applied to the acupoints to protect the patient's skin and facilitate xation of moxa cone (produced by Nanyang lvying moxa grass Biological Products Co.Ltd.,12 mm×15 mm) and ignited with alcohol lamp,3-5 minutes each moxa cone, when patients felt burning hot, changed the moxa cone, after acupoint, gently pressed the acupoints with cotton swab,3 moxa cones/per acupoint, 6 acupoints per day, a total of 18 moxa cones for 30 minutes one time treatment in a day. Once every other day, 15 times as a course of treatment, 2 days after rest to continue the next course of treatment, a total of 2 courses. Drug group: oral donepezil hydrochloride[Produced by Weicai (China) Pharmaceutical Co.Ltd:H20050978, batch number:131224a, 130820a, 15060111610044, 5mg×7 tablets] 5 mg, once every night, 30 days as a course of treatment, a total of 2 courses.

Proton MRI and spectroscopy
The 3.0T Siemens superconducting MRI system(Siemens, Verio 3.0T) was used to collect the multi voxel spectrum of MCI and NC groups. PRESS-CSI sequence were used. The TR/TE was 1700/135ms and the frequency was 1200Hz. FOV is 160mm×160mm,and matrix size is 16×16. The volume of interest (VOI) is 80mm×66mm×15mm for HIP and 50mm×50mm×15mm for PCG. The voxel size is 10.0mm×10.0mm×15.0mm for two of them. The VOI size of HIP was designed to matching bilateral HIP, which is locate in the skull base, to avoid the interference from other factors, such as skull, cerebrospinal uid. Automatic pre scanning program was used to adjust the gain of voxel, receive/transmit, semiautomatic shimming, weak water suppression, FWHM < 25Hz, Water suppression level > 95%. After the scanning, A series of post-processing steps are used to obtain the MRS data include water reference processing, Filter, Zero-lling, Fourier transformation, Frequency shift correction, Baseline correction, Phase correction, and Curve tting The chemical shifts of metabolites were 2.02ppm for NAA, 3.03ppm for tCr and 3.22ppm for Cho. tCr as internal parameter, NAA/tCr, Cho/tCr ratio was calculated. MMSE (Chinese version) and MoCA (Beijing version) were evaluated by the same clinician before MRS examination in MCI group and NC group. consists of several tasks, testing direction, memory, attention, calculation, language (naming, repetition, auditory comprehension, reading and writing) and visual special ability, the highest total score is 30. MoCA tests 8 cognitive areas, visual spatial ability, attention, executive function, immediate memory, delayed memory, language, abstraction, calculation and orientation, the highest total score is 30. If the subjects' education years are less than 12 years, the measured total score is the test score plus 1 point.

Neuropsychological assessment
All tests were administered and graded by professionals trained in neuropsychological testing.

Statistical analysis
Statistical analyses were performed using SPSS 22.0. Demographic clinical characteristics and metabolite ratios relative to tCr across the MCI and NC groups were assessed by one way ANOVA tests and repeated measurement analysis, followed by post hoc multiple comparisons of least-signi cant difference (LSD). Statistical signi cance was set at P < 0.05.

Demographic and neuropsychological assessment
The demographic and neuropsychological assessment data of the four groups are listed in Table 1. ANOVA revealed group differences of MMSE scores (F = 78.428, P < 0.0001) and MoCA scores(F = 78.428, P < 0.0001),with a signi cant decrease in MMSE scores and MoCA scores in the three treatment groups compared to the NC (P < 0.0001). Changes of metabolites in hippocampus and posterior cingulate gyrus of four groups before treatment The changes of metabolites in bilateral hippocampus and posterior cingulate gyrus were shown in Table 2. NAA/tCr in bilateral hippocampus, posterior cingulate gyrus and Cho/tCr in right hippocampus of the three treatment groups were lower than those in NC group(P < 0.05).There was no signi cant difference in Cho/tCr in left hippocampus and bilateral posterior cingulate gyrus between the three treatment groups and NC group(P > 0.05).There was no signi cant difference in NAA/tCr and Cho/tCr in bilateral hippocampus and posterior cingulate gyrus among the three treatment group(P > 0.05).

Changes of metabolites in hippocampus and posterior cingulate gyrus of four groups after treatment
After two months of treatment, NAA/tCr in bilateral hippocampus, posterior cingulate gyrus and Cho/tCr in right hippocampus were increased in the three treatment groups, and there was no signi cant difference compared to NC group(P > 0.05, Table 3). NAA/tCr in right hippocampus and Cho/tCr of right posterior cingulate gyrus of acupoint group was signi cantly higher than those in sham point group(P < 0.05) and NC group(P < 0.05, Table 3). There was no signi cant difference in NAA/tCr and Cho/tCr in bilateral hippocampus and posterior cingulate gyrus between drug group and sham point group acupoint group (Fig. 2).

Comparison of MMSE scores in four groups at different time points
There were signi cant differences in MMSE scores between the three treatment groups and the normal control (F = 22.460, P < 0.0001). The results showed that acupoint group and sham point group had better effect, and there was interaction between group and time (F = 21.910, P < 0.0001). The MMSE scores at different time points after treatment also had statistical signi cance (F = 107.689, P < 0.0001), showing that the effect was the best after 2 months of treatment, MMSE showed a downward trend in the fth month compared with that after 2 months of treatment (Table 4 and Fig. 3).

Comparison of MoCA scores in four groups at different time points
There was signi cant difference in MoCA scores between the three treatment groups and the normal group (F = 8.621, P < 0.0001). The MoCA scores of acupoint group and sham point group were higher than those of drug group. There was interaction between group and time (F = 24.415, P < 0.0001), and MoCA scores at different time points after treatment were also signi cantly different (F = 111.378, P < 0.0001). After 2 months of treatment, MoCA scores of acupoint group and sham point group were increased, and still maintained good effect in the fth month of follow-up.The MoCA score in the drug group was better after 2 months of treatment, but with the extension of time, the MoCA score at the fth month of follow-up was lower than that after 1 month and 2 months of treatment (Table 5 and Fig. 3). One month = After one month of treatment; Two months = After two months of treatment; Fifth month = Return visit in the fth month.Compared with the same group before treatment,**P < 0.01,*P < 0.05; Compared with the same group after 2 months of treatment, # P < 0.05.

Discussion
As far as we know, this is the rst study using MRS to observe moxibustion in the prevention and treatment of mild cognitive impairment. Moxibustion is a Chinese medical therapy widely used in East Asian countries [58]. As part of traditional acupuncture practice, moxibustion applies the heat of burning moxa oss, primarily from Artemisia vulgaris, to stimulate speci c spots, known as acupuncture points on the surface of skin [59]. Heat transfer and radiation play an important role in the therapeutic mechanism of moxibustion [60]. The temperature on surface during the burning moxa was controlled to be lower than 60°C, which is the highest temperature that human skin can endure and without scarring,it was found Tissue up to 3 cm deep can be heated during the treatment by an examination magnetic resonance imaging (MRI) of the three-dimensional temperature elevation [61]. Increasing of the temperature in skin via burning moxa can lead to activation of thermosensitive receptors (TRPV channels),which can induce different biological and chemical reactions in the body [62]. It has been suggested recently TRPV1 activation as a novel therapeutic targe that activation of TRPV1 in rodents protects neurons from cytotoxic effects of Amyloid-βpeptide (Aβ), and reverses hippocampal damage and memory impairment [63]. This study observed that after moxibustion treatment for 2 months, the long-term effect of MMSE and MoCA scores was superior between acupoint and non acupoint groups than that of drug group. We speculate that heat transfer and radiation of moxa cone on the acuponint are the critical mechanisms for improving brain metabolites in MCI patients.
A systematic review showed that moxibustion might help prevent AD, and its mechanism of action might include increasing neurotrophin and heat shock protein activity, regulating cell cycle factors, and/or suppressing apoptosis and in ammation [47].  . We found that NAA/tCr and Cho/tCr in the right hippocampus of MCI patients were lower than those in the normal control group, suggesting that NAA/tCr and Cho/tCr in the right hippocampus may be sensitive markers for early diagnosis of MCI.
Our results suggest that right hippocampus NAA/tCr and Cho/tCr ratios measured at MRS may be sensitive markers of future progression to dementia in a clinically de ned elderly population with isolated, mild memory complaints. Moxa cone moxibustion can effectively regulate the metabolism of NAA/tCr and Cho/tCr in the brain of MCI patients, so as to improve the cognitive function. The application of 1 H-MRS for the metabolic evaluation of consequences of hippocampus and posterior cingulate gyrus in the brain may help understand disease symptoms and progression as well as the mechanisms of brain plasticity in general.

Limitations
There were some limitations in our research. First of all, we used long TE in this study, which may lead to biased results, because the T2 relaxation time between queues is unknown, which is a de ciency of this study. Secondly, due to the in uence of time, manpower and material resources, this research did not study the stimulation amount of moxibustion. Is the in uence of stimulation amount of moxibustion on clinical e cacy related to the size of moxa cone, temperature, time and times of moxibustion? These are the problems that need to be optimized and further studied in the future.

Conclusions
The NAA/tCr and Cho/tCr ratios in right hippocampus may be sensitive markers of future progression to dementia. Moxa cone moxibustion can effectively regulate the metabolites of NAA/tCr and Cho/tCr in the brain of MCI patients, so as to improve the cognitive function. The application of 1 H-MRS for the metabolic evaluation of consequences of hippocampus and posterior cingulate gyrus in the brain may help understand disease symptoms and progression as well as the mechanisms of brain plasticity in general.

Declarations
The studies involving human participants were reviewed and approved by the Medicine Ethics Committee of the First A liated Hospital, Guangxi University of Chinese Medicine. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identi able images or data included in this manuscript.

Consent for publication
All authors have approved of the manuscript and agree with its submission.  Figure 1