In recent years, previous studies have shown that acupuncture might be effective in improving the cognitive function of AD patients [
29,
30]. The results of this study will contribute to a better understanding that whether EA has additional effects to donepezil for improving the cognitive function of patients with AD.
Although clinical diagnosis is the core criteria for AD patients, preclinical diagnostic criteria, such as bio-marker evidence for AD, are also used in clinical practice and research [
17]. Medial temporal lobe atrophy (MTA) as the typical imaging appearance can differentiate AD from ageing with a sensitivity and specificity of 80–85%. The severity of MTA combined with clinical information seems justified to be taken into account when diagnosing AD in clinical practice [
31]. Therefore, the included criteria of this trial are based on the degree of atrophy of hippocampal or medial temporal lobe volume (measured by MRI). Moreover, as the MTA-scale scores differentiate between moderate to severe AD patients with a sensitivity of 70–100% and a specificity of 67–96%, according to the severity of the condition of participants, subgroup analysis will be applied in the statistical analysis. Furthermore, donepezil is approved by the FDA to treat the symptoms of mild to moderate AD [
24,
25] and studies have shown that it can improve the neuropsychiatric, cognitive, and global functions [
32‐
34]. Accordingly, donepezil was chosen as a positive control in this trial. Additionally, an appraisal of drug evolution for AD has shown that the primary outcomes of trials can be described within cognitive, functional, global change, and severity domains [
17]. On the basis of this appraisal, the primary outcome is the ADAS-cog score. The ADAS-cog, assessing the severity of impairment of selective areas of cognition, is a sensitive and reliable neuropsychological test [
35]. Moreover, a previous trial [
36] which had a similar design also evaluated the change in ADAS-cog score after treatment, but follow-up effectiveness was not assessed. Therefore, a 12-week treatment with a follow-up at 24 weeks is expected to better demonstrate the additional effect of EA to donepezil in treating AD.
However, a limitation of this trial is that participant blinding is not possible due to the two different therapeutic strategies employed for each group. Therefore, to minimize any bias that may result from the two treatments, the process of randomization will be under strict control and the investigators and data analysts are to be blinded. Acupuncture as a complex intervention produced greater response expectancies and placebo effects [
37,
38]. However, the expectancy effects of acupuncture cannot be estimated in this trial due to the impaired cognitive function of AD patients.