Stimulated by Wu et al 2024.[1]
IA – intradermal acupuncture
SSRI – selective serotonin reuptake inhibitor
MDD – major depressive disorder
EA – electroacupuncture
MA – manual acupuncture
IF – impact factor
Zung SDS – Zung self-rating depression scale
PSQI – Pittsburgh sleep quality index
FC – functional connectivity
MCID – minimal clinically important difference
GPT – generative pre-trained transformer– key to acronyms
I have highlighted papers on, or related to, depression a fair bit on this blog (more than 5% of blogs so far) – see the category depression. MDD comes up less often, indeed this is only the third time it has appeared in the title of a blog. Type MDD into the search box on the right to see the others.
This paper is somewhat unusual because rather than using EA or MA, they plump for the less well-known acronym IA, where ‘I’ stands for intradermal. It is also the first time acupuncture has appeared in the title of a paper published in the journal Neuropsychopharmacology (IF 6.6).
This is a multicentre 3-arm trial (n=120) based in Hangzhou, China. Active IA was compared with sham IA and no IA in 3 parallel arms. IA was applied using the Seirin Pyonex needles (the description sounds like these although the paper only mentions Seirin). The length and gauge of the needles is not recorded, but personally, I always use the pink packets in adults (0.20×1.5mm). There is a convenient sham version that is identical apart from the absence of the tiny needle. The IA groups had either active or sham versions applied to 4 bilateral points (HT7, PC6, SP6, LR3) approximately every 72 hours for a total of 10 sessions over 6 weeks. The devices were removed the day before the next session. We are not told about disposal of contaminated sharps, albeit tiny ones.
All patients were stable on an SSRI for 6 weeks prior to inclusion, so apart from the IA groups, the no IA group simply continued on their SSRIs alone. This is a design that gives some measure of efficacy as well as effectiveness, which can be useful in acupuncture research due to the problems with sham techniques. The latter are often quite effective in their own right, but you don’t get to measure that without a third (no acupuncture) arm. The downside of 3 groups, of course, is a reduction in statistical power and consequent increased risk of type 2 errors.
The primary outcome was the HAMD-17, which was also used to determine inclusion of course. Scores of 0 to 7 are considered normal on this measure. For this study they included patients with a score of 17 or above, which equates to moderate to severe depression, mild depression sits between 8 and 16. The baseline mean was just over 20.
Other outcomes included the Zung SDS and the PSQI, but they also used fMRI to evaluate a subgroup of patients (n=20) along with 20 healthy controls. The patients were taken from the active IA arm of the trial. So, we got a comparison of the functional connectivity (FC) in the MDD brain compared with the healthy brain and what changed in the MDD brain after a course of active IA treatment alongside an SSRI.
All three groups had a clinically significant improvement in HAMD-17 score (that is a reduction of 3–4 points for moderate to severe depression). The active IA and SSRI group improved by 9.6 points at week 6 (after treatment) and 10.6 and week 10 (end of follow-up). This was almost double the improvement of the sham IA plus SSRI group and the SSRI alone group, which was very highly significant. Similar results were seen in the Zung SDS outcome, perhaps unsurprisingly; however, there were no differences in the PSQI improvements. The latter improved to clinically significant degree (MCID is 3) from baseline in all 3 groups, which was enough to go from a mean score bordering on moderate to severe sleep disturbance down to the mild to moderate boundary.
Perhaps the most interesting result was in terms of adverse events. There were no adverse events related to the needles of course, as these were only 1.5mm long. Infection risk is the main thing we worry about with these. There was a significant reduction in drug-related side effects in the active IA group. Palpitations, somnolence, and nausea were almost eliminated completely in the active IA group but were present in 15% to 20% in the other groups. This is something I noted previously on the blog in the first NMA of acupuncture in MDD: Acupuncture and MMD.
The fMRI findings necessarily revolved around the predefined ROIs, which were the striatum, the ventral tegmental area, the dorsal raphe nucleus, and the median raphe nucleus. All areas related to the dopamine reward pathways and the 5-HT system.
Whilst there was quite an array of differences between the MDD and healthy brain in terms of FC, the changes in the MDD brain following IA were more limited and seemed to focus on striatum / frontal cortex and cerebellum / limbic system connectivity, modulating activity both up and down, but mostly down.
For the first time, I used Scholar GTP (one of ChatGPT’s many GPTs) to analyse this paper and I continued a discussion about a variety of specific aspects of the paper with this GPT. It was a very useful endeavour, but as you will see, if you follow the link, I had to challenge some responses. I finished by asking whether or not IA (as opposed to AI) should be used in the UK NHS… you can guess the response, or you can read it here.
References
1 Wu X, Tu M, Yu Z, et al. The efficacy and cerebral mechanism of intradermal acupuncture for major depressive disorder: a multicenter randomized controlled trial. Neuropsychopharmacology. Published Online First: 8 December 2024. doi: 10.1038/s41386-024-02036-5
The BMAS Blog 
You must be logged in to post a comment.