Stimulated by Gu et al 2025.[1]
EA – electroacupuncture
PD – Parkinson’s disease
NMS – non-motor symptoms
MA – manual acupuncture
IF – impact factor
PDSS – Parkinson’s disease sleep scale
REM – rapid eye movement sleep
RBD – REM sleep behaviour disorder
RLS – restless legs syndrome
SDB – sleep disordered breathing
EDS – excessive daytime sleepiness– key to acronyms
I attended an academic session of an acupuncture conference recently and the presentations were mainly research projects undertaken by students as part of their course work. The research was not groundbreaking of course, but I found that I enjoyed this part of the meeting the most. One of the presentations was a review on non-motor symptoms (NMS) in PD, which previously had not really been on my horizon as a possible focus for acupuncture. Insomnia is one of the top NMS in PD, along with depression, fatigue, and cognitive impairment (in more advanced PD).
I have highlighted a single centre trial (n=78) on MA for sleep quality in PD on the blog last year – Acupuncture for sleep in PD 2024.[2] That research was from Guangzhou and published in JAMA Network Open (IF 10.5 at the time). This time the trial is multicentre (8 hospitals in Shanghai), somewhat larger (n=136), and published in one of the journals of the International Parkinson and Movement Disorder Society – Movement Disorders Clinical Practice (IF 2.7).
Both trials used the PDSS for the primary outcome.[3] I described this outcome on the blog (link above) related to the previous paper, but briefly, it is a 15 item scale with scores from 0 to 150 and larger scores equate to better sleep quality.
Whilst the Guangzhou paper used MA, this one from Shanghai used EA, although it is not clear whether or not, and how, all the needles were connected. The protocol describes all the points in some detail. There were 4 in the midline of the scalp from GV20 to Yintang. The rest were all bilateral – PC6, HT7, Anmian (halfway between GB20 and TE17), LI4, SP6, KI3, LR3, GB34, GB5, GB6. I make that 24 needles in total and the EA device they used had 6 output channels, so I guess they could have connected them all up using 2 devices, but this is really not at all clear.
The sham technique was a non-penetrating form using what was described as a ‘pragmatic placebo needle’, and presumably it was applied to the same points as listed above.
The treatment schedule was similar to that in the previous trial, but longer – 3 sessions per week for 8 weeks (rather than 4 weeks) and then a further 12 weeks of follow-up (rather than 4 weeks). Both used non-penetrating sham controls, but the Guangzhou paper describes the process in more detail.
Another difference of the Shanghai trial was the use of objective sleep parameters in the form of polysomnography, which was performed at baseline and at the end of the 8 week treatment phase.
The PDSS improved significantly in the EA or MA groups of both trials compared to the sham controls. The change from baseline in the current trial was just over 15 points, but the previous smaller trial with a shorter treatment course registered a change of almost 30 points. The sham procedures were both non-penetrating and there was also a marked difference in the PDSS improvement in these groups: 2 points vs 10 points.
There were marked baseline differences in the PDSS with the current trial PD population starting at 70 and the previous trial at 85. This reflects the inclusion of some patients with more severe disability (Hoehn and Yahr stage 4) in the current trial. The previous trial limited inclusion to stages 1 to 3.
Generally, I assume that when the baseline is worse there is a greater possibility for improvement but that probably reflects the fact that I am often looking at pain outcomes in acupuncture trials. In this case, some of the central changes in advanced PD that impact on sleep outcomes may be irreversible.
Moving to some other news, this trial assessed 4 other coexisting sleep disorders, including REM sleep behaviour disorder (RBD), RLS, SDB, and EDS. The most prevalent of these was SDB with 48% of patients having this comorbid sleep disorder. RBD came in at 46%, RLS at 35% and EDS at 28%.
A subgroup analysis of the results in these comorbid sleep disorders revealed a dramatic difference in results for just one of them. Patients with RBD were much more likely to respond to EA than those without it. It would be interesting to know whether the RBD improved as well, but that was beyond the scope of this trial.
Finally, the other new findings from this trial were those derived from polysomnography (objective sleep parameters). Patients receiving real EA had highly significant differences (mostly improvements) in all objective sleep parameters with an increase in total sleep time of over 2 hours from a baseline mean of around 6 hours. The sham group mean increased by 40 minutes. In the real EA group, deep sleep (stage N3) and REM sleep increased and the other sleep stages (N1 and N2) reduced. The biggest changes in the real EA group look as though they were in REM sleep. I think that is a new feature for me – EA improving REM sleep.
References
1 Gu S-C, Yin P, Yang M, et al. Efficacy and Safety of Electroacupuncture on Insomnia in Parkinson’s Disease: A Multicentre, Randomized, Controlled Trial. Mov Disord Clin Pract. Published Online First: 28 May 2025. doi: 10.1002/mdc3.70139
2 Yan M, Fan J, Liu X, et al. Acupuncture and Sleep Quality Among Patients With Parkinson Disease: A Randomized Clinical Trial. JAMA Netw Open. 2024;7:e2417862. doi: 10.1001/jamanetworkopen.2024.17862
3 Chaudhuri KR, Pal S, DiMarco A, et al. The Parkinson’s disease sleep scale: a new instrument for assessing sleep and nocturnal disability in Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2002;73:629–35. doi: 10.1136/jnnp.73.6.629
The BMAS Blog 
You must be logged in to post a comment.