Acupuncture for PSD 2025

Stimulated by Zhang et al 2025.[1]

PSD – post-stroke dysphagia
RCT – randomised controlled trial
MA – manual acupuncture
SRT – swallowing rehabilitation training
IF – impact factor
c – circa (from Latin and written before a number to mean it is an approximation)
WST – water swallowing test (30ml of water)
VFSS – video fluoroscopy scanning study
SSA – standardised swallowing assessment
MBI – modified Barthel index
sEMG – surface electromyogram
MCID – minimum clinically important change

– key to acronyms

This is a relatively large (n=254) pragmatic RCT of MA compared with SRT in patients with PSD. I should own up here to the fact that I have used this acronym before on this blog for 2 quite different conditions – post-stroke dysphagia and post-stroke depression.

The research was based in Harbin, which is in the far northeast of China, almost directly above the Korean peninsula. It is closer to Vladivostok than to Beijing, which are 500km southeast and 1000km southwest of Harbin respectively.

The paper is just out in the journal Complementary Therapies in Medicine (IF 3.5), a journal that I became familiar with when Adrian White took over as its editor in chief. He took over from Andrew Vickers in c1999 and continued for several years before he took over Acupuncture in Medicine from me and Jacky.

The patients in this study had a stroke confirmed by imaging and at least grade II WST, which is at least a mild impairment of swallowing. The WST has 5 grades, where grade I is the ability to swallow 30ml of water in one gulp without coughing or choking and grade V is the inability to swallow water. The grades in between include increasing hesitancy and coughing. VFSS was also used to confirm some form of swallowing disorder.

Patients were randomised to either MA or SRT. Both treatments were performed for 30 minutes per day, 5 days per week, for 2 weeks. The SRT involved various exercises for the lips, mouth, tongue, and jaw, followed by swallowing training exercises.

MA was performed at a variety of points in the head and neck. The paper lists 12 named points, of which, 3 are in the midline, giving a total of 21 needles per treatment, if my sums are correct. Only 4 of the points listed are numbered points on meridians – GB20, CV23, ST7, LI20. The others are presumably all extra points. Some are well known, but at least 3 of them are not to be found in any of my points resources, at least not with the names used in this paper.

I was slightly alarmed to see that the needles used were 0.35mm by 60mm and I was amused to read that they were inserted by a ‘performer’ rather than an acupuncturist. Whilst the needles were rather longer than absolutely necessary, none were inserted more than 30mm and most a lot less.

The one needle that was inserted 30mm was at CV23 – my favourite tongue point, which lies in the midline, just above the hyoid bone. If inserted 30mm, this needle would pass between the anterior bellies of the digastric muscle, then through the fibres of mylohyoid, between the bellies of geniohyoid, and ultimately reach the genioglossus (mentioned previously here).

The other points included in the protocol targeted a number of both suprahyoid and infrahyoid muscles as well as other regional muscles indirectly related to swallowing, such as masseter and semispinalis.

The primary outcome was the SSA, which involves a clinical examination of neurological and oral motor status (eg conscious level, lip closure, tongue movement etc), swallowing 5ml water, and swallowing 60ml water. It has 18 items, and the total score is from 18 to 46, where 18 equates to normal swallowing ability.

Secondary outcomes were the MBI and sEMG assessment of suprahyoid and infrahyoid muscles. The MBI is a standard test of 10 activities of daily living, including feeding, bathing, grooming, dressing, toileting (3 items), and mobility (3 items). It is scored from 0 to 100, and has been modified from the original index simply by allowing a finer measure of each item, so rather than 5-point steps from 0 to 5, 10, or 15, the MBI uses single point steps for each item.

Outcomes were measured at baseline, after 1 week of treatment, after 2 weeks of treatment, and then after a further 1-month follow-up. The baseline SSA was 33 to 34 and it dropped by just over 10 points in the MA group and just under in the SRT group. The difference between groups was just less than 2 points, and this became significant at the end of the treatment phase and was maintained at the 1-month follow-up.

There were no significant differences in the MBI scores, although they improved in both groups by just over 20 points. The MCID for the MBI is around 10 points.

sEMG results did show a few significant differences between the groups; however, there were only 7 out of 96 probability tests that reached the standard threshold for statistical significance. At this threshold you would expect 4 to 5 simply by chance, but a couple of results would still have been significant at a more stringent threshold.

This is a relatively large study that demonstrates modest benefit of acupuncture over standard rehabilitation, from which we can confidently say that it is at least as good. Since the contact time was the same in both groups, the costs of treatment are likely to be similar, assuming a similar cost of practitioner time.

References

1          Zhang S, Liang B, Tang Q, et al. Therapeutic effect of acupuncture on post-stroke dysphagia: a multicenter, randomized controlled trial. Complement Ther Med. 2025;103200. doi: 10.1016/j.ctim.2025.103200

---

Declaration of interests MC