"Does acupuncture really work after a stroke?" "Will having acupuncture and physiotherapy together help the patient recover faster?" Many patients and families face this kind of choice as they look for a way forward in rehabilitation. Eye-catching claims such as "astonishing results from acupuncture" and "works with a single needle" are common. But to make a responsible medical decision for a family member, one must never rely on individual "success stories" alone, and must look at the level and quality of the clinical evidence.
Looking purely from the physiotherapy professional angle and citing the highest level of evidence-based medical literature (including Cochrane systematic reviews, the AHA/ASA clinical guideline, and a meta-analysis comparing real and sham needling), this article looks objectively at one core question: for movement function and self-care after a stroke, can acupuncture bring an "extra benefit"? And should its relationship with physiotherapy be defined as "replacement" or "supplement"? This article does not mean to belittle acupuncture, but it will also never inflate an unproven hypothesis into a "miracle effect".
The key point first. For regaining movement of the arms and legs and self-care after a stroke, the highest level of medical research has not confirmed that acupuncture brings an "extra benefit". Studies found that the effect of real acupuncture was not clearly different from that of sham acupuncture, and the overall quality of evidence is generally low to very low. The rehabilitation method with reliable research behind it is active, repetitive training aimed at real daily movements within physiotherapy. Patients can treat acupuncture as a supplement if they wish, but should not give up or delay physiotherapy because of it.
Understanding the levels of medical evidence
To judge whether a therapy truly works, one or two research reports claiming that it "works" are not enough, because an individual clinical trial may have too small a sample or flaws in its design, and studies with "positive results" are often more easily published in journals than those showing "no effect" (known in academia as publication bias). Evidence-based medicine therefore stresses pooling similar studies systematically and rating their quality strictly before drawing a conclusion. Acupuncture for stroke is an excellent example.
The most authoritative pooled research on acupuncture and stroke rehabilitation at present is the Cochrane systematic review (Yang et al. 2016). It pooled data from 31 randomised controlled trials with 2,257 patients in the subacute and chronic stages of stroke and reached a fairly conservative conclusion: the benefit of acupuncture for improving dependence in daily life (Barthel Index) and some neurological deficits was rated only low / very low quality by GRADE criteria. Most of the trials were insufficiently rigorous and small, and there was publication bias, so the existing evidence is not enough to support listing it as routine treatment.
Why does it matter that 29 of the 31 trials are concentrated in the same region?When nearly all of the research data come from a single region, and the funnel plot also shows publication bias, it means the research field may tend to publish only results that "work", making the overall effect look exaggerated. This is not aimed at any country. It is the objective standard that evidence-based medicine must check when judging whether research results can be trusted.
Comparing the effect of real and sham acupuncture
To tell whether a therapy has "a real physiological effect" or is purely a "psychological comfort effect", the most rigorous approach is to set up a sham acupuncture control group, in which patients are led to believe they have been needled when the relevant points were not actually stimulated. If the rehabilitation results of real and sham needling show no difference, it reflects that the effect mostly comes from non-specific factors or a placebo effect and not from the needling itself.
Sze and colleagues (2002) carried out exactly this kind of pooled analysis. In a random-effects model comparing real acupuncture with sham acupuncture, the pooled estimate for improvement in disability was only 0.07 (95% confidence interval −0.34 to 0.48). Because this confidence interval crosses zero, it means that compared with sham acupuncture, real needling has no statistically significant extra benefit in improving motor function or reducing disability.
What this set of data says
- The pooled effect is only 0.07, and the confidence interval is −0.34 to 0.48, crossing zero in the middle
- Compared with sham acupuncture, real acupuncture did not bring a statistically clear extra improvement
- This suggests that some of the cases that seem to "work" in everyday experience may come only from psychological comfort or other non-specific factors
What this set of data does not rule out
- It does not mean "acupuncture is harmful", and reports of side effects from acupuncture are generally few clinically
- It does not deny the subjective feeling of comfort that patients have after acupuncture
- It does not say that acupuncture cannot be done at all. It only says the research evidence is not enough to support treating acupuncture as the main treatment
This is exactly the professional line of "reporting honestly and not rejecting blindly". A placebo effect can indeed make a patient feel relieved, but when deciding whether to make a therapy the main means of stroke rehabilitation, the medical community must work out "how much real benefit the therapy has left once the placebo effect is taken away". The highest-level research available shows that this benefit has not been confirmed to date.
What the international clinical guideline (AHA/ASA) says about acupuncture for stroke
Besides pooled research data, the recommendations of authoritative international clinical guidelines are also valuable. The adult stroke rehabilitation guideline published by the American Heart Association / American Stroke Association (AHA/ASA) in 2016 (Winstein, Stein et al. 2016) states clearly that acupuncture is not recommended for improving activities of daily living and upper limb function, and that the scientific evidence for using it to promote recovery of limb movement and walking ability is also insufficient.
A 274-person RCT cited in the guideline (comparing acupuncture alone, physiotherapy alone, and the two together) more directly supports the conclusion that "acupuncture cannot replace physiotherapy": the data show that adding acupuncture on top of usual physiotherapy brings no synergistic effect. In other words, adding acupuncture while having physiotherapy does not make rehabilitation progress "extra fast".
The rehabilitation method with reliable research support: active training in physiotherapy
Having made clear the limits of acupuncture in scientific evidence, it is more important to recognise "which therapies do have firm evidence". The core of stroke rehabilitation backed by solid medical evidence is active, highly repetitive, task-specific training built on promoting neuroplasticity, together with strengthening and gait rebuilding. This is essentially different from the passive stimulation pattern of acupuncture. Through a large amount of active practice, physiotherapy directly stimulates the brain to reorganise damaged neural networks in the golden period of recovery.
- The patient practises target movements actively, repeatedly and task-specifically
- Drives neuroplasticity, reshaping damaged circuits in the golden period
- Strength and gait rebuilding have reliable evidence
- The training dose can be adjusted to severity and stage
- Passive stimulation, with the patient not actively driving learning
- Extra benefit for movement / self-care has not been confirmed
- Difference between real and sham needling crosses zero (Sze 2002)
- Can serve as a supplement by personal choice, and should not replace active training
The neuroscience behind the golden period of recovery (such as sensitive periods and the window of brain plasticity), and the course of recovery at different stages, are not covered in detail on this page. To learn more, see the article on the theory of stroke recovery (analysing the neuroscience of the golden period) and the stroke recovery timeline. The most important principle is to hold on to the golden period and make sure active training reaches a sufficient dose, which is the evidence-based path of rehabilitation. Never lose this key period by "waiting for acupuncture to work".
The difference between the "dry needling" used by physiotherapists and Chinese medicine acupuncture
When "needle treatment" is discussed in Hong Kong, it is easy for people to confuse two quite different techniques. Some registered physiotherapists in Hong Kong also perform needling, but that is "dry needling". Its theoretical basis comes from Western neurophysiology and pain science, it targets myofascial trigger points in muscle, and it aims to relieve pain and regulate abnormal muscle tone. This differs completely, in both theoretical basis and treatment aims, from traditional acupuncture in Chinese medicine, which emphasises "pattern differentiation and treatment" and inserts needles at points along the meridians.
Which "needle" does this page discuss? The research evidence this article examines is about treating Chinese medicine meridian acupuncture as the main therapy for stroke rehabilitation, and not about the dry needling technique that physiotherapists use to deal with pain. The two are entirely different practices. Even if a physiotherapist is sometimes seen using needles, it does not mean that Chinese medicine acupuncture can cure hemiplegia after a stroke.
Where acupuncture stands for stroke, and points to note
Taking all the medical evidence above together, an objective and balanced position can be set for acupuncture. Because reports of side effects from acupuncture are generally few, there is nothing wrong with patients using it as a supplementary treatment by personal choice. What is really worth noting is not "whether acupuncture can be done" but "where acupuncture should sit in the whole rehabilitation plan".
- Patients may treat acupuncture as a supplement by personal choice. Its side effects are few, and it is not something that can never be done
- Active task training within physiotherapy is the focus of rehabilitation that research supports
- Acupuncture should not replace the active training of physiotherapy that has scientific research behind it
- Do not delay training in the golden period of recovery by waiting for acupuncture or expecting a miracle
- Before deciding to add any supplementary therapy, it is best to ask the treating medical team for advice first
The professional division of work between swallowing difficulty and daily self-care training
Finally, the professional scope of this page needs to be made clear. This article looks purely from the physiotherapy angle and concentrates on the clinical evidence of acupuncture for recovery of motor function, strength and movement skills after a stroke. It does not cover Chinese herbal medicines or supplements. For other areas of rehabilitation, see the matching specialist pages below.
To learn more about these directions of rehabilitation, see speech therapy after stroke, occupational therapy after stroke and balance and fall prevention exercise.