Claims acupuncture 'staves off dementia' are missing...

"Acupuncture may help elderly people retain their memory, research suggests," the Daily Mail reports.

But the research the news is based on isn't new; it is in fact a review of previous trials, most of which were judged as being of poor quality.

This was a review that pooled the results of five Chinese studies assessing the effectiveness of acupuncture in treating what is known as mild cognitive impairment (MCI).

MCI describes when people develop problems with thinking and memory which are not severe enough to have a significant impact on daily life. A cause of concern is that around 1 in 10 people with MCI will go on to develop a form of dementia within a year; usually Alzheimer's disease.

The review stated that "acupuncture appears effective [for treating MCI]". However, there are a number of important caveats to consider before taking the statement at face value.

The Chinese studies compared acupuncture as an intervention with a drug called nimodipine. This isn't licensed to treat MCI in the UK (in fact there are currently no licensed treatments). So it's very difficult to draw any comparisons or implications from such findings.

The studies were overall of poor quality with a high risk of bias, covered a relatively small non-Western population, didn't look at dementia outcomes, and didn't provide adequate safety information.

In conclusion, this review doesn't provide evidence that acupuncture is safe for people with MCI or will prevent them developing dementia.

Where did the story come from?

The study was carried out by two researchers from Wuhan University in China. No sources of funding are reported and the authors declare no conflicts of interest.

The study was published in the peer-reviewed medical journal Acupuncture in Medicine.

The Daily Telegraph reported the findings of the study at face value without recognising its many limitations.

The Daily Mail's headline: "Could acupuncture starve off dementia," is incorrect as the study didn't even look at dementia outcomes. But the Mail did include analysis from independent experts, such as Professor Edzard Ernst of the University of Exeter, who said: "This is a perfect example of the 'rubbish in, rubbish out' phenomenon which is well-known to authors of systematic reviews – if the primary studies are flawed, the review of such studies will be flawed as well."

What kind of research was this?

This was a systematic review and meta-analysis which aimed to gather the available evidence from randomised controlled trials (RCTs) looking at the effectiveness and safety of acupuncture for treating mild cognitive impairment (MCI). This is a pre-dementia state when people start to have some problems with memory and thinking. It is thought that around 10 to 15% of people with MCI will develop dementia within one year.

There are currently no drugs or treatments licensed to slow the progression of MCI in the UK. Some studies from other countries have suggested that traditional Chinese acupuncture may have a beneficial effect in various brain diseases, including Parkinson's, vascular dementia and Alzheimer's. Some have also studied MCI, which the authors of this review aimed to look at.

A well conducted systematic review is the best way of gathering the available evidence on an intervention, but the pooled findings will only ever be as good as the studies they include.

What did the research involve?

The researchers searched several literature databases up to July 2015 to identify randomised, or partially randomised, controlled trials that compared a group who received acupuncture (alone or with other treatment) for MCI with a control group receiving another active treatment. They focused on MCI with predominantly memory loss symptoms (amnestic) rather than thinking problems (non-amnestic).

Studies needed to have looked at cognitive outcomes using at least one validated scale such as the Montreal cognitive assessment (MoCA), mini-mental state examination (MMSE), clock drawing task (CDT) or Wechsler memory scale (WMS).

Two reviewers assessed the quality of studies for inclusion and extracted data.

Five trials met eligibility criteria and pooled in meta-analysis. All five studies were published 2012-13, and they all appear to be Chinese.

They include 568 people with MCI, 288 receiving acupuncture and 280 in control groups who all received nimodipine. In two of the trials the acupuncture group also received nimodipine.

In the UK, nimodipine is only licensed for the treatment of neurological problems following subarachnoid haemorrhage (bleeding between the membranes covering the surface of the brain). There has been some research, again mainly in China, looking at the effects of nimodipine in treating MCI as there was speculation it may help improve blood flow in the brain.

The size of the individual studies ranged from 26 to 94 people. In four trials acupuncture was given for eight weeks (three to five times a week), in one trial treatment was for 12 weeks. They used acupuncture points specified in Chinese medicine.

What were the basic results?

The three trials comparing acupuncture with nimodipine found that acupuncture was significantly more effective. It improved MMSE scores by an average 0.99 points compared with nimodipine (95% confidence interval (CI) 0.71 to 1.28). Two of the trials also assessed picture recognition scores and found they were also better with acupuncture. Two trials assessed CDT scores – one not giving any results and the other reporting acupuncture didn't help.

The two trials comparing acupuncture plus nimodipine with nimodipine alone also found that the combination improved MMSE scores (average difference 1.09, 95% (CI) 0.29 to 1.89). One reported that it also increased picture recognition.

Adverse effects were reported by three of the studies. Adverse effects of acupuncture included redness at the injection sites and, in one study, fainting.

How did the researchers interpret the results?

The researchers conclude: "Acupuncture appears effective for [amnestic] MCI when used as an alternative or adjunctive treatment; however, caution must be exercised given the low methodological quality of included trials. Further, more rigorously designed studies are needed."

Conclusion

This review aimed to gather evidence for the safety and effectiveness of acupuncture to treat mild cognitive impairment.

It found some evidence that acupuncture may have some efficacy, but there are many important cautions to this research:

All trials compared acupuncture with nimodipine, which is not licensed for this use in the UK. Since there are no treatments or interventions licensed in the UK to prevent progression of MCI, it's hard to draw any comparisons or implications from such findings.
There are only five relatively small studies, all of which seem to be Chinese populations. We don't know that study populations or acupuncture practices could be applied to the UK.
The trials overall seem to be of poor quality. Only one of the five trials used an acceptable method of randomisation. In the remaining ones it wasn't clear that they were properly randomised. No trials used a placebo/sham acupuncture intervention, and it had to be assumed that both participants and assessors were aware of the treatment that had been given. All of these things may introduce bias.
The trials have only assessed changes in cognitive test scores, such as the mental state score. They've not actually looked at progression to diagnosed dementia as an outcome.
The duration of acupuncture was 8 to 12 weeks, but we can't say anything from this on how long acupuncture courses would have to be or whether any effects would be sustained after treatment stopped.
Side effects were poorly reported by these trials. We don't know that this treatment would be safe.

Overall, this review doesn't provide evidence that acupuncture will prevent people with mild cognitive impairment developing dementia.

There is currently no treatment or intervention available for MCI in the UK, and there is no evidence to say that this is going to change anytime soon. It's not possible to know which people with MCI will progress to dementia.

Nevertheless it is helpful for family, friends, and the person themselves to recognise if they are having problems with thinking and memory as this can help to make sure they get the support they need.

If you are concerned about someone you know, encouraging them to see their GP is a useful first step.