Article Text


In this issue
  1. Adrian White

Statistics from

One special application of acupuncture is in the operating theatre where the aim is to prevent symptoms that have yet to occur, such as pain or nausea. This is quite different from treating chronic pain or other symptoms that exist already. We have two papers on acupuncture in surgery, and they illustrate two different approaches to acupuncture research, illustrating an important principle.

In the first study, Dias and colleagues (see page 65) describe a randomised controlled trial of acupuncture for pain, done without any preliminary testing. The overall results are disappointing – for reasons discussed in Hayhoe's commentary. In the second study, on nausea in children after throat surgery, Norheim and colleagues (see page 74) decided to undertake a pilot study first to make sure all the detailed procedures were in order, before going on to a definitive study. They used needles during surgery and acupressure bands afterwards in the postoperative ward (see figure). The study was a ‘success’ in suggesting some important modifications to the subsequent, definitive study: it did not show an effect but that was not its purpose, and it should be regarded as a positive study in meeting its aims. (The main study has already been completed but the results are not yet publicly available).

The important principle arising from seeing these two papers together is the need to be sure about every aspect of a study before committing resources to a definitive trial. Probably, many ‘negative’ studies of acupuncture have failed, for purely technical reasons, to demonstrate an effect that does exist. This problem is known as poor ‘model validity’ – the study design is less than optimal.

Another aspect of this model validity in acupuncture research is the lack of objective outcome measures, particularly biological markers (they minimise the response bias from lack of blinding of the practitioner). Lo and colleagues (see page 74) had a neat idea for an objective measure: they knew that the frequency of migraine attacks is related to cortical excitability, and that can be measured. Regrettably, the measurements they obtained did not match the clinical improvement in their patients.

Figure 1

PC6 acupuncture in paediatric surgery

Another objective measure for acupuncture – this time for a physiological effect, not a clinical one, is a change in finger flexion response to vibration. The method is already well accepted, but Takakura and colleagues (see page 78) wanted to refine the details of the analysis, and here report their success. This is rather technical, but important in principle.

Also ‘rather technical, but important in principle’ – are the STRICTA criteria which have now been revised (see page 83). These criteria define how authors should describe the acupuncture intervention they have used, and so are crucially important in improving the quality of scientific reports in the international literature. They have been developed and revised by an international team led by MacPherson, who has been instrumental in getting the STRICTA criteria accepted as an official extension to the CONSORT criteria. These are themselves internationally accepted as valuable in improving the quality of scientific reporting. Also like CONSORT, the STRICTA list can provide a check-list for the aspects of needling that researchers need to define when planning a study.

If all of this seems too technical for regular acupuncturists treating everyday conditions in ‘real’ patients, then please turn to the Case report for inspiration (see page 94). A patient with chronic neck dystonia was being managed with regular botulinum toxin injections, but then underwent a course of acupuncture. After that, she could reduce the frequency of her injections – saving herself some side effects and saving the health service some money. This kind of report reminds us not only where all this research starts out from – careful clinical observations – but also where it hopes to go – cheaper and more effective treatment for patients

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.