Sir, I read with interest the recent article on gag reflex (BDJ 2006; 201: 721–725) but would point out that a randomised double blind clinical trial (RCT) with cross-over on the effect of acupressure on controlling the gag reflex while taking maxillary impressions has been published previously, demonstrating no apparent beneficial effect.1 While the same meridian points are stimulated for either acupuncture or acupressure, the possibility that one may be more efficacious than the other can only be established if a similar RCT methodology is followed for both.

In that regard, the following comments could be constructive for any future research. Firstly, the title of the article describes it as being an audit, when in fact it was a non-randomised clinical trial. Had it been an audit, the stated aim 'to test if acupuncture was able to control the gag reflex' would have been accepted as proven, and the study would have investigated operator compliance with clinical standards of relevance to acupuncture instead.

Secondly, the need for ethical approval for the study was dismissed because an informal enquiry to a local ethics committee had deemed it unnecessary so long as the acupuncture would be a supplement to the dentists' usual techniques for controlling the gag reflex and no placebo procedures would be involved.

The study's methodology then describes that the first maxillary impression was done normally while the second one was carried out after acupuncture, in order to measure the difference in gagging between the two. However, unless each second impression was also clinically required for every patient's dental treatment, ethical approval should have been sought. Equally, unless the participating dentists' usual technique for controlling the gag reflex was always nothing, the use of acupuncture to test its potential efficacy would have committed them to unethically refraining from using any, for fear that otherwise would confound the results.

Next, the authors describe the use of two gagging indices but they fail to calibrate the 21 separate users or to subsequently assess inter-observer parity and intra-observer consistency. However, the main deficiency is that none of the patients were asked to rate their own levels of nausea after each impression, perhaps using a 100 mm visual analogue scale as described elsewhere.1 While clinicians can use indices to score the severity of gagging in relation to the effect on treatment compliance, only patients can score how nauseous they feel during impressions, bearing in mind that both subjectively1 and objectively2 differences between operator perceptions and patient evaluations exist.

In addition, to ensure that every impression invoked a similar challenge to gag, the standardisation of each of them should have been verified by casting and then measuring the mid-line palatal lengths of the untrimmed study models.1

Equally, since the operators were not blind, for those patients who might have demonstrated a gag severity on the borderline between two grades of the index, the possibility that a higher grade before acupuncture and a lower grade after being selected cannot be excluded, with a concern that as a result the recorded statistical differences in gagging between the two approaches could be due to bias.

Similarly, the authors discount the possibility of the placebo effect with acupuncture, on the basis that once the first impression had activated a gag reflex, the second would automatically initiate a similar response, unless acupuncture was effective. This view would only be valid if the patients knew that the second procedure was going to be performed in exactly the same way as the first. However, the fact that they knew otherwise might well have raised their hopes to the point where it could have resulted in a placebo response.

It is therefore commendable that the authors suggest an RCT should be undertaken to answer the question of the potential efficacy of acupuncture in controlling a severe gag reflex, but with the present study's limitations it would be unsound to suggest as yet that the technique has anything definite to offer.

The author of the study, Dr Palle Rosted, responds: Thank you for your comment on our paper regarding the use of acupuncture for controlling the gag reflex. 3 We are well aware that a study regarding the use of acupressure in controlling the gag reflex has been published in the past. 4

However, it concerns two different techniques, acupuncture and acupressure, which from a neurophysiological point of view have a different mode of action. Acupuncture is an invasive procedure and has both a peripheral and central effect involving neurotransmitters such as enkephaline, dynorphine, serotonine and endorphines.5 Acupressure is a non-invasive procedure, and a complete understanding of the mode of action is still under debate. However, the maximal effect one would expect is a peripheral effect, releasing neurotransmitters such as enkephaline and maybe dynorphine only.

Secondly, the aim of our study was to investigate the possible effect of acupuncture on the gag reflex and not on nausea. We accept that gagging and nausea in many ways can be triggered by the same type of stimulation, touch, smell etc, but they represent different stages of a process which might culminate in full blown vomiting. Furthermore gagging is a rather simple reflex, whereas nausea is a more complex reaction.

In our study, the aim was to investigate if acupuncture made a difference in the severity of the gagging, assessed by previously tested scales,6 and if there was a difference in the number of patients who accepted dental treatment before and after acupuncture. We demonstrated a total change in the attained variable of 53% (p <0.001). Moreover, we demonstrated that prior to acupuncture 37 out of 37 patients did not accept dental treatment; after acupuncture 30 patients out of 37 patients accepted dental treatment.

In the study regarding acupressure the endpoint was to assess if a reduction in the sensation of nausea related to taking a maxillary dental impression could be noticed.4 In the mentioned study there was no difference in the outcome in the acupressure and placebo group.

However, this is not surprising. In both cases pressure was delivered on the forearm. In the test group on a well known acupuncture point (PC-6), in the placebo group on a random point on the forearm. However, this is not a true placebo procedure and does not tell us anything about the efficacy of acupuncture.

The correct conclusion should have been: pressure on the forearm seems to give a mean reduction in nausea of 30% in both groups. Apparently there is no difference if the pressure was delivered to an acupuncture or a non-acupuncture point.

Unfortunately, this misinterpretation is not uncommon in acupuncture studies as researchers often forget to analyse the neurophysiological mechanism behind the intervention. Pressure on the forearm will in all circumstances activate the spinal segments and thus will at least give a segmental effect involving enkephaline and dynorphine mechanism. If a central effect will occur, involving release of endorphine and serotonine at all is dubious.

As a consequence of these major differences in the two studies3,4 a direct comparison is not possible.

Dr Chate mentioned correctly that our study should have been done as a proper RCT, and we agree. However, we have at no stage pretended that this study was a proper designed RCT. Moreover, Dr Chate mentioned that it is essential if the second impression taking was clinically required. As none of the test patients accepted an impression on the first attempt, the second attempt was obviously clinically required.