Statistics from Altmetric.com
Editor—As an endoscopist who routinely uses hypnosis in lieu of sedation for gastroscopy, I was intrigued by Conlong and Rees' study comparing the use of hypnosis in gastroscopy with intravenous sedation.1 Unfortunately, their paper only serves to prove the point that a study is only as good as the question being asked. And, in this case, they asked the wrong question.
It has already been established that gastroscopy can be carried out safely and effectively without intravenous sedation.2 The advantages of doing so have been cited on many occasions. However, many endoscopists remain reluctant to do so. If more endoscopists are to consider carrying out gastroscopy without sedation, the question that needs to be asked is whether or not using hypnosis confers any advantages over endoscopy without sedation.
This was, indeed, a secondary question in Conlong and Rees' paper, and they do describe a significant reduction in the level of agitation, at least, of hypnotised patients.
However, I believe that their study fails to adequately assess hypnosis as it contains a number of methodological flaws, particularly as regards hypnotic technique.
The authors chose a singularly uninspired induction technique. In the time allowed, they could have introduced many more sensory modalities, thus increasing the likelihood of achieving a deeper level of trance.
They also failed to use patient-generated “special place” imagery, which has repeatedly been shown to be highly effective in reducing pain perception and anxiety during medical procedures.3
By using a single blind trial design, they ignored the contextual and relational aspect of hypnosis and left the endoscopist unable to reinforce and maintain trance throughout the procedure. (I believe that the confidence and trust instilled by the hypnotist-endoscopist contributes significantly to the depth of trance achieved by the patient.)
Finally—and most significantly in light of their conclusion—they failed to include any suggestions for amnesia at the end of the procedure. Hypnotic amnesia does not happen spontaneously, so it hardly seems reasonable to claim that hypnosis was ineffective at inducing amnesia when it was not even suggested. I therefore dispute their claim that sedated patients were “significantly more amnesic” because they did not, in fact, compare midazolam-induced amnesia with anything!
The disturbing thing about this paper is that, as people tend to read only abstracts (or even just “blipverts” in the back of theBMJ 4), the conclusion that has filtered down to the general audience (at least, if my training days in Leeds are at all representative) is that “hypnosis is no good for endoscopy”. And that, I am sure Conlong and Rees would agree, is an unfortunate—and erroneous—conclusion.