Skip navigation

Acupuncture for surgery

Clinical bottom line

This review tells us that we can be sure that no good evidence that acupuncture is useful during surgery has yet been presented.


H Lee, E Ernst. Acupuncture analgesia during surgery: a systematic review. Pain 2005 114: 511-517.

Systematic review

Randomised studies involving people undergoing surgical procedures were included. Acupuncture meant needle insertion. Outcomes were related to pain during surgery, or those supporting anaesthesia during surgery.

A number of different general and specialist databases were searched for trials. Positive studies were defined as acupuncture significantly more effective than control, neutral when there was no significant difference, and negative when acupuncture was significantly less effective than control, using a p value of 0.05.


Nineteen trials were found, and the overall results are shown in Table 1. There were more positive results for acupuncture in low quality trials, compared with high quality trials. Only four of 15 trials comparing acupuncture with conventional anaesthesia had high validity scores, where bias was unlikely. Only four trials used a sham-acupuncture control.

Eleven studies were published in English, five in Chinese, and one each in Croatian, French, and German. Statistical significance was highly dependent on the origin of the study, with 63% of non-English articles being significant, compared with 18% of those published in English.

Table 1: Overall conclusions of trials according to trial quality

Trial quality
High quality
Low quality
All trials

Particular outcome measures were infrequently reported. Outcomes reported in at least three trials included intraoperative fentanyl consumption in four trials with 108 patients, showing a slight reduction of fentanyl consumption with acupuncture, though the most recent, and largest, trial showed no difference.

A conclusion that the number of patients with successful analgesia as judged by the anaesthetist was higher with acupuncture than without acupuncture overall depended on the result of a single study (Figure 1). This Chinese study depended on its statistical significance on only 40% of patients having successful anaesthesia without acupuncture, and was the only one of the six studies to demonstrate statistical significance.

Figure 1: Successful anaesthesia with and without acupuncture


This is a good review. If it lacks one perspective that may be important, it was the observation that trials in some parts of the world are always positive (Bandolier 71). This explains the discrepancy that most positive trials were published in languages other than English. Given also that acupuncture was not blinded in most of the studies, means that there is a significant risk of bias.

On the basis of what has been published, we can be sure that no good evidence that acupuncture is useful during surgery has yet been presented. It is perhaps just too early to say that it definitely is not useful, but that point is close.