Feb 19 2015

Phantom Acupuncture

There are two basic schools of thought when it comes to acupuncture, which is the practice of placing thin needles into alleged acupuncture points in order to have a therapeutic or symptomatic effect. The “traditional” interpretation is that the needles are stimulating a physiological response of some kind at the acupuncture points. Within this school there is a range of opinions as to whether this response is due to a biochemical, neurological, or another known biological response or whether it is due to the still more traditional (but actually less than a century old) belief that the needles are manipulating the life force or Qi.

The other school holds that acupuncture is essentially an elaborate placebo. (Note Рthis article contains all the references necessary to support my statements below, so I will not repeat them.) Any apparent response is a non-specific response to the attention of the practitioner, expectation, distraction from pain, simple regression to the mean, and other illusory effects.

Each school makes different predictions about the various lines of evidence that can be brought to bear to resolve this question. There have been in total several thousand clinical studies looking at the apparent effects of acupuncture. These have failed to convincingly reject the null hypothesis, meaning that they have not demonstrated a clear biological response to acupuncture for any indication. The better controlled studies consistently show that needle location does not matter (sham acupuncture), and that needle insertion does not matter (placebo acupuncture). You can literally have a non-acupuncturist randomly poke someone with toothpicks and get the same response as the full acupuncture treatment.

The other line of evidence regarding acupuncture are studies that look at physiological responses to having needles poked through the skin. These studies, completely unsurprisingly, show that “stuff happens” when you stick a needle through the skin. There is a local reaction to the trauma, and the brain reacts in a predictable way. I do not think these studies in any way distinguish between the two interpretations of acupuncture. Even if acupuncture is nothing but a placebo, we would still expect these types of responses to sticking needles into tissue.

One way in which these studies might differentiate the two interpretations is if it could be convincingly shown that acupuncture points are real – the body responds in a consistently and functionally different way when acupuncture points are jabbed than when non-acupuncture points are jabbed. However, the evidence does not support this conclusion. In fact the totality of the evidence strongly supports the conclusion that acupuncture points have no basis in reality – they don’t exist.

There is now a new line of evidence that is very interesting, and one that I had not previously considered – phantom acupuncture. A recent study looked at performing acupuncture on a phantom limb. They used a now well-established technique of tricking the brain into incorporating a dummy body part as if it were real. They placed subjects in front of a table so that one of their arms was below the table, with a rubber arm above the table placed in such a way that visually the rubber arm looked like their own arm. They then stroke the rubber arm and the subjects real arm simultaneously. The brain sees and feels the rubber arm being stroked, and this sensory feedback is often enough for the brain to create the sensation of ownership over the rubber arm.

The researchers then placed acupuncture needles into the rubber arm that subjects had incorporated as their own. Obviously there is no possibility of any physiological response from the needle penetrating the rubber arm. I further think it is reasonable to conclude that placing a needle into a rubber arm cannot activate acupuncture points (if they existed) or alter Qi (if it existed). This experiment nicely eliminates local physiological responses and any Qi responses to the needles.

The researchers performed functional MRI scanning (fMRI) on subjects while needles were placed in their phantom rubber limbs. Keep in mind that fMRI research involves collecting lots of data and aggregating it. So the researchers are not looking at brain reactions in real time, just the aggregate brain activity of many subjects over many trials. They then look for statistical associations in the activity.

What they found was the same brain activation that previous studies have found with acupuncture of real limbs.

When the rubber hand was fully incorporated with the real body, acupuncture stimulation to the rubber hand resulted in the experience of the DeQi sensation as well as brain activations in the dorsolateral prefrontal cortex (DLPFC), insula, secondary somatosensory cortex (SII), and medial temporal (MT) visual area. The insular activation was associated with the DeQi sensation from the rubber hand.

If these results hold up, this implies that the brain is simply responding to the expectation and visualization of the needle penetration. Actual needle penetration is unnecessary. The most parsimonious interpretation of this data is that acupuncture is all in the mind. There is no need to hypothesize the existence of Qi, acupuncture points, or a specific physiological mechanism for acupuncture.

There are plenty of studies that show that the perception of pain is easily manipulated by simple things such as distraction. Swearing, distorting body image, and crossing your arms while one of them is pricked will all reduce pain perception. Pain perception is closely tied to attention, and so simple distraction is effective. It’s no surprise, therefore, that the brain responds to phantom stimulation.

At the very least this study demonstrates that all prior studies looking at fMRI responses to acupuncture needle insertion were likely simply showing a non-specific brain response to the expectation and/or visualization of needle insertion, without the need to invoke any specific physiological responses.


The totality of evidence strongly indicates that there is nothing specific to acupuncture. Acupuncture points don’t exist, Qi does not exist, and the elaborate details of acupuncture treatment do not matter. In other words – acupuncture is an elaborate (and unnecessarily so) placebo.

We can now add phantom acupuncture to sham and placebo acupuncture as lines of evidence demonstrating that acupuncture is no more than a placebo. The researchers indicate their desire to take the next step – to see if there is a clinical response to phantom acupuncture. The placebo hypothesis predicts that there should be, at least to some extent, depending on proper blinding (if possible).

What all of this means is that the very concept of acupuncture adds nothing to our understanding of the universe, and biology and medicine specifically. It is a failed concept. We lose nothing by discarding it. Any part of acupuncture that “works” is mere placebo and therefore not specific to acupuncture. Anything specific to acupuncture does not work.

Any non-specific symptomatic benefits to acupuncture (which do not appear to be clinically significant, by the way) can be achieved without sticking needles through the skin (which entails some risk), does not require special training, and does not require an expensive elaborate procedure. Further, persisting in the myth of acupuncture fosters misunderstanding of science, biology, and medicine which has insidiously negative effects.

Acupuncture itself is a phantom phenomenon and should go the way of the ether and N-rays.

194 responses so far