Phantom acupuncture for back pain

Stimulated by Lee et al 2019 & Wei et al 2019.[1,2]

Photo by Brad Lloyd on Unsplash.

Back pain was a regular topic on this blog at its inception in March 2016 – around the time that the draft of NG59 was published. NG59 is the current NICE guideline for back pain and sciatica, although it seems somewhat out of step with other guidelines in the Western world,[3] and one of the key interventions recommended has been shown to be entirely ineffective in practice.[4]

I have found myself remarking ironically in lectures since that “…of course acupuncture doesn’t work for back pain!” The data on acupuncture efficacy in NG59 had errors in both draft and final versions, but in spite of this it still exceeded many of the subsequently recommended interventions. See the discussion in previous NICE blogs.

The papers I am focussing on here represent two very different approaches in back pain research that have been published this month. The relation between them is simply the back pain – a label with such potential for clinical heterogeneity that it is rather tenuous to make any link. But I was running out of time to make my choice, and both have interesting features… and of course this is just a blog!

The first is an fMRI study examining brain connectivity in patients with back pain who received either real acupuncture (to leg points) or so called ‘phantom acupuncture’.

Phantom acupuncture

So, what on earth is phantom acupuncture? Well it is fairly rare on PubMed with only 2 hits for phantom acupuncture [ti] (in the title text) and 4 for phantom acupuncture [tiab] (in the title or abstract text).[1,5–7]

It is quite a clever technique that popped up in 2014.[5] Basically the subject watches a video screen as the acupuncturist pretends to perform acupuncture, but instead of seeing the image in real time, they are shown a video of an acupuncture needle being inserted and manipulated in the same place. The video may be when the procedure was performed on the same subject previously, or it may be from a video on another subject. So, the subject only gets the visual cue relevant to the context of acupuncture treatment, and no direct somatosensory stimulation, not even the acupuncturists touch. I guess the reason for the latter is that this technique was developed for fMRI studies, and touch alone can light up the somatosensory cortex and other areas downstream.

The recent study that I am highlighting is the third in a sequence by the same group.[1,5,7] First they tested out the phantom acupuncture procedure and measured automonic and psychophysical responses compared with those associated with real acupuncture. Then they studied the fMRI changes in patients with back pain undergoing real or phantom acupuncture. Finally, in this latest study they have looked functional connectivity between brain regions already identified as relevant in patients with back pain.

Now the acupuncture was performed at three points: ST36; SP11; SP13. Why not needle the back I immediately thought. Well in order to do the fMRI, you need the patient to be lying with their head in a tunnel with a huge magnetic, and I have only ever seen this done with subjects face up. Secondly, they had the subjects lying on an inflatable cushion place under the most painful part of the back, and they inflated the cushion to extend the lumbar spine until the back pain reached a certain level of intensity. With all that going on there is not much space or opportunity to needle the back!

SP13 is an odd one though. It is pretty much directly over the deep inguinal ring, so not a great place for needling over the abdominal wall. Having said that the combination of points do cover all the ventral segments of the lumbar spine (L1-5).

I am going to skip the explanation of how you measure functional connectivity between different regions of interest (ROIs) in the brain, but I will mention the networks of interest.

DMN – default mode network

SMN – sensorimotor network

SN – salience network

The DMN is the network that is active when your mind is wandering ie you are not focussed on a specific task. For the SMN is clue is in the name, and the SN is a bit of a mystery, but it is thought to mediate a switch from the DMN to the central executive network ie to snap you out of your daydreaming to focus on a task, presumably triggered by a salient stimulus.

Finally, what did Lee et al find in the brain connections of their subjects lying uncomfortably in an MRI scanner and being subjected to either real of phantom acupuncture stimulation?

Well both groups (real and phantom) had a significant reduction in subjective pain, without there being any significant difference. Of course, the study was way too small (n=43) to detect any difference in subjective pain.

In the group that received real acupuncture, the pain reduction correlated with a reduction in the connectivity between the medial prefrontal cortex (mPFC) and the posterior insula (pINS), whereas in the phantom acupuncture group the decrease in pain correlated with and increase connectivity between the anterior insula (aINS) and the posterior cingulate cortex (PCC).

So, what does that mean?

Well the real acupuncture seems to have resulted in a detachment from the pain, with a reduction in the effect of the pINS (where deep somatic sensations are processed) on the mPFC (responsible for self-referential processing). Perhaps this indicates more of a focus on the external environment created by the stimulation of acupuncture needles and away from the internal environment where the self is ruminating on its chronic back pain. By contrast the phantom acupuncture mediated pain reduction associated with increased communication between the aINS (a key part of the SN) and the PCC (a key part of the DMN). Neither of these regions responds directly to the physical aspects of stimulation but rather to the cognitive emotional aspects.

All this is a bit rarefied isn’t it! So, why not move out of the lab and into the real world of treating back pain? That brings me to the next paper I want to highlight. It is a protocol for a huge pragmatic observational study underway in China. They hope to include over 2000 patients, record their outcomes, and then through regression analysis determine whether or not any particular methods are more or less successful. Currently they plan to compare:

Traditional Chinese acupuncture style vs Microacupuncture style (Scalp acupuncture, Ear acupuncture, Wrist and ankle acupuncture, Eye acupuncture, Umbilical acupuncture, Abdomen acupuncture), Local acupoint selection vs Non-local acupoint selection and Single Acupuncture vs Combined therapy.


I was excited to see the 2000 figure in the abstract, but now I am starting to worry that they may not find any difference between any forms of needling, and then where will we be? Plus, where is the electroacupuncture? Unfortunately, it only pops up in the reference list.

Oh well, time will tell, and meanwhile I am off for a walk to try to get the image of ghostly acupuncturists covered in white sheets out of my mind before Halloween is upon us.


1         Lee J, Eun S, Kim J, et al. Differential Influence of Acupuncture Somatosensory and Cognitive/Affective Components on Functional Brain Connectivity and Pain Reduction During Low Back Pain State. Front Neurosci 2019;13:1062. doi:10.3389/fnins.2019.01062

2         Wei X, Liu B, He L, et al. Acupuncture therapy for chronic low back pain: protocol of a prospective, multi-center, registry study. BMC Musculoskelet Disord 2019;20:488. doi:10.1186/s12891-019-2894-4

3         Qaseem A, Wilt TJ, McLean RM, et al. Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians. Ann Intern Med Published Online First: 14 February 2017. doi:10.7326/M16-2367

4         Juch JNS, Maas ET, Ostelo RWJG, et al. Effect of Radiofrequency Denervation on Pain Intensity Among Patients With Chronic Low Back Pain: The Mint Randomized Clinical Trials. JAMA 2017;318:68–81. doi:10.1001/jama.2017.7918

5         Lee J, Napadow V, Kim J, et al. Phantom acupuncture: dissociating somatosensory and cognitive/affective components of acupuncture stimulation with a novel form of placebo acupuncture. PLoS One 2014;9:e104582. doi:10.1371/journal.pone.0104582

6         Leem J, Park J, Han G, et al. Evaluating validity of various acupuncture device types: a random sequence clinical trial. BMC Complement Altern Med 2016;16:43. doi:10.1186/s12906-016-1026-z

7         Makary MM, Lee J, Lee E, et al. Phantom Acupuncture Induces Placebo Credibility and Vicarious Sensations: A Parallel fMRI Study of Low Back Pain Patients. Sci Rep 2018;8:930. doi:10.1038/s41598-017-18870-1

Declaration of interests MC