Hot flushes 2019

Inspired by Lund et al BMJ Open 2019.[1]
Photo by Ryan Moreno on Unsplash
– representing waking up soaking wet as some of my patients describe…

This paper hit the ground running, so to speak. I guess it must have been press released by BMJ, because it ended up on the BBC News website late on Saturday 9th March 2019. It seemed to get quite a bit of attention and surprisingly little criticism from the sceptics. The latter may well have been because ‘it could all be a placebo effect’ was prominent, and links to BBC items on the placebo effect were added. The news item included the immortal phrase “We can’t explain the underlying mechanism behind acupuncture…”,

…which prompted a Trumpesque tweet from me ;-).

On reflection I may have been a little hasty in my response.

The study had a pragmatic design, very like the ARC (Acupuncture in Routine Care) trials from the Modellvorhaben Akupunktur. Patients were randomised to a short course (5 weekly sessions) of standardised acupuncture (CV3, CV4, LR8, SP6 and SP9) or a six week wait for their acupuncture. The sessions were performed by a local GPs trained in acupuncture (9 Danish primary care practices were involved). The ARC studies had a 3-month treatment cycle with 12 individualised sessions or a 3 month wait for the same intervention. The other big difference from ARC was the size of this trial. This one included just 70 women, whereas the largest ARC trial included over 15k patients – 200 times bigger! Although to be fair only about 3k agreed to randomisation – the others made up a large prospective cohort. With this size in mind it is remarkable how much press attention the study achieved.

Another interesting factor in this report is the prominent use of the term Western medical acupuncture. Here we see the influence of a former BMAS stalwart Palle Rosted, who has influenced training and education in Denmark over the last 15-20 years. He was always a firm advocate of local and segmental needling, as well as having a special interest in the use of acupuncture in dermatology and dentistry.

But the authors got past peer review and editorial oversight without justifying why their approach was Western medical in style. I am not complaining, but we are keen to promote such mechanistic justifications in clinical acupuncture research questions. I believe it is unsafe to ask a research question without considering the potential mechanisms involved otherwise we risk missing the point![2]

This group did not miss the point though, as this was not a sham controlled study like the one from Annals in 2016 that was the subject of my first blog.[3] The Annals study failed to show a difference between real acupuncture and a non-penetrating sham, and they went on to find that expectancy after the first treatment did not predict response.[4] This might argue against the effect of acupuncture in hot flushes being entirely a placebo effect.

But now I am left with the task of explaining how acupuncture might work in the treatment of hot flushes, in keeping with the implication of my Trumpesque tweet. I cannot simply deny I tweeted it!

I had thought the concept was quite well understood. A natural decline in circulating oestrogens leads to a heightened sensitivity in the thermoregulatory centre of the hypothalamus. Acupuncture results in a rise in central endorphin release and consequent stabilization of the centre. A little simplistic I’m afraid, since I have subsequently discovered that we do not understand the mechanism of the hot flushes in the first place![5]

We do not understand the mechanism of hot flushes.

Our old friend CGRP seems to be strongly implicated in hot flushes,[6–8] and acupuncture seems to affect circulating levels of this neuropeptide,[9] at least in the cerebrospinal fluid of rats, as well as immunoreactivity in different parts of their brains. One idea is that falling levels of oestrogens result in generally lower levels of circulating CGRP and thus upregulation of CGRP receptors. Areas with a high density of these receptors, such as the thermoregulatory centre in the hypothalamus (of females) and blood vessels in skin, then amplify the response to this neuropeptide.

The medial preoptic area of the hypothalamus, involved in thermoregulation, contains over 25-fold more CGRP-immunoreactive cells in female rodents compared with male rodents.[6]

Well there are quite a few loose ends to be tied up here, but perhaps I can compromise with, “We don’t know exactly how it works, but we have a few pretty plausible theories – they just need to be properly pinned down” ;-).


1         Lund KS, Siersma V, Brodersen J, et al. Efficacy of a standardised acupuncture approach for women with bothersome menopausal symptoms: a pragmatic randomised study in primary care (the ACOM study). BMJ Open 2019;9:e023637. doi:10.1136/bmjopen-2018-023637

2         Cummings M. Commentary: Controls for acupuncture – can we finally see the light? BMJ 2001;322:1578.

3         Ee C, Xue C, Chondros P, et al. Acupuncture for Menopausal Hot Flashes: A Randomized Trial. Ann Intern Med 2016;164:146–54. doi:10.7326/M15-1380

4         Ee CC, Thuraisingam S, Pirotta M V, et al. Expectancy after the first treatment and response to acupuncture for menopausal hot flashes. PLoS One 2017;12:e0186966. doi:10.1371/journal.pone.0186966

5         Andrikoula M, Prelevic G. Menopausal hot flushes revisited. Climacteric 2009;12:3–15. doi:10.1080/13697130802556296

6         Oliveira MA, Lima WG, Schettini DA, et al. Is calcitonin gene-related peptide a modulator of menopausal vasomotor symptoms? Endocrine 2019;63:193–203. doi:10.1007/s12020-018-1777-z

7         Wilhelms DB, Dock H, Brito HO, et al. CGRP Is Critical for Hot Flushes in Ovariectomized Mice. Front Pharmacol 2018;9:1452. doi:10.3389/fphar.2018.01452

8         Hay DL, Poyner DR. Calcitonin gene-related peptide, adrenomedullin and flushing. Maturitas 2009;64:104–8. doi:10.1016/j.maturitas.2009.08.011

9         Wyon Y, Hammar M, Theodorsson E, et al. Effects of physical activity and acupuncture on calcitonin gene-related peptide immunoreactivity in different parts of the rat brain and in cerebrospinal fluid, serum and urine. Acta Physiol Scand 1998;162:517–22. doi:10.1046/j.1365-201X.1998.0317e.x

Declaration of interests MC