An acupuncture bait and switch on hypertension

Acupuncture is a theatrical placebo, but it's hard not to grudgingly admire just how—shall we say?—malleable or adaptable a placebo it is. What I mean by this is that, if you believe its practitioners and adherents, acupuncture can treat almost literally any disease or health problem. Any! Pain? Acupuncture. Allergies? Acupuncture. Biliary colic? Acupuncture. Infertility? Acupuncture. Menstrual problems? Acupuncture. Prostate problems? Acupuncture. Asthma? Acupuncture.

Well, maybe not asthma, but that doesn't stop acupuncturists from claiming that acupuncture can treat it.

Oh, and if you're a middle aged white dude with a touch of hypertension like me, a new study claims that acupuncture can help you, particularly if your antihypertensive medications are causing side effects. I've been fortunate enough that this is not the case for me (indeed, I've been amazed at how side effect-free my treatment has been), but, sadly, that's not the case for everyone. Maybe some of those people would find having needles stuck into them on a regular basis preferable to pharmacology, diet changes, and exercise.

First, let's take a look at the press release on EurekaAlert!. It's full of tropes about traditional Chinese medicine that drive me crazy:

Irvine, Calif., Aug. 19, 2015 -- Patients with hypertension treated with acupuncture experienced drops in their blood pressure that lasted up to a month and a half, researchers with the Susan Samueli Center for Integrative Medicine have found.

Their work is the first to scientifically confirm that this ancient Chinese practice is beneficial in treating mild to moderate hypertension, and it indicates that regular use could help people control their blood pressure and lessen their risk of stroke and heart disease.

"This clinical study is the culmination of more than a decade of bench research in this area," said Dr. John Longhurst, a University of California, Irvine cardiologist and former director of the Samueli Center. "By using Western scientific rigor to validate an ancient Eastern therapy, we feel we have integrated Chinese and Western medicine and provided a beneficial guideline for treating a disease that affects millions in the U.S."

Here we go again. Note the appeal to antiquity, as acupuncture is referred to as an "ancient Chinese practice" when in fact acupuncture as we know it today is far more a phenomenon that arose during the 20th century. It was popularized in China by Chairman Mao Zedong and brought to the West with the help of credulous reporters. Indeed, traditional Chinese medicine, of which acupuncture is a major part, was the very first example of a conscious attempt to "integrate" quackery with real scientific medicine. Back before, say, the 20th century, acupuncture resembled a practice common in the "West" as well as the "East," namely bloodletting—and a particularly crude and barbaric form of bloodletting. Indeed, until well into the 20th century, no attempt was made to insert needles near acupuncture points; needles were simply inserted near the point of pain.

And, of course, unless the ancient Chinese were considerably more advanced than we knew, there was no electricity, which makes Dr. Longhurst's boasting of "using Western scientific rigor to validate an ancient Eastern therapy" so utterly risible that reading his weasel words of woo (as Kimball Atwood liked to call them) makes my head feel as though it's about to explode. Doesn't he realize how silly he sounds? Does he realize how racist he sounds? As I've said so many times before, the "East" versus "West" dichotomy assumed by advocates of quackademic medicine with respect to many forms of alternative medicine that they seek to "integrate" into "Western" medicine in that it assumes that the "East" (i.e., Asians) are more "wholistic," touchy-feely, and mystical, while the "West" (i.e., those of us descended from Europeans) are cold, scientific, and reductionist. If I were Asian, I'd be insulted by the construct.

In examining any study of acupuncture, there are several things to look for. Indeed, there are several general categories of deficiencies that apply to nearly all of them, particularly the "positive" studies like this one:

  • Inadequate or no blinding
  • The "bait and switch" of using electroacupuncture instead of acupuncture
  • Small sample size
  • Improper or inadequate controls
  • Statistical problems
  • "Soft" subjective outcomes measured

There are others, but these are the most common.

So which of these apply to this study? Let's take a look.

The investigators studied 65 subjects. Subjects stopped taking their antihypertensive medications for at least 72 hours before the electroacupuncture (EA) treatments began, and during the study they underwent 24 hour ambulatory blood pressure monitoring. They were also instructed to start taking their antihypertensive medications if their average systolic BP went over 180 or their average diastolic BP went over 110. They noted that none of them had to do this, which tells me that they didn't have particularly severe hypertension.

Right here, though, is a problem with this study that shows just how ineffective institutional review boards are at evaluating the ethics of clinical trials of "complementary and alternative medicine" (CAM). Taking subjects with hypertensive off of their antihypertensive medications is not without risk. Sure, if they don't run blood pressures over 180 systolic or 110 diastolic, the risk is very low, but it is not nonexistent. Weigh this against a fantastical therapy with minimal chance of benefit, and if I were on the IRB for this study there'd be no way I would vote to approve it.

Study design for acupuncture (electroacupuncture) versus hypertension Study design for acupuncture (electroacupuncture) versus hypertension

On to the design, which is summarized in the figure above. This study, for some reason, randomized patients into two different acupuncture groups, one receiving 30-minutes of EA at acupuncture points PC 5-6 + ST 36-37 or at LI 6-7 + GB 37-39 once weekly for 8 weeks. How these were decided upon appears to be based on animal experiments. (Lovely.) The PC 5-6 acupuncture points iares located above proximal to the wrist between the tendons of the palmaris longus and flexor carpi radialis muscles. The ST 36-37 acupuncture points have been discussed here before, being located on the leg lateral to the anterior border of the tibia. The LI 6-7 acupuncture points are on the thumb side of the arm above the wrist, while the GB 37-39 acupuncture points are on the lateral part of the lower leg, above the ankle.

Basically, the 33 patients receiving EA at PC 5-6 and ST 36-37 were the experimental group, while the rest were the controls. This is better than the average acupuncture study in that at least the investigators tried to evaluate "irrelevant" points (based on what, who knows) as controls. Unfortunately, certain other controls are notably absent, namely a sham EA control for PC 5-6, which could be accomplished quite simply by sticking the needles in, but not actually applying electricity, the machine being set up so that the practitioner doesn't know whether current is being applied or not.

Speaking of blinding, another big problem with this study becomes rapidly evident when you look at the methods. The study was only single blinded (the patients). True, the data analysis was also blinded, but practitioners were not blinded. Also, there was no post-study assessment of blinding to see how effectively patients were blinded to their study group. At the very least, post-study assessment of subjects to make sure that their assessment of which group they were in is consistent with random guesses, suggesting that they didn't know what group they were in. If you really wanted to make sure the patients were blinded, you'd stick the needles in all four acupoints and then only two of them would receive current.

So what were the reported results? After eight weeks, the experimental group showed decreased peak and average SBP in 70% of the participants, an average of 6 to 8 mmHg for systolic blood pressure and 4 mmHg for diastolic blood pressure. These improvements reportedly persisted for a month and a half. There was no detectable change in either systolic or diastolic blood pressure in the control group. Also, reportedly, After treatment, the plasma concentration of norepinephrine (which increases blood pressure), which was initially elevated, declined 41%. Similarly, the renin-angiotensin system activity, which increases blood pressure (indeed was reduced, with renin decreased by 67% and aldosterone by 22%. Blood pressure is commonly treated with angiotensin converting enzyme (ACE) inhibitors like lisinopril, which lower the activity of the same system.

The reported decreases in blood pressure were small, with peak and average systolic BP decreasing by 8 and 6 mmHG, respectively, and diastolic BP decreased by 4 mmHg. The authors claim that this is clinically significant. I suppose that might be possible, but only if a patient has some incredibly mild hypertension, so that such a small reduction would bring blood pressure down to target levels as described in clinical practice guidelines.

So let's assume for the moment that the findings are "correct" (in other words, EA really does reduce systolic BP by around 8 mmHg and diastolic BP by 4 mmHg). This requires weekly treatments. Hypertension is a chronic condition. Are people really going to undergo EA treatments at least weekly for the rest of their lives, hundreds and hundreds of treatments? Is that really better than taking a pill. The assumption is that it is better, but I'm not so sure, particularly given that the longer one regularly sticks needles into one's skin the greater the chance that one of those uncommon acupuncture complications will occur.

Overall, this study is just not that convincing. It has small numbers, inadequate controls, and, of course, the classic bait and switch substituting EA for acupuncture. Even if its results are totally valid, it does not validate acupuncture, but rather transcutaneous electrical stimulation (TENS), which is a science-based treatment for some conditions. In the end, acupuncture is a modality that is touted as being able to treat everything. In reality, it is based on a prescientific understanding of how the human body works and requires a lot of "help" in the form of electricity even to give the appearance of working.

Basically, acupuncture is to CAM as words were to Humpty-Dumpty. When acupuncturists use the word "acupuncture" it means what they choose it to mean, neither more nor less, and the most important thing is to be master of what it means. Unfortunately, the very malleability of acupuncture is what makes it so unconvincing as a treatment. A treatment that is claimed to be able to treat anything in the end probably treats nothing effectively.


More like this

At the risk of repeating myself (but, then, since when did such concerns ever stop me before?), I'll just start out by mentioning that, of all the non-herbal "alternative" medicine remedies out there, I used to give a bit of a pass to acupuncture. No, I never did buy any of that nonsense about how…
Every so often, our "friends" on the other side of the science aisle (i.e., the supporters of "complementary and alternative medicine"—otherwise known as CAM or "integrative medicine") give me a present when I'm looking for a topic for my weekly bit of brain droppings about medicine, science, and/…
Back when I started this blog, I hadn't yet become aware of the phenomenon known as quackademic medicine. This phenomenon, as you recall, is the infiltration of academic medical institutions that should be bastions of science- and evidence-based medicine by outright quackery. In quackademic…
It's always disappointing to see a good journal fall for bad medicine, particularly when it's in your field. For example, the Journal of Clinical Oncology (affectionately referred to by its abbreviation JCO) is the official journal of the American Society of Clinical Oncology (ASCO) and probably…

After eight weeks, the experimental group showed decreased peak and average SBP in 70% of the participants, an average of 6 to 8 mmHg for systolic blood pressure and 4 mmHg for diastolic blood pressure.

Not that I am an expert, but why wouldn't a variation of this magnitude be within normal variations for somebody with a moderately active lifestyle? It would not surprise me if blood pressure readings were substantially higher immediately after exercise, because during exercise you have greater need to transport oxygen to the muscles in question. (It would also partially explain the phenomenon of people having heart attacks while jogging.)

By Eric Lund (not verified) on 20 Aug 2015 #permalink

Just what exactly was their "decade" of bench research?

Longhurst (the last author on this "eureka" studay) published a paper in 2013 titled "Electroacupuncture modulation of reflex hypertension in rats: role of cholecystokinin octapeptide.". This off-the-deep-end cardiologist so wants acupuncture to work that, as you note, Orac, he somehow thinks putting a voltage to that inserted acupuncture needle and passing current is somehow the same as just inserting the needle, which it is most clearly not. For comparison, selective electrical vagal nerve stimulation can reduce blood pressure as well ( it is only the nerve stimulation, and nothing else that is important for the blood pressure change. The problem, unfortunately, is trying to only stimulate the right nerve fibers in the vagal nerve and not the whole nerve makes this not a technique ready for prime time. .

I don't get why they were allowed to call this acupuncture. It isn't. What they did is electrophysiology, and not very intelligent electrophysiology at that. Researchers who are doing real electrophysiology research ought to be pissed off at crap like this.

By Chris Hickie (not verified) on 20 Aug 2015 #permalink

As to the question of do the study authors really think patients will be able to substitute acupuncture for pills, I may have an answer to that, though I certainly don't find it better than medication given the limited effects of the treatment (at best IMHO).

You can buy home accupuncture kits. I expect the woo meisters will encourage the sale of said kits, classes to "teach" people how to "self-administer" accupuncture, and less frequent "tune up" visits with accupuncturists so nobody loses a job.

You can buy home accupuncture kits.

I shouldn't be surprised by this, even though acupuncture is one of those things that I'd put in the category of "Don't Try This At Home, Kids!" Does your average home acupuncturist know how to sterilize the needles properly? Will they sterilize the needles properly after each use? Will people share acupuncture needles? (Yes, I'm getting into serious squick territory here.)

Given the hysteria in the US about providing clean IV needles to drug addicts, I wouldn't be surprised if some enterprising DA used anti-drug laws to shut this thing down, at least in states where such laws are sufficiently broad.

By Eric Lund (not verified) on 20 Aug 2015 #permalink

Longhurst et al. found a slight benefit for hypertensive patients after administering a mild subcutaneous electric shock (via acupuncture needles) at certain acupuncture points.
What do real acupuncture studies show?
They show degree of continuity.
Flachskampf et al. 2007. Adjunct A better than sham, but no difference between A and sham for non-adjunct treatment groups.
The SHARP study, 2006. No difference between non-adjunct A and sham.
Dong-Ze Li et al. 2014. Meta-analysis of acupuncture - no electricity involved. Adjunct A significantly lowers blood pressure; A alone does not.
Zhao XF et al. 2015. Systematic Review and Meta-analysis. Adjunct A works, insufficient evidence for A alone. They warn that the reliability of current evidence for A overall is questionable and that larger high quality trials are needed.
This evidence is pretty clear. Acupuncture is of no benefit for hypertension. Whether a mild electric shock delivered via acupuncture needles is superior to placebo has yet to be demonstrated.
What the evidence of Longhurst et al. does not support is the claim by University of California that the "ancient Chinese practice" of acupuncture has been "scientifically confirmed" to work for hypertension.
That’s simply hyper-nonsense on stilts.

By Leigh Jackson (not verified) on 20 Aug 2015 #permalink

Doesn't figure 1 show that they included people with systolic levels as low as 123mm/Hg and diastolic as low as 71mm/Hg. This does not seem very hypertensive.

By peter mcdonnell (not verified) on 20 Aug 2015 #permalink

I found another Longhurst paper published earlier this year:

Integrating acupuncture into the cardiology clinic: can it play a role?

In the second paragraph Xiaorui Zhang's notorious 2002 report for the WHO is quoted as evidence of acupuncture's proven benefits.

Anyone who quotes that report has zero credibility since it systematically excluded sham controlled studies which showed no superiority for acupuncture over placebo as necessarily being wrong since other studies had shown them to be superior!

"In many published placebo-controlled trials, sham acupuncture was carried out by needling at incorrect, theoretically irrelevant sites. Such a control really only offers information about the most effective sites of needling, not about the specific effects of acupuncture. Positive results from such trials, which revealed that genuine acupuncture is superior to sham acupuncture with statistical significance, provide evidence showing the effectiveness of acupuncture treatment. On the other hand, negative results from such trials, in which both the genuine and sham acupuncture showed considerable therapeutic effects with no significant difference between them, can hardly be taken as evidence negating the effectiveness of acupuncture. In the latter case, especially in treatment of pain, most authors could only draw the conclusion that additional control studies were needed. Therefore, these reports are generally not included in this review."

The report was biased before it was even written.

By Leigh Jackson (not verified) on 20 Aug 2015 #permalink

@ Chris#2

Found a BBC link to a 2005 rat study. Note that a photo of someone getting acupuncture is placed alongside the statement "acupuncture combined with electrical stimulation can lower high blood pressure". Acupuncture on man, electro-stimulation on rats. Message: acupuncture works.

Credit to Mike Cummings of the British Medical Acupuncture Society for advising caution. Credit to him for also slamming Xiarorui Zhang's report for the WHO as being hopelessly biased, worthless junk.

By Leigh Jackson (not verified) on 20 Aug 2015 #permalink

Maybe they just knew how to take blood pressure readings correctly during this study.

I have Stage 1 hypertension that was poorly controlled with lisinopril, or a betablocker or HCTZ (all while on CPAP therapy). But recently, my miniscule dose of 12.5 mg HCTZ has brought my readings into the normal range.

The only thing that has changed is the office staff/nurses are now letting me sit for a few minutes in the chair where taking it and they are providing support for my arm when they take it. No more of me holding it out straight for them.

It has proven to be a miracle cure.

By Not a Troll (not verified) on 20 Aug 2015 #permalink

using Western scientific rigor to validate an ancient Eastern therapy

When you know in advance what the answer is to be, when you regard research as a tool of advertising, that's when a double-blind design would be a really good idea.

By herr doktor bimler (not verified) on 20 Aug 2015 #permalink

herr doktor bimler @ 11: Ha, you beat me to it. I caught that item about "validate" too. "East/West" talk might have been OK in the 1970s when scientists were first starting to study Hindu and Buddhist meditation practices, but the word "Negro" was also OK in the 1960s.


Home acupuncture kits: Good God!, and let's have dozens or hundreds of people end up in the ER with various forms of subcutaneous infections, some of which may involve antibiotic-resistant bacteria. Not to mention the various "sharing needles" risks if more than one person in a household is using them. How is it even legal to sell those kits?


Simple decision tree:

If the choice is between pills and needles, take the pills.

If the choice is between needles and dangerous contagious diseases, take the needles. Different type of needles though, and thankfully it usually means only one;-)

If one of the choices is magical hand-waving, take the other choice.

If someone says "energy" other than as a synonym for alertness or arousal ("I have a lot of energy today") run like hell in the other direction.

By Gray Squirrel (not verified) on 20 Aug 2015 #permalink

A cousin was recently admitted to hospital after acupuncture treatment. Diagnosis: punctured lung. The needles are so thin this can apparently happen quite easily.

Cousin gained the impression that the medical staff were not particularly surprised. I wonder whether they had seen this before?

Home acupuncture! Imagine the over-zealous deep acupuncturist.

Thank the Good Lord for skepticism. ?

@ MadisonMD #14

Thanks! Overdone, and/but I was helpless with laughter.

'Acupuncture is based on the premise that if you stick pins in somebody who complains, they will stop complaining when you stop sticking pins in them'

Paraphrase of Mike Royko on Acupuncture.

Home acupuncture kits may actually solve people's willingness to fall for this crap. A few cases of infections will do - because I don't believe that an average citizen would take proper care of the needles, even less so a woo follower who may or may not believe in pathogens, sterilization and all this oppressive Western scientific dogma.

By kultakutri (not verified) on 22 Aug 2015 #permalink

I have patients protest the "inconvenience" of taking a daily pill, or having to come in and see me a few times a year. Are those same people actually going to go see an acupuncturist weekly?

Hypertension is notoriously variable, and "white coat hypertension" (it reads as high just because of the anxiety of a doctor/nurse checking it in clinic) is a well known phenomena. The numbers they show for "improvement", seem well within the normal variability of blood pressure. Also, even though the patients were "blinded", they still knew they were in a "study" of some sort, and the Hawthorne Effect alone probably could account for any improvement. (

By NewcoasterMD (not verified) on 23 Aug 2015 #permalink