Acupuncture is often promoted as a totally safe therapy. This may be good advertising but, as we have so often discussed on this blog, it is not true. A recent paper evaluated 17 systematic reviews on this subject. Four major categories of adverse effects (AEs) were identified:
- organ or tissue injuries (13 reviews, median: 36 cases, median deaths: 4),
- infections (11 reviews, median: 17 cases, median deaths: 0.5),
- local AEs or reactions (12 reviews, median: 8.5 cases, no deaths were reported),
- other complications such as dizziness or syncope (11 reviews, median: 21 cases, no deaths were reported).
The incidence of any AEs per patient was reported to be 42.4% with traditional acupuncture, 40.7% with minimal acupuncture and 16.7% with non-invasive sham acupuncture. But, as there is no AE-monitoring system, the frequency of serious AEs is essentially unknown, and more AEs are being reported regularly.
Here are two further case reports that are too recent to be included in any review.
Greek doctors reported a case of acupuncture-induced haemothorax in a 37-year-old healthy woman who had acupuncture treatments for neck and right upper thoracic non-specific musculoskeletal pain. Subsequently, she developed dyspnoea and chest discomfort. After some delay while trying other treatment, she was eventually transferred to the emergency department where a chest X-ray revealed a right pneumothorax and fluid collection. She was admitted to hospital and a chest tube inserted into the right hemithorax drained 800 mL of bloody fluid (haematocrit (Hct) 17.8%) in 24 h and 1200 mL over the following 3 days. Her blood Hct fell from 39.0% to 30.8% and haemoglobin from 12.7 to 10.3 g/dL. The patient recovered completely and was discharged after 9 days of hospitalisation.
Another case report told the story of a patient who had received the acupuncture treatment from a practitioner licensed in acupuncture, while on an international business trip. Following the acupuncture treatment, the practitioner realized that a needle had fractured and remained in the patient’s neck. After failing to retrieve the needle, the patient was sent for imaging. Radiograph revealed that the patient had two needle fragments located in his cervical tissue. After determining that the needles did not pose an immediate threat, the patient boarded a flight home to the United States. Following his flight, the patient presented to an American hospital where it was discovered that the needle fragments had migrated during the flight, with one needle now located 2 mm from the patient’s vertebral artery. Surgical intervention was required to retrieve the needles, resulting in the patient needing physical therapy to increase cervical range of motion and mediate pain relief. The authors concluded that clinicians utilizing single-use filiform needles in their practice, whether for acupuncture or dry needling, should be aware of the potential for this type of adverse event. Further, to minimize the risk of similar adverse events occurring in the future, clinicians should make sure that they are using high quality needles and make a habit of counting in and counting out the needles that they use to verify that all needles are accounted for.
I think this is sound advice for clinicians. To patients, I would furthermore recommend the following:
- Ask yourself whether you really need acupuncture. Once you have done some research, you might find that, for your condition, there are more effective treatments.
- If you want acupuncture nevertheless, consult a practitioner who is competent not just in the process of needling but in recognising AEs, diagnostics, contra-indications, etc. In my book, this is someone who has studied medicine, ie a doctor.
- Ask the clinician of your choice to provide you with the full information relating to the following questions: How many treatments will you need? What will this cost? What are the chances of success? What other therapeutic options are there? What are the risks?
- If the clinician cannot provide you with satisfactory answers, don’t let him/her treat you.
13 Responses to Acupuncture: much more than meets the eye!
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
Thanks Edzard, your advice section is also perfect to use for patients attending their medical practitioner:
I think this is sound advice for clinicians. To patients, I would furthermore recommend the following:
Ask yourself whether you really need that drug or surgical process. Once you have done some research, you might find that, for your condition, there are equally effective and less harmful treatments ie placebo effect, natural course of the illness, natural remedies, diet and lifestyle changes, etc
If you want the drug or surgical procedure nevertheless, consult a practitioner who is competent not just in the process of needling but in recognising AEs, diagnostics, contra-indications, and also someone that is able to recognise alternative approaches that may be more aligned to your needs
Ask the clinician of your choice to provide you with the full information relating to the following questions: How many treatments will you need? What will this cost? What are the chances of success? What other therapeutic options are there? What are the risks?
If the clinician cannot provide you with satisfactory answers, don’t let him/her treat you.
if you read it again, it is mostly about consulting a medical practitioner.
@ Greg: You are a mixed up dude.
I thought we weren’t to trust anything that came from China? Or is it different when you agree with the conclusions?
not the first time that you thought wrongly
”
infections (11 reviews, median: 17 cases, median deaths: 0.5),
”
Median deaths is a rather strange way of reporting adverse outcomes here, and makes it rather hard to work out what is going on. I am assuming that you are giving the median of the number of deaths reported in each review, which doesn’t actually tell us very much. It can’t even be correct, as with an odd number of reviews (11), after arranging them in order of number of deaths reported in each review, the median will be the number of deaths in the 6th review, which has to be a whole number.
If there were an even number of reviews then the median deaths could be 0.5, which would tell us that in half of them there were no deaths, and in at least one of them there was exactly one death, but potentially many more in the rest.
Wouldn’t it be more informative simply to give the total number of deaths?
Like some of the readers of this blog, I, too, am an ND (Not Doctor). In fact, I am not a medical practitioner at all. That means I bring no preconceived ideas to these things except that I am pretty sure I will consider a treatment if there is evidence it works and pretty sure I won’t if there is none.
The downside of not being a medical practitioner is that sometimes, I am completely baffled by what I read.
What, exactly, is “non-invasive acupuncture”? Isn’t acupuncture invasiveness the point? (Pun not intended but you have to admit, it’s a good one.) Do they leave the needles on a table in another room and try to convince you that you will soon start to feel better? Do they just pretend to stick needles in you? How does it work?
Thanks, in advance.
An example of Ernst said here “Acupuncture is often promoted as a totally safe therapy.” is reported in the following link of Medscape: https://www.medscape.com/viewarticle/911641
The physiotherapist John Burns claimed acupuncture has no risk of side effects. It looks like Burns and Medscape staff that allowed this sentence to published should know better. What a disservice to patients and other health professionals.
Interesting article. None of my patients have ever had any negative side effects. Maybe the study mentioned turned out to be a malpractice case?
Ghosh, that’s unusual!
~10% of all patients have unwanted side effects
@EE
Wrong wrong wrong…. don’t twist the fact to make it say what you want.
What was said is that “only around 10% of serious adverse drug reactions are reported to agencies that monitor drug safety.”
So likely the problem is much bigger than is even being acknowledged.
C’mon professor…. don’t play dumb.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC48133/
https://pubmed.ncbi.nlm.nih.gov/14718331/
Said where? I can’t find a link.
In the UK we have a system whereby doctors can report adverse drug reactions by filling in a yellow card which is sent to the monitoring agency. The purpose is to identify new, rare or unexpected reactions so that other prescribers can be alerted. However, only a small proportion of actual reactions are reported, because most adverse drug reactions are not unexpected. It would be a complete waste of everybody’s time (including the monitoring agency’s) if I reported every time somebody on chemotherapy dropped their white blood count to a dangerous level, or even every time somebody had an anaphylactic reaction to penicillin, as that would be neither new nor surprising, and is not the purpose of the monitoring system.
Since I don’t know what system (or indeed what country) you are referring to, I don’t, of course, know if this is the explanation for the “only around 10%” to which you refer.