Like acupuncture, Weebles wobble, but they don’t fall down.

The practice of medicine, particularly our pharmaceutical and surgical interventions, involves a constant struggle between risk and benefit. If the physiology or anatomy of the human body is altered, even with the best of intentions, there is always a potential downside. There are certainly instances where the risk to benefit ratio is extremely favorable or unfavorable and the right recommendation is obvious, and unfortunately there are times when it isn’t entirely obvious what the next step should be. But there has been a trend of steady progress in regards to improved safety and efficacy over the past several decades.

The treatment of pain has of late been one of those areas where the picture is becoming a bit less cloudy. We are learning more and more about the potential negative outcomes related to the long term use of opioid medications, such as physical dependence, addiction and even chronic pain. The way that these drugs have been prescribed in many patients has caused more harm than expected, and in some instances more hurt than help. Doctors generally strive to alleviate pain and suffering but, once again, good intentions don’t decrease risk.

In the neonatal and young infant population, the management of pain has had a rocky history. I’ve written about pediatric pain in the past, in particular the potential difficulties in managing acute pain. I won’t go into detail (read my prior post), but we have truly come a long way since the days of performing major surgery on newborns without any analgesia at all. There are areas where we need to do better, however. Children are still less likely than adults to be adequately treated for pain.

But things have improved. And as more children receive appropriate management for pain, the side effects of that management must increasingly be dealt with by healthcare professionals, the patients and their families. One of the issues that is typically observed and managed in neonatal and pediatric intensive care units is physical dependence and the subsequent occurrence of withdrawal symptoms.

What are dependence and withdrawal?

No different than with adults, children experience pain for a variety of reasons, and some require the use of opioid medications for extended periods of time. Physical dependence with opioids can occur after just five to seven days of daily exposure. In the setting of the intensive care unit, there are children that are sometimes on medications for pain and sedation for weeks and even longer.

The American Academy of Pediatrics, in their 2014 clinical report on the subject, provides the following definition for physical dependence:

Physical dependence is a state of adaptation that is manifested by a drug class–specific withdrawal syndrome that can be produced by abrupt cessation, rapid dose reduction, decreasing blood level of the drug, and/or administration of an antagonist.

Withdrawal symptoms occur when there is less of the drug in the patient’s system, which happens for essentially three reasons. As the patient recovers, opioid pain medication dosing is often decreased or spaced out. Intravenous medications are switched to oral formulations, which isn’t always a predictable transition. And sometimes the absorption of medications is decreased because of illness or injury to the gastrointestinal tract. Also, as referred to in the above definition, a medication that blocks the action of another drug, such as naloxone (Narcan) used for opioid overdose, can lead to withdrawal symptoms even when there hasn’t been a drop in the blood level.

Not every pediatric patient experiences withdrawal, and when they do it can be mild enough to escape the awareness of caregivers or to be blamed on something else. Is the baby fussy because of mild withdrawal from morphine, or is it just gas, for example. Of course the symptoms of more serious conditions can be blamed on withdrawal as well. When significant withdrawal does occur, it can interfere with recovery of the patient, expose them to increased risk of medication side effects and increase caregiver anxiety tremendously.

The most common and obvious changes seen during withdrawal in children are behavioral. Anxiety, agitation, difficulty sleeping and jitteriness are frequent examples. Also common in withdrawal are stiffening of the muscles, nausea and vomiting, poor feeding/appetite, increased respiratory and heart rate, elevated body temperature, sweating, and elevated blood pressure.

Allow me to plant a seed and point out that there can be significant subjectivity to the assessment of most of these manifestations when caring for neonates and older infants. Was that just a little spit up or should we count it as an episode of vomiting? Is the baby not interested in the bottle because of withdrawal or because premature infants who have been ill sometimes just need more time to work up on feeds? Is the baby agitated, or just a little overstimulated? Perception is reality in the case of withdrawal.

There are validated assessment tools, the most well-known probably being the modified Neonatal Abstinence Scoring System and the Withdrawal Assessment Tool-1 (WAT-1), that can used to help determine if an infant is undergoing symptoms of withdrawal. These tools can be used in the setting of potential withdrawal from opioids or benzodiazepine medications. The WAT-1, which I’ll focus on for reasons that will soon become clear, involves eleven items with a total possible score of twelve and is performed every twelve hours by a nurse caring for the patient.

How is the WAT-1 performed?

It breaks down as follows:

  1. The nurse will review what happened over the previous twelve hours. Infants score a point (points are bad) for having had any loose/watery stools, retching/gagging/vomiting, or temperature elevations above 37.8°C (100°F) up to a maximum of 3 points. Again, I’d like to point out the subjectivity inherent in this assessment. The nurse must decide if the baby’s stool is loose/watery compared to normal baby stools, which are often considered loose/watery. Also perfectly healthy babies have spit ups and brief gagging episodes all the time.
  2. The nurse will then observe the baby for two minutes to get a sense of their behavioral state prior to any stimulation. A baby will only get 0 out of 5 possible points if they are asleep or awake and completely calm. They score a point for any crying, moderate to severe tremor (subjective), sweating, moderate to severe uncoordinated/repetitive movements (subjective), or more than one yawn or sneeze.
  3. The nurse will then progressively stimulate the baby over one minute. They get a point for a moderate to severe startle reaction to touch or if they demonstrate increased muscle tone, both subjective assessments. Finally, the baby is left alone and the nurse documents how long it takes for the baby to calm. One point is scored if it takes the patient between two and five minutes. Greater than five minutes results in a score of two points.

I hope it doesn’t sound like I am trying to say that an assessment like the WAT-1 is worthless. It definitely isn’t. It is an improvement over previous methods which were much more cumbersome, involving many more items and more frequent assessments. And it is certainly less subjective than relying solely on the clinical judgement of the nurses and physicians. Although, tools like the WAT-1 are validated by comparing them to just such previous methods. They all must be taken with a grain of salt. Their greatest benefit is likely the fact that their use forces us to consider the possibility of withdrawal and to approach it systematically.

How is withdrawal managed?

The best approach to withdrawal in the pediatric population, and I imagine in adult patients as well, is anticipation and prevention of withdrawal if at all possible. If a child has been on a daily opioid for more than one week, although there are reports of withdrawal occurring after only five days, the medication should probably not be discontinued abruptly. And an assessment tool like the WAT-1 should be used consistently and as recommended.

It is common practice in hospitals to take into account how long the patient has been on a daily opioid, how high the doses are at the time weaning begins, concurrently administered medications (benzodiazepines, paralyzing agents, etc.), and the patient’s overall medical condition. It is also standard of care to follow an approved weaning protocol, generally with only one drug (if the patient is on more than one associated with withdrawal) being weaned at a time and the rapidity of the wean being determined by the length of time the patient had been on the drug. Naturally there are potential individual differences between patients, so signs and symptoms of withdrawal trump following a strict algorithm.

A typical approach would be to transition the patient to a stable dose of a longer-acting form of the drug to be weaned, and then to decrease the amount by 10%-20% every couple of days if excessive rescue medication dosing hasn’t been needed. An approved assessment tool should be used to determine signs or symptoms of withdrawal and the need to delay a decrease in dosing or to give rescue dosing of a short-acting version of the drug being weaned. A WAT-1 score greater than or equal to three, for example, is a typical cut-off for consideration of rescue dosing. Depending on the individual circumstances it can take a few days to weeks for a child to be successfully weaned and ready for discharge home.

Is there a role for acupuncture in the management of withdrawal in neonatal and pediatric intensive care units?

No, no there isn’t.

Weaning babies off of these medications can go smoothly or be a frustrating roller coaster ride. I confidently speak for parents, patients and medical professionals when I say that we would all love a safe and effective means of weaning these medications more quickly and with less occurrence of withdrawal. If such a means exists, there is no reason to think that it would consist of placing tiny steel needles through the skin or pressing on the ears in order to manipulate the flow of nonexistent mystical energy through nonexistent pathways in the human body.

But thanks to a team of pediatric anesthesiologists at Stanford University, there is now another in a very long line of studies seemingly designed to be positive and to serve as fodder for believers looking for some science, any science, to hold up as proof of their beloved alternative medical belief system. This “research” was published in the October issue of Medical Acupuncture, a journal which has been discussed in the pages of Science-Based Medicine before, and not in a good way. The Editor-in-Chief happens to be Richard Niemtzow, the man who brought us “battlefield acupuncture.”

The authors start off by lamenting the need for painful procedures in young infants and the possible repercussions of poorly managed pain in the very young, such as an amplified pain response down the road. This is true…sort of. Some neonates exposed to a few painful experiences, such as heel sticks or circumcision, may have an exaggerated response to the jab from routine immunizations during the first year of life. But babies who experience repeated and prolonged pain, such as those riding vents for weeks or having major surgeries, are actually at risk of having blunted responses to pain during the first year of life. What is clear however, is that pain causes remodeling of our response to it, perhaps for a long time, and we should do our best to reduce it whenever it is safe and possible to do so.

They also point out the downsides to our new and improved approach to pain management, which to be honest is simply the fact that we have an approach to pain management at all in neonates. But, as I discussed above, it can lead to dependence and withdrawal as well as possible acute side effects with overdoses such as respiratory suppression and death. They mention the possible need for adding additional medications to the mix, like clonidine, to help manage withdrawal. I didn’t get into that in detail, but it’s true. And like every other drug, there are potential risks when prescribed.

So far, so good.

The authors then go off track, although not unexpectedly, when they begin to praise the successes of acupuncture for withdrawal. They state that there is a long history of safety and efficacy starting with the incidental discovery of the indication in 1972 when “acupuncture anesthesia” was performed on a man addicted to opium in China. As Dr. Atwood’s excellent four part series explains, acupuncture anesthesia was and is a fiction, and anything associated with it is highly suspect.

They describe the protocol for acupuncture detoxification as established by the National Acupuncture Detoxification Association (NADA), which initially involved multiple points on the body and ears as well as electrical stimulation. Because of logistical issues and “reasons of efficiency in clinics”, both the electricity and the body points were eventually abandoned, luckily without any negative impact on the effectiveness of the treatments. That’s curious. The current NADA protocol consists of “bilateral ear acupuncture of the following five points: Kidney; Liver; Lung; Shen Men; and Sympathetic.

They claim that acupuncture treats anxiety, pain and agitation in a variety of pediatric contexts. After admitting that the only prior study related to pediatric withdrawal, which involved using acupressure in neonates suffering from abstinence syndrome, did not work, they point out the “suggestive trend toward less need for pharmacologic support.” All of this is used to bolster their case for an “exploration of the feasibility and efficacy of performing acupuncture on infants in the ICU.”

The study

The authors enrolled ten neonates and infants in the study with the goal of determining the feasibility and efficacy of acupuncture as part of the management of withdrawal in the ICU. They obtained IRB approval from Stanford and informed consent from caregivers. These infants had few similarities. They were on different doses of medications or were on different medications entirely. One was on a ventilator and still getting doses of a paralyzing agent. They were at varying stages of their wean off of medications, and had extremely different reasons for why they were on these medications in the first place.

The treatment protocol consisted of NADA-approved ear points but they also threw in some body points to “help stabilize symptoms of autonomic dysfunction.” They state this without a reference, as if it’s just a well-accepted thing that doctors do. Like stating that “fevers were treated with standard dosing of acetaminophen.” The authors did their own acupuncture and inserted the needles until they felt the De Qi sensation. After needling the babies, acupressure beads were placed on the points for 24 hours or until the next session. The NADA protocol they were supposedly following only calls for 30 minutes. All but one of the subjects had five sessions. That one went home after three days.

The authors report that all ten of the patients improved while receiving acupuncture and acupressure. Based on WAT-1 assessments, they needed fewer doses of rescue medications and their opioid and/or benzodiazepine infusions were weaned. There were no major complications, although one child suffered a minor skin abrasion from an acupressure bead. And although the nature of the study did not permit any statistical analysis, a survey of bedside caregivers (nursing, physicians, parents) revealed universal belief that the patients benefited from the intervention. The authors even provide several comments from the surveys, such as “Can we please do this for other patients?” and “She really likes her acupuncture.”

So to sum up, ten babies received an intervention for a highly-subjective process, and were assessed without any blinding using a tool that is very susceptible to placebo. The authors of the study, two pediatric anesthesiologists, performed the intervention themselves at the bedside for all to see and included in their methods for no clear reason a means to make sure that anyone assessing the patients during a different shift knew which babies were involved. Then to cap it off, they included a likely non-anonymous survey that is even more subjective than the WAT-1 as part of their evidence for efficacy. I would have been shocked if the results were any less positive.

There are three possible explanations for the results of this study that I can think of. One is that acupuncture and/or acupressure works, not only for withdrawal from opioids but also benzodiazepines, and in a patients with a variety of medical histories. Or we accept that acupuncture functions as a theatrical placebo and the biased perception of the caregivers led to the infants receiving fewer rescue medications and continuing on their weans. If this is the case, which in my opinion it almost certainly is, some or all of the infants may not have received medication that they truly needed. Finally, it all may just be the result of random factors. Placebo didn’t play a role, acupuncture doesn’t work, the process of withdrawal and weaning is highly variable and this was a very small number of subjects.

Regardless, I would love to have been privy to those IRB discussions and to the informed consent process. The latter isn’t described in the paper unfortunately. But this was not one of those papers where the researchers attempt to come up with a scientific-sounding explanation for the proposed benefits of acupuncture. In addition to the “De Qi” reference, they state that “children may require less stimulation than adults because their Qi may be more responsive to stimuli.” So did they really discuss Qi with the parents? Did they mention that a better study looking at this very issue, which was published in the same journal in 2011 and even cited by the authors, showed no benefit whatsoever?

Why does this matter?

Although merely a collection of ten worthless anecdotes published in a journal that warps the concept of peer review beyond all meaning or recognition, there are already individual acupuncture practices touting it as evidence that acupuncture “truly is for everyone!”

Stanford researchers have found that acupuncture helps newborns in intensive care. Acupuncture has been effective for pain relief and sedation, the study showed. “High doses of opioids and benzodiazepines are often required for neonates and infants for the purposes of pain management and sedation. Cessation from medications lead to withdrawal symptoms and irritability. The researchers cite acupuncture’s documented ability to reduce pain, irritability and withdrawal symptoms in adults.”

Here is a website offering continuing medical education credits based on the study.

This study, a painfully unnecessary attempt to determine the feasibility and efficacy of acupuncture in neonatal and pediatric intensive care unit patients at risk of withdrawal, was a complete waste of time. You cannot determine efficacy at all based on it and the question of feasibility was answered a long time ago. This study added nothing to our understanding of withdrawal or even of acupuncture.

Considering the reams of research showing that there are no benefits associated with the insertion of needles into, or pressing on, particular points on the body beyond placebo, it never should have happened. But it can’t be undone. It’s out there, adding to the cultural inertia of acupuncture. It will find its way to more websites and it will be cited in future poorly designed studies. Although less of a fuss was made about this paper than I would have expected, there are almost certainly many more people that are believers now because of it.

Although many forms of alternative medicine are like this to varying degrees, acupuncture really is the best Weeble of the bunch. It has proven time and time again to be able to right itself despite devastating blows in the literature. It’s nearly ubiquitous in academic centers and the modality that shruggies will usually cite when pressed for a reason for their lack of motivation. It’s widely accepted by the public as legitimate. Acupuncture doesn’t appear to be going anywhere anytime soon. In the meantime, why don’t you make supporting the Society for Science-Based Medicine one of your resolutions for 2015.




  • Clay Jones, M.D. is a pediatrician and has been a regular contributor to the Science-Based Medicine blog since 2012. He primarily cares for healthy newborns and hospitalized children, and devotes his full time to educating pediatric residents and medical students. Dr. Jones first became aware of and interested in pseudoscience in medicine while completing his pediatric residency at Vanderbilt Children’s Hospital twenty years ago and has since focused his efforts on teaching the application of critical thinking and scientific skepticism. Dr. Jones has no conflicts of interest to disclose and no ties to the pharmaceutical industry. He can be found on Twitter as @SBMPediatrics.

    View all posts

Posted by Clay Jones

Clay Jones, M.D. is a pediatrician and has been a regular contributor to the Science-Based Medicine blog since 2012. He primarily cares for healthy newborns and hospitalized children, and devotes his full time to educating pediatric residents and medical students. Dr. Jones first became aware of and interested in pseudoscience in medicine while completing his pediatric residency at Vanderbilt Children’s Hospital twenty years ago and has since focused his efforts on teaching the application of critical thinking and scientific skepticism. Dr. Jones has no conflicts of interest to disclose and no ties to the pharmaceutical industry. He can be found on Twitter as @SBMPediatrics.