The title of this post isn’t an exaggeration. I will be discussing a paper published late last year in the Journal of Chiropractic Medicine that retrospectively analyzed the outcome after six infants with a common musculoskeletal condition were placed under general anesthesia, with a breathing tube and everything, in order for a chiropractor to manipulate their neck.

This isn’t satire.

A pediatric anesthesiologist was involved, as was a pediatric radiologist. Nurses were there and a pediatric surgeon was present in one case. This happened at a children’s hospital in Switzerland, I assume with the approval of the institutional review board, and was written up by a member of the faculty at the University of Zurich. Again, this isn’t satire. This isn’t a joke. This is an egregious breach of medical ethics. It is medical child abuse.

What were they thinking?

This question has been rattling around my head for a couple weeks now, and I’m still at a bit of a loss. It’s easy to chalk it up to the numerous cognitive errors that plague mankind, and the way that the unconscious need to avoid cognitive dissonance serves as a powerful motivation to rationalize even the most horrific beliefs and behaviors. That’s ultimately how this study came to be conceived and implemented, of course, but it isn’t the whole picture.

For this study to happen, for these six children to be so fundamentally failed by the Swiss healthcare system, it required more than just a reckless chiropractor with delusions of legitimate research methodology and a seemingly preternatural ability to fool herself. She needed access to a medical facility and the necessary personnel to pull it off. She also needed parents to consent to it, although I try to not blame the victims. They were lied to.

The faculty member mentioned above as the author of the study, Inga Parvicini, is also the deluded chiropractor. She selected the six victims featured in this paper from the infants treated by an unnamed Swiss chiropractor with manipulation under anesthesia (MUA) over the past 25 years. In each case, she had assisted in the procedure.

Paravicini is also the curriculum director for the UZH master’s course in chiropractic medicine. So in Switzerland, it would seem abject quackery has been fully integrated into pediatric medical practice. Some of these kids were actually even referred for this treatment by their pediatrician.

What is torticollis?

Also known as wry neck, torticollis is a common condition with many causes that result in twisting of the neck to one side because of pain, stiffness, or neurological defects. In older children and adults, it is frequently caused by a self-limited soreness of the muscles of the neck and shoulder, the classic crick in the neck. By far the most common form, however, occurs in young infants and is known as congenital torticollis. We have discussed torticollis before, including that time a chiropractor broke a baby’s neck in a vain and unnecessary attempt to fix torticollis.

Congenital torticollis is diagnosed when a newborn or young infant presents with difficulty turning their head, with a classic resting position where the head is tilted to one side but turned to the opposite. The most common subset of congenital torticollis is congenital muscular torticollis (CMT), which occurs in as many as 15% of newborns, most of which simply have a positional preference based on their position in the womb that quickly resolves without intervention. More than 80% of torticollis cases fall into this category.

Roughly 1 to 4% of newborns will develop a more severe version of CMT, which is caused by an injury to one of the sternocleidomastoid (SCM) muscles on either side of the neck. These muscles run from the sternum and clavicle to the bony protuberance just behind the ear and are used to turn the head up and to the opposite side, or, when acting in unison, to look straight up or down. The most likely cause of tightness in one of these muscles is injury while still in the womb, which is why it tends to present in the first few weeks of life. Later presentations are more likely to have an acquired cause such as an infection or traumatic injury.

CMT is something that newborn healthcare providers are always on the lookout for because earlier treatment tends to result in faster resolution. It is a primarily clinical diagnosis based on positioning of the head at rest, reduced range of passive movement, presence of tightness or a mass in the SCM, and a lack of concerning signs of a non-muscular cause. Radiographic imaging is rarely necessary in babies that have a typical presentation, although sometimes ultrasound of the SCM muscle can be helpful in confirming injury and an X-ray or CT scan can rule out bony abnormalities when a child presents in an unusual way.

Infants with congenital torticollis do sometimes have a non-muscular etiology, one of which is an abnormality in the cervical spine, though this kind of anomaly is quite rare. These vertebrae need to be able to move freely for the neck to turn appropriately with a normal range of motion.

One of these cervical spine abnormalities is a true subluxation of the first and second vertebra, which is different from the chiropractic subluxation that we have written about many times on SBM in that it exists. Also known as an atlantoaxial rotary subluxation, it involves a displacement of the normal alignment of the first and second vertebra caused by swelling and inflammation from an infection, trauma, or recent surgery in the area. It typically occurs in older children and in most cases is treated with pain control, rest, and a soft cervical collar. In severe cases, immobilization and surgery may be necessary.

A chiropractic understanding of congenital torticollis

According to the author of the study, there is another category of newborn torticollis known as the arthrogenic type. Try looking this term up and you won’t find any legitimate discussion of it. You will find the study we are discussing and that’s about it. It appears to be completely fictional. According to the author, however, it has German origins:

In Germany, this condition sometimes is called kopf-gelenk indurzierte symmetrie storung…this type of torticollis indicates a joint dysfunction caused by capsular adhesion at the atlantoaxial joint.

In other words, it is just another nebulous description based on a functional defect with a cause that can’t be tested or disproved. Paravicini goes into horrific detail regarding treatment of this fictional entity:

Because the fixation of the atlantoaxial joint is not due to a neuromuscular reflex mechanism but rather due to capsular adhesions, the specific mobilization is without impulse and the sound that is heard during the mobilization is not a cavitation but rather the breaking up of adhesions.

Remember that she is talking about the treatment of infants and describing manipulation of a child’s neck vigorous enough to make a cracking sound. Most chiropractors these days will admit that the sound being made is simply the escape of gas from the joint space, so-called cavitation, and that it is not meaningful. Paravicini has invented an alternate explanation in order to give a false impression that her preferred approach is fixing a problem. That tingling sensation means it’s working!

She claims that 2.5% of children with “arthrogenic torticollis” will not respond to standard chiropractic interventions and may be candidates for manipulation under anesthesia. Where does she get this estimate from? Keep in mind that we are talking about exposing infants to general anesthesia and a breathing tube while reading her answer:

These numbers are anecdotal data originating from 1 specialized chiropractor in Switzerland who treats approximately 400 infants per year with arthrogenic newborn torticollis. From these 400 infants, 10 will end up needing MUA due to the reasons mentioned above.

There have been no studies looking into the safety and efficacy of MUA in infants. Nobody with a shred of ethical integrity would have even considered this as an intervention. The “reasons mentioned above” are not worth your time because they are manufactured rationalizations conjured up in an attempt to force the variation of congenital muscular torticollis outcomes, assuming that this is even a reliably given diagnosis in every case seen by the mysterious unnamed chiropractor, into a convenient pattern. Arthrogenic torticollis does not exist. Infant MUA is an abomination.

And now the study

Paravicini essentially looked at the files of a bunch of infants who had undergone MUA for torticollis, and whose treatment she had assisted in for the unnamed Swiss chiropractor with, and picked six of them to write up. She doesn’t explain why she chose these six cases. She does explain that the indication for MUA in each was arthrogenic newborn torticollis verified by x-ray, which the pediatric radiologist I spoke with laughed at. She also claims that each patient had failed to improve despite several weeks of standard treatment, which might have included chiropractic, osteopathy, or craniosacral therapy in addition to physical therapy or home-based interventions. I do not have confidence in her claim that conventional treatment was appropriately performed.

The six cases included male infants aged 4, 4.5, 5.5, 8.5, 9, and 10 months who each had descriptions consistent with classic congenital muscular torticollis without any need to invoke the arthrogenic label. Each had X-rays done to rule out a rare bony abnormality that would make neck manipulation dangerous, though three of them required CT scans for added reassurance. All of this radiation was likely completely unnecessary.

The infants were put to sleep using an IV agent and a breathing tube was placed under the care of a pediatric anesthesiologist. The author assisted the unnamed Swiss chiropractor in providing “mobilization” of each child’s first two cervical vertebrae “with minimal force and with no impulse”. This essentially means that they didn’t do a high-velocity low-amplitude adjustment but simply moved the neck around a bit while gently pushing on the neck. Although the noise claimed to be caused by breaking up adhesions was again mentioned, she did not actually say if it occurred in these cases.

The primary outcome that they looked at was active (patient moves on their own) and passive (the patient is moved by an examiner) range of motion of the neck. They documented this before and after MUA at visits 2 and 6 weeks later. Unsurprisingly, they found improvements. This is a problematic assessment, however, because this obviously was not blinded. It is easy to provide a little more or less force when assessing passive range of motion. Also, assessing active range of motion in infants, one of which had severe developmental delays, is very unreliable and subjective.

The secondary outcome was parental feedback at the 2 and 6 week visits and then by phone interview anywhere from 6 to 72 months later depending on their availability. This again is challenging to interpret because of the random timing of the follow-up but mostly the lack of blinding. The parents knew that their child had undergone MUA and they had to have known that this was an unusual intervention. It likely was not cheap and some of them had to drive for hours to seek out care from this chiropractor. So there was ample motivation to both perceive their child as being subjectively better and to perhaps become more adherent to more conventional treatment approaches.

The author concludes that MUA is safe and may be effective in the treatment of infants. Neither claim is justified by her case series. She points out that this was the first study of its kind and suggests that there are likely infants out there who have suffered harm because MUA wasn’t performed on them. She downplays the potential risk from radiation and general anesthesia.

The potential risk from radiation and general anesthesia

Important in any discussion of medical interventions is an assessment of the risks and benefits of both doing and not doing that intervention. In an infant, there is a small but real risk associated with exposure to radiation. The younger the patient, the more time there is for cellular damage from ionizing radiation to result in cancer. One CT scan using pediatric protocols is unlikely to cause lymphoma down the road, but we don’t know what the future holds for thess children regarding a potential need for additional studies for different indications.

More worrisome is the effect of general anesthesia on the developing brain. There have long been concerns based on a lot of animal data showing neurocognitive harm that exposure to anesthesia in young children may result in problems with learning and development later in life. The FDA and the International Anesthesia Research Society started SmartTots to help address these concerns.

What we have learned is that a single, brief (less than 1-2 hours) exposure is unlikely to cause any major issues, but we can’t say that it is completely safe in this regard. There may be minor delays or dips in a child’s eventual IQ that haven’t been picked up. The AAP recommends that general sedation be avoided in infants and toddlers under age 3 years unless it is deemed to be truly necessary. It often is, but a rational assessment of risk and benefit is necessary. As with radiation exposure, we don’t know if these children will need additional unrelated procedures that may require sedation in the near future.

In the case of MUA for infant torticollis, risk is far greater than potential benefit

CMT sometimes resolves on its own without intervention, although that is uncommon. Untreated CMT often results in permanent craniofacial asymmetry that is of cosmetic concern. So for most infants, treatment is strongly indicated.

More than 90% of cases will respond to simple interventions such as changes in positioning and handling, environment adaptations, and physical therapy at home if started before 12 months of age. Sometimes professional physical therapy is indicated, which almost always works in that subset of patients. The longer it takes for treatment to be initiated, the longer treatment takes to work, but it almost always works. Surgery is typically only required when treatment is initiated late but results in full function and resolution of craniofacial asymmetry that has developed in almost 90% of cases.

So essentially it is unlikely that conventional treatment failed the 6 infants described in the Paravicini paper. And I am extremely skeptical of claims that MUA truly hastened resolution. I am even more skeptical of the existence of arthrogenic torticollis in anything other than the minds of certain chiropractors.

This paper is useless as a means of answering the question of MUA. It’s a collection of a small number of self-selected anecdotes after the fact. There was no control group. The collection of outcome data was highly suspect. And again I want to be clear in my condemnation of the author, the University of Zurich, and the actual medical professionals that let this happen on their watch.


Posted by Clay Jones

Clay Jones, M.D. is a pediatrician and a regular contributor to the Science-Based Medicine blog. 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 the incursion of pseudoscience into his chosen profession while completing his pediatric residency at Vanderbilt Children’s Hospital a decade ago. He has since focused his efforts on teaching the application of critical thinking and scientific skepticism to the practice of pediatric medicine. Dr. Jones has no conflicts of interest to disclose and no ties to the pharmaceutical industry. He can be found on Twitter as @SBMPediatrics and is the co-host of The Prism Podcast with fellow SBM contributor Grant Ritchey. The comments expressed by Dr. Jones are his own and do not represent the views or opinions of Newton-Wellesley Hospital or its administration.