Chiropractic Researchers Defend Infant Neck Study Amid Diagnosis Debate

A scholarly dispute over how to interpret measurements of the upper cervical spine in infants has escalated into a public exchange in the pages of Pediatric Radiology, one of the field’s leading journals. The controversy centers on a cross-sectional study by a team from Balgrist University Hospital and the University of Zurich, which examined the correlation between examiner judgment and radiological digital pictures in infants with suspected upper cervical spine dysfunction. After a critical letter from Nawaz Tahir and colleagues challenged aspects of the work, the Swiss team has now published a formal reply, defending the scope of their study while conceding broader points about the state of evidence in pediatric manual therapy.

The original study, published on 18 August 2026, set out to answer a deceptively simple question: when a trained examiner palpates and assesses the upper cervical spine of an infant, does that clinical judgment correspond to what is visible on radiological digital images of the same region? The upper cervical spine, comprising the atlas and axis vertebrae at the top of the spinal column, is a technically demanding area to evaluate in babies. Infants cannot report symptoms, the anatomy is small and largely cartilaginous, and positional asymmetries of the head and neck, often presenting as torticollis, can arise from many different underlying causes. Establishing whether hands-on assessment tracks with imaging findings is therefore a methodologically meaningful question, and one with direct implications for how clinicians decide when imaging is warranted.

In their letter, Tahir and colleagues acknowledged the methodological quality of the Swiss work and expressed general agreement with the authors’ interpretation of the findings. Their critique, however, focused on two distinct issues. The first concerned the use of radiological digital pictures of the upper cervical spine in infants presenting with torticollis. The second concerned the use of spinal mobilization as a treatment approach for the same condition. Crucially, the letter framed both objections in the context of congenital muscular torticollis, a specific diagnosis in which shortening or fibrosis of the sternocleidomastoid muscle causes the head to tilt and rotate. That framing, the Swiss authors argue, is where the critique loses its footing.

In their reply, published on 9 September 2026, Jan Peter, Miklos Csato, Inga Paravicini, Martina Wehrli, Petra Schweinhardt and Anke Langenfeld point out a fundamental mismatch between the critique and their study population. The infants enrolled in the cross-sectional study had been referred for chiropractic assessment and treatment only after undergoing evaluation for congenital muscular torticollis by the referring pediatrician or chiropractor. None of the infants had actually received a diagnosis of congenital muscular torticollis. In other words, the study examined a group of babies whose head and neck posture problems remained unexplained after the most common muscular cause had been ruled out, a population that is clinically distinct from the one the letter’s cited guidelines address.

This distinction matters because torticollis is not a single disease but a descriptive sign. Congenital muscular torticollis is the most frequent cause in infancy, but positional plagiocephaly, ocular imbalance, skeletal anomalies, and neurological conditions can all produce a similar tilted head posture. Guidelines recommending one management pathway or another are typically written for a specific diagnosis, and the Swiss authors note that the guidelines cited in the critical letter refer specifically to congenital muscular torticollis, not to other forms of torticollis. Applying those recommendations to a cohort deliberately composed of infants without that diagnosis, they contend, misreads what the study was designed to show.

The second point of contention, spinal mobilization in infants, touches on one of the most sensitive debates in pediatric musculoskeletal care. An international evidence-based position statement for physiotherapists, published in 2024 by Gross, Olson, Pool and colleagues in the Journal of Manual and Manipulative Therapy, advises against any form of spinal mobilization or manipulation in infants, irrespective of diagnosis. The Swiss authors state that they are very much aware of this position statement and do not dispute its content. Instead, they draw attention to its evidentiary basis: the statement, they note, rests on the limited availability of high-quality evidence, meaning that no firm conclusion can currently be reached about the effectiveness or safety of manual treatments in this age group.

The authors are equally emphatic that their own study cannot fill that gap. The objective of the cross-sectional study, they stress, was not to evaluate treatment effectiveness or to compare therapeutic approaches. Accordingly, they write, the findings should not be interpreted as evidence supporting or rejecting the appropriateness or effectiveness of any particular intervention. This is a point of scientific logic as much as of clinical policy: a study that measures agreement between two diagnostic assessments, in this case examiner judgment and radiological imaging, says nothing about whether treating the identified findings changes outcomes for the child. Conflating diagnostic correlation studies with treatment trials is a common source of misinterpretation in the literature, and the reply serves as a pointed reminder of the difference.

At the same time, the exchange highlights a genuine and unresolved tension in the field. If high-quality evidence on manual therapy in infants is scarce, then both the clinicians who use these techniques and the bodies that advise against them are operating in an evidentiary vacuum. The Swiss authors close their reply by fully agreeing that high-quality evidence on both the effectiveness and the safety of manual treatments in infants, including spinal mobilization and manipulation, is urgently needed. That concession is notable: rather than dismissing the critique, they accept its most important implication while defending the narrower claims of their own research. The letter, they conclude, was critical yet constructive, and they welcome it as such.

For readers outside the specialty, the dispute is a case study in how carefully study populations must be defined before research findings can be generalized. A cohort of infants referred for chiropractic assessment after congenital muscular torticollis was excluded is not a cohort of infants with congenital muscular torticollis, and conclusions drawn from one cannot be transplanted to the other. Similarly, a measurement that is reliable, meaning that repeated assessments produce consistent results, does not automatically constitute a reliable diagnosis, a point captured in the very title of the critical letter. Reliability speaks to consistency; diagnostic validity speaks to whether the measurement identifies a true underlying condition. The two concepts are related but distinct, and much of the friction in this exchange flows from the gap between them.

The reply, authored by clinicians and researchers from the Department of Chiropractic Medicine and the Department of Radiology at Balgrist University Hospital and the University of Zurich, was received on 24 August 2026, revised on 28 August, accepted on 2 September, and published as an open-access Matters Arising contribution. Corresponding author Anke Langenfeld and Petra Schweinhardt drafted the response, with the remaining authors critically reviewing and approving it. The authors declare no competing interests. Whether the exchange will stimulate the large, rigorous trials that both sides agree are lacking remains to be seen, but the episode demonstrates how the peer-reviewed correspondence column continues to function as a venue for sharpening the questions that matter, particularly in a field where the patients are too young to speak for themselves and the stakes of getting the evidence right could not be higher.

Subject of Research: Diagnostic correlation between examiner judgment and radiographic imaging of the upper cervical spine in infants with torticollis

Article Title: Reply to: “A reliable measurement is not the same as a reliable diagnosis” by Nawaz Tahir et al.

Article References: Peter, J., Csato, M., Paravicini, I., Wehrli, M., Schweinhardt, P., & Langenfeld, A. (2026). Reply to: “A reliable measurement is not the same as a reliable diagnosis” by Nawaz Tahir et al.. Pediatric Radiology. https://doi.org/10.1007/s00247-026-06783-z

Image Credits: AI Generated

DOI: 10.1007/s00247-026-06783-z

Keywords: pediatric radiology, torticollis, congenital muscular torticollis, upper cervical spine, chiropractic medicine, spinal mobilization, radiography, diagnostic reliability, infants, manual therapy, position statement, Balgrist University Hospital