New Research Challenges One of the Oldest Assumptions About Spinal Manipulation
By Jerome Fryer, DC
For decades, one of the most common explanations for spinal manipulation has been remarkably simple:
The adjustment restores spinal motion.
Whether described as “freeing a stuck joint,” “correcting a fixation,” or “improving segmental mobility,” the mechanical model has been deeply ingrained in chiropractic and manual therapy. But a newly published randomized controlled trial asks an important question:
What if pain improves after spinal manipulation—even when measurable spinal motion doesn’t?
Rather than weakening the case for spinal manipulation, this study may actually help us refine our understanding of how it works.
A New Look at an Old Question
A recent randomized controlled pilot study published in the Journal of Manual & Manipulative Therapy investigated whether lumbar spinal manipulation produces measurable changes in segmental lumbar motion using diagnostic ultrasound. Forty-four adults with low back pain were randomly assigned to receive either a high-velocity, low-amplitude (HVLA) spinal manipulation or a carefully designed sham intervention. Researchers measured the distance between adjacent lumbar spinous processes before and immediately after treatment while participants were positioned in end-range lumbar flexion. Pain, disability, trunk flexion, and perceived recovery were also assessed.
The findings were both interesting and clinically relevant.
Patients who received spinal manipulation experienced greater reductions in pain and disability than those receiving the sham treatment.
However…
The ultrasound measurements showed no significant increase in segmental lumbar motion following manipulation.
At first glance, that might sound surprising.
Does This Mean Manipulation Isn’t Mechanical?
Not necessarily.
In my opinion, this is where the discussion becomes particularly interesting.
The study measured one specific mechanical variable:
the distance between adjacent spinous processes during lumbar flexion.
That is an elegant and reliable measurement, but it represents only a small part of the biomechanics occurring during spinal manipulation.
It does not directly measure:
- Facet joint separation
- Intervertebral disc deformation
- Changes in intradiscal pressure
- Ligament strain
- Capsular tension
- Foraminal dimensions
- Fluid movement within the disc
- Mechanoreceptor activation
- Load redistribution throughout the motion segment
In other words, the absence of increased interspinous motion does not necessarily mean there were no meaningful biomechanical changes.
It simply means this particular measurement did not change.
That distinction is important.
My Research Changed How We Think About Cavitation
This question reminds me of work my colleagues and I published in PLOS ONE in 2015.
Using real-time magnetic resonance imaging (MRI), we became the first researchers to directly visualize cavitation during spinal manipulation. Rather than vertebrae “snapping back into place,” we observed a rapid separation of the facet joint immediately before the audible release—the familiar “pop” associated with manipulation.
The findings supported the process of tribonucleation, where dissolved gases rapidly form a cavity within synovial fluid as joint surfaces separate.
This demonstrated something important:
Spinal manipulation unquestionably produces a real biomechanical event.
But that biomechanical event is transient.
Joint separation during manipulation does not necessarily result in a lasting increase in gross spinal range of motion.
Those are two very different concepts.
Interestingly, the new ultrasound study is entirely compatible with that interpretation.
Temporary Mechanical Events vs Permanent Motion Changes
Many clinicians instinctively assume that if manipulation works mechanically, the spine should move more afterward.
But biology isn’t always that simple.
Consider stretching.
A tendon may experience substantial tensile loading without producing a permanent increase in joint motion after a single stretch.
Likewise, spinal manipulation may briefly alter:
- facet joint loading
- capsular tension
- disc pressures
- ligament strain
- mechanoreceptor firing
without producing measurable increases in overall lumbar flexion.
Transient mechanical events can still have important physiological consequences.
Mechanotransduction May Be Part of the Story
One area receiving increasing scientific attention is mechanotransduction—the process by which cells convert mechanical forces into biological responses.
Virtually every tissue within the spine responds to mechanical loading.
Intervertebral disc cells, cartilage, ligaments, bone, muscles, and even inflammatory pathways all react to:
- compression
- tension
- shear
- pressure
- fluid flow
Cells do not require large visible movements to detect mechanical stimuli.
In fact, relatively small changes in tissue strain may initiate complex biochemical signaling pathways.
This is one reason why simply measuring spinal range of motion may fail to capture the entire therapeutic picture.
Pain Is More Than Motion
The authors appropriately discuss that improvements following spinal manipulation likely involve multiple interacting mechanisms rather than a single explanation. They note that pain modulation, descending inhibition, psychological factors, contextual effects, and the therapeutic alliance may all contribute to patient improvement. At the same time, they caution that their study did not directly measure these mechanisms.
That balanced interpretation is important.
The study does not conclude that spinal manipulation works only through neurophysiological mechanisms.
Nor does it demonstrate that biomechanics are irrelevant.
Instead, it shows that one commonly assumed mechanical explanation—an immediate increase in ultrasound-measured segmental motion—was not supported in this sample.
Those are very different conclusions.
Perhaps We’ve Been Asking the Wrong Question
Rather than asking:
“Does spinal manipulation increase spinal motion?”
perhaps we should be asking:
- What mechanical events occur during manipulation?
- Which tissues are being loaded?
- How are mechanoreceptors stimulated?
- How does transient joint separation influence surrounding tissues?
- Which biological responses matter clinically?
These questions may ultimately prove far more informative than measuring spinal range of motion alone.
Where Do We Go From Here?
This pilot study should not be viewed as disproving mechanical models of spinal manipulation.
Instead, it challenges us to think more precisely.
As imaging technology continues to evolve—including ultrasound, MRI, elastography, dynamic fluoroscopy, and other advanced biomechanical assessment tools—we may discover that the most clinically meaningful mechanical events are not the easiest ones to measure.
Science often progresses this way.
We begin by measuring what is convenient.
Eventually, we learn to measure what truly matters.
Clinical Take-Home Message
For clinicians, this study offers a practical reminder.
Patients improved.
Pain decreased.
Disability improved.
Yet those benefits were not accompanied by measurable increases in ultrasound-assessed lumbar segmental motion.
Rather than viewing this as contradictory, we should see it as an invitation to better understand the remarkable complexity of spinal manipulation.
The spine is not simply a collection of moving bones.
It is a living, adaptive biological system where mechanics, neurophysiology, tissue biology, and patient expectations interact in ways we are only beginning to understand.
Perhaps the most important lesson from this study is not that spinal manipulation fails to increase motion.
Perhaps it’s that we’ve been measuring only one small piece of a much larger puzzle.
References
- Winslow J, Costello M, Pantina V, et al. Ultrasound-assessed segmental lumbar motion and clinical outcomes following spinal manipulation: a randomized, controlled pilot study. Journal of Manual & Manipulative Therapy. Published online July 26, 2026. DOI:10.1080/10669817.2026.2701340.
- Kawchuk, Fryer JC, Mykietiuk C, Symons B, et al. Real-Time Visualization of Joint Cavitation. PLOS ONE. 2015;10:e0119470.
