Autonomic nervous system and endocrine system response to upper and lower cervical spine mobilization in healthy male adults: a randomized crossover trial.

The study was set up to detect a large effect, but the effect it actually found was moderate, so a real difference could have been missed. The authors say this themselves. Only 20 people were tested, and cortisol and heart rate variability varied a lot between them, which makes a single significant result at one timepoint easier to produce by chance. Everyone was a healthy man aged 21 to 35 who was not stressed, so the results say nothing about your patients with persistent pain, illness or a disrupted stress response. Women were excluded entirely. There was no sham or control group, so it is impossible to know whether the cortisol change came from the mobilisation or from lying still, being in the clinic, or simply time passing. The authors state that testing efficacy was not the aim. The conditions were tightly controlled to keep the physiology clean, which the authors say limits how far the findings apply outside that room. The data were not normally distributed, so the researchers could not test whether the two techniques and the passage of time interacted. Only one measure was significant on the between-group comparison at 30 minutes, and it pointed in the opposite direction to the within-group result reported for lower cervical mobilisation, which makes the finding hard to interpret. In a crossover trial this small, a carryover effect from the first session into the second is possible, although the authors report their checks ruled this out.

The work was supported by the Stanley Paris PhD OMT Scholarship and the Mark Steptoe Memorial Trust Fund. The declaration supplied lists no other funding or competing interests.

Do not conclude that choosing between upper and lower cervical mobilisation lets you steer a patient's stress hormones. The trial had only 20 healthy, unstressed young men and no sham or control group, the authors state that testing efficacy was not the aim, it was powered for a large effect that turned out moderate, and the significant cortisol difference appeared at a single 30-minute timepoint while the overnight measure showed nothing.