Effects of mobilization or manipulation of the thoracic spine on autonomic nervous system markers in symptomatic and asymptomatic participants - a systematic review and meta-analysis

OBJECTIVES: To identify the effects of thoracic mobilization/manipulation on autonomic nervous system responses.

METHODS: Four electronic databases were searched for controlled trials published before February 2024. Studies on mobilization/manipulation of the thoracic spine were included evaluating effects on the autonomic nervous system (ANS). Risk of bias was assessed by two independent assessors using the Cochrane risk-of-bias tool 2, the RoB-2 tool for crossover studies or the ROBINS-I tool. Meta-analyses using random-effects models present the overall combined mean effects.

RESULTS: 2139 articles were identified, 20 studies (863 participants) were included in the qualitative data analysis and 15 in meta-analyses. Four studies had a high risk of bias in one or more domains. Meta-analyses indicated no statistically significant effect of mobilization or manipulation on markers of the ANS. The ratio of low-frequency-to-high-frequency power did not significantly decrease after thoracic mobilization/manipulation compared to any type of control intervention (-0.28; 95% CI -0.59 to 0.04; p=0.09). Skin conductance and root mean square of successive RR interval differences as well as LFab (ms^2; absolute power of the low-frequency band) did not significantly increase after thoracic mobilization/manipulation. Subgroup and sensitivity-analyses indicated no significant effects.

DISCUSSION: Methodological limitations and heterogeneity (I2=0-94%) in reported outcomes, reduce the level of evidence. Future studies with a rigorous methodological approach and studies on symptomatic participants with longer follow-ups are warranted.

CONCLUSION: No significant effects of mobilization/manipulation of the thoracic spine on ANS markers were found. The direction of changes towards increased or decreased sympathetic or parasympathetic nervous system activity was ambiguous.