Chiropractic + Naturopathic Doctor • December 2019 • Research review

STUDY TITLE: The acute effects of joint manipulative techniques on markers of autonomic nervous system activity: a systematic review and meta-analysis of randomized sham-controlled trials

Research into biomechanical and neurophysiological mechanisms have yielded incomplete answers, as has initial research into the effect of joint manipulative therapies (JMT) on the autonomic nervous system (ANS). Effects on the ANS such as somato-autonomic reflex modulation have been proposed as potential mechanisms. Indeed, the chiropractic profession initially professed to “normalize” autonomic activity as an explanation for the treatment effect of high-velocity spinal manipulation. The true mechanism of effect, however, remains elusive. Several literature reviews have attempted to systematically review and meta-analyze results from studies investigating the mechanism of JMT, although they are limited by methodological concerns. As such, the goal of this study was review and critique the existing literature and compare the acute changes in ANS markers between patients receiving JMT on spinal or peripheral joints with those undergoing a sham treatment.

Study Characteristics: 29 studies met the inclusion criteria, after initial screening of 2267 studies. Interventions included mobilizations (n = 16 studies), ‘atypical’ mobilization (n = 1), sustained natural apophyseal glides (SNAGs) (n = 5) and high-velocity, low-amplitude thrust manipulation (HVLA) (n = 7). Treatment was generally localized to the spine, with 2 studies additionally including peripheral joint treatment.

Risk of Bias Assessment: Risk of bias was unclear in all but 4 eligible studies, based largely on uncertainty regarding blinding of participants, the data extraction process and blinding of statisticians. With respect to specifics, 2 studies (both on mobilization) were determined to have low risk of bias while 3 (2 of which were on HVLA spinal manipulation) were deemed to have a high risk of bias.

Overall Technical Quality: Technical quality was judged to be acceptable in 25/29 studies. The remaining studies (2 on mobilization, 2 on manipulation) were deemed to be technically deficient, based on a low technical score.

Sham Procedures: 25/29 studies defined a sham intervention as an “inactive” manual contact (i.e. no movement). 2/29 studies used a sham treatment mechanically similar to the intervention; 1/29 used a sham similar to the true intervention with less pressure and 1/29 did not describe the sham intervention.

• Outcome – Skin Conductance: Moderate evidence (10 studies) suggests that treatment causes a bilateral increase in skin sympathetic nerve activity. Data pooled from 3 studies indicate significant increases over sham (mean difference 13.75, 95% CI 1.36 to 26.14, I2 = 51%, random effect, p = 0.03; 3 studies, 96 subjects) between baseline and intervention period and between baseline and the post-intervention period (mean difference 9.34, 95% CI 2.85 to 15.83, I2 = 0%, p = 0.005; 3 studies, 96 subjects).

• Outcome: Skin Temperature: Very low quality evidence (8 studies) indicates no effect on skin sympathetic nerve activity or skin temperature. No pooling of data was possible.

• Outcome – Skin Blood Flow: Very low quality evidence (2 studies) suggest no effect on blood flow (1 study in fact found dual effects in opposite directions [increase and decrease]).

• Outcome – Heart Rate: Very low quality evidence (5 studies) found no effect on heart rate (beats per minute) during the intervention (mean difference −0.83 bpm, 95% CI -5.47 to 3.81, I2 = 0%, p = 0.73; 83 subjects, 2 studies) and the immediate post-intervention period (mean difference − 1.23 bpm, 95% CI -4.47 to 2.02, I2 = 0%, p = 0.46, 121 subjects). 2 other studies found a significant increase in heart rate between baseline and intervention period, while 1 study reported a significant increase in the post-intervention period.

• Outcome – Blood Pressure: Very low quality evidence (5 studies) indicates no effect on systolic blood pressure during the intervention period (mean difference − 2.02 mmHg, 95% CI -6.96 to 2.92, I2 = 31%, p = 0.42, 83 subjects) and the post-intervention period (mean difference − 1.02 mmHg, 95% CI -5.77 to 3.72, I2 = 0%, p = 0.67, 83 subjects). No effect was noted on diastolic blood pressure during either period (mean difference − 0.07 mmHg, 95% CI -3.09 to 2.94, I2 = 0%, p = 0.96, 83 subjects; mean difference 0.32 mmHg, 95% CI -2.49 to 3.14, I2 = 0%, p = 0.82, 83 subjects). One study noted no effect on arterial blood pressure; 2 studies noted significantly increased systolic blood pressure compared to sham (no pooled analysis).

• Outcome – Heart Rate Variability: Low quality evidence (1 study) indicates no effect on heart rate variability.

• Outcome – Respiratory Rate: Very low quality evidence (3 studies) indicates a significant increase in respiratory rate compared to sham. No pooled analysis was possible.

• Outcome – Alpha amylase activity: Very low quality evidence (1 study) indicates a significant decrease in sympathetic activity in the salivary glands within 10 minutes of treatment.

• Outcome – Skin Conductance: Low quality evidence (4 studies) indicates no effect on skin conductance in the intervention period or post-intervention period. Pooled analysis (2 studies) found no effect in the intervention (mean difference 4.62, CI 95% -2.31 to 11.55, I2 = 0%, p = 0.19, 2 studies, 60 subjects) or post-intervention (mean difference 3.99, CI 95% -3.47 to 11.44, I2 = 0%, p = 0.29, 2 studies, 60 subjects) periods.

• Outcome – Skin Temperature: Very low quality evidence (2 studies) indicates no effect on skin temperature. No pooled analysis was possible.

• Outcome – Skin Conductance: Very low quality evidence (1 study) suggests a significant increase in skin conductance compared with sham.

• Outcomes – Skin Temperature, Blood Flow: Very low quality evidence (1 study) indicates a significant increase or decrease in skin temperature and blood flow compared to sham.

• Outcomes – Heart Rate/Blood Pressure: Very low quality evidence (1 study) suggest an increase in heart rate and blood pressure when compared with sham.

• Outcome – Heart Rate Variability: Low quality evidence (4 studies) indicated no effect on heart rate variability. • Outcome – Heart Rate: Very low quality evidence (3 studies) found no effect on heart rate immediately after treatment (mean difference − 1.67 bpm, 95% CI -5.33 to 1.98, I2 = 1%, p = 0.37, 3 studies).

• Outcome – Blood Pressure: Very low quality evidence (1 study) found no effect on blood pressure immediately, 10 min and 24 hr after treatment.

• Outcome: Pupil Diameter: Low quality evidence (1 study) found no acute effect on pupillary control within 5 min of treatment.

• Outcome – Plasma concentrations of epinephrine and norepinephrine: Low quality evidence (1 study) found no effect on sympathoadrenal activity.

• Outcome – Oxy-hemoglobin concentration: Very low quality evidence (1 study) found no effect on muscle sympathetic nerve activity immediately, 5 min or 30 mins after treatment.

The authors conclude that one type of joint manipulative therapy – mobilizations with oscillatory movements – probably produce effects on skin sympathetic nerve activity. This finding, however, has limited clinical relevance. The remaining interventions (SNAGs, HVLA) have no acute effect on the studied markers of ANS activity. They suggest that future research consider findings over longer follow-up periods and in patients with chronic pain. It is worth remembering that this study included ONLY studies that employed a sham control comparison. This is the best way to study these effects, but as a result some papers wouldn’t have made ‘the cut’.

COMMENT: This is an area where unfortunately, the claims of some practitioners are outpacing our science. This relationship is interesting and the fact is, some clinical outcomes cannot be explained at this point via the existing research. There is much more work to do and I think this will be a burgeoning area in coming years.... The main thing to address, in my opinion, is the length of treatment intervention in these studies. (Many papers in this area employ only one, or just a few, treatments.) Changes in the ANS likely take time to manifest and then logically would take time to reverse. We simply need more research in this area....

DR. SHAWN THISTLE is a practising chiropractor, educator, international speaker, knowledge-transfer leader, evidence-based health care advocate, entrepreneur and medicolegal consultant. He founded RRS Education in 2006 and currently acts as the company’s CEO. RRS Education helps chiropractors and other manual medicine clinicians around the world integrate research into patient care via weekly Research Reviews, Online Courses and Seminars. Rrseducation.com