Introduction
-
In relation to sham controlled studies, i.e. ‘ effect’ studies:1 -Is there an effect of SM on ‘brain function’?2 -If there is an effect, for how long does it last?3 -If there is an effect, is it associated with any clinical benefits?
-
In relation to other controlled studies (‘inactive control’ or ‘another physical stimulus’), i.e. the ‘ differences in outcome’ studies:4 -Is there a difference in ‘brain function’ after SM vs. an ‘inactive control’?5 -Is there a difference in ‘brain function’ after SM vs. ‘another physical stimulus’?
Methods
Search for literature
Eligibility criteria
Screening
Extraction of information
Descriptive information
Information related to methodological quality
Information related to the results
Classifying articles by their methodological quality
Data analysis and synthesis
Results
Description of studies (n = 18)
1st author Yr Ref | Design | Type of study subjects | Number of study subjects (males/females) | -Age (range) -Mean | -Type of spinal manipulation -Type of control -Sham | How was cerebral activity measured? | When was cerebral activity measured? | Clinical outcomes assessed (measurement tool and time of assessment) |
---|---|---|---|---|---|---|---|---|
Spars 2017 [9] | Randomized controlled trial | Symptomaic: volunteers (unknown origin) with mechanical neck pain < 6 weeks of duration | 24 (4 / 8) manipulation group / (4 / 8) sham group | -? /? -36 manipulation group / 40 sham group | -HVLA midthoracic (X1) -‘No’ control -Sham: similar positioning of the subject and investigator’s hands which were placed across the skin with minimal pressure (to mimic the HVLA procedure) | Blood oxygenation level-dependent signal (in response to noxious stimuli) | Before After: immediately | Pain intensity (11-point numerical pain rating scale) (before spinal manipulation or sham procedures and after the final fMRI) |
Lelic 2016 [14] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: volunteers (origin unknown) with recurrent spinal ache, pain or stiffness and evidence of spinal dysfunction but who did not yet sought treatment for this and pain free at the time of the study. | 19 (9 / 10) | -? -26 | -HVLA (where needed, in any spine level or sacroiliac joints, nb unknown _ may be at several levels) -‘No’ control -Sham: passive and active movements of the head, spine, and body, similar to what was done for HVLA intervention, without loading and thrust | SEP amplitudes: N30 peaks Strength of brain sources: contralateral somatosensory cortex, prefrontal cortex, cingulate cortex, and bilateral secondary somatosensory cortex | Before After: exact time unknown | None |
Baarbé 2018 [15] | Randomized controlled trial | “Subclinical neck/spinal pain”: volunteers (unknown origin) with recurrent mild neck pain and muscle tension, but minimal acute pain on the day of testing and who never sought treatment for this neck complains. | 27 (6 / 8) intervention group / (5 / 8) sham group | -18–27 intervention group / 19–24 sham group -21 (for both groups) | -HVLA cervical (X2 to 4 per subject) -‘No’ control -Sham: neck gently moved into lateral flexion and rotation in a similar manner to the actual neck manipulation, without applying the HVLA thrust | Cerebellar inhibition | Before After: exact time unclear (said to be immediately after motor acquisition task, i.e. cerebellar inhibition was re-measured about 20 min after spinal manipulation) | None |
1st author Yr Ref | Design | Type of study subjects | Number of study subjects (males/females) | -Age (range) -Mean | -Type of spinal manipulation -Type of control | How was cerebral activity measured? | When was cerebral activity measured? | Clinical outcomes assessed (measurement tool and time of assessment) |
---|---|---|---|---|---|---|---|---|
Kelly 2000 [16] | Randomized controlled trial | Healthy: volunteer chiropractic students with evidence of upper cervical “subluxation”. | 36 (9 / 9) intervention group / (11 / 7) control group | -20-37 (both groups) -24 (both groups) | -Toggle (X1) -Control: 2 min of resting | Mental rotation reaction-time task | Before After: exact time unknown | None |
Dishman 2002 [17] | Non-randomized controlled trial | Healthy: healthy college students, volunteers | 24 (? /?) (repartition in each group not reported) | -? /? -25 intervention group / 27 control group | -HVLA L5-S1 (X1) -Control: side posture positioning without lower limb flexion, truncal torque, or manual contact | MEP amplitudes | Before After: -immediately (20 to 120 s) -5 min −10 min | None |
Dishman 2008a [18] | Randomized controlled trial | Healthy: healthy chiropractic students, volunteers | 72 (21 / 5) intervention group / (15 / 8) control 1 / (14 /9) control 2 | -? (3 groups, said to be between their 20s and 30s) -? (3 groups) | -HVLA L5-S1 (X1) -Control 1: L5-S1 preloading -Control 2: side posture positioning | MEP amplitudes | Before (10 MEP recorded during 100 s) After: immediately (10 MEP recorded during 100 s) | None |
Fryer 2012 [16] | Crossover controlled trial (order of interventions randomized) | Healthy: healthy university students, volunteers | 14 (10 / 4) | -18-50 -23 | -HVLA L5-S1 (X2 to 4) -Control: bilateral side-posture positioning without truncal torque, or manual contact | MEP latencies and amplitudes Silent periods | Before After: exact time unknown (according to the Discussion approximately 10 min after) | None |
Ogura 2011 [19] | Crossover controlled trial (order of interventions “counterbalanced”) | Symptomatic: volunteers, recruited at the local university, with mechanical cervical pain and shoulder stiffness. | 12 (12 / 0) | -21–40 -28 | -Instrumentally assisted manipulation (location and nb of spinal levels adjusted unknown) -Control: 20 min of resting | Regional cerebral metabolic rate (rate of glucose consumption) | No before measurementAfter: between 35 to 55 min post- intervention or resting | -Stress Response Scale (immediately after interventions) -European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (immediately after interventions) -Pain intensity (visual analogue scale) (before and immediately after spinal manipulation, not before- after-20 min of resting) |
Inami 2017 [8] | Crossover controlled trial (order of interventions randomized) | Symptomatic: volunteers (unclear origin – probably the same as Ogura et al. 2011) with mechanical cervical pain and shoulder stiffness. | 21 (21 / 0) | -? -26 | -Instrumentally assisted manipulation (where needed, anywhere at the spine, sacroiliac joints and/or scapulae, mean of 8 per subject) -Control: 20 min of resting | Regional cerebral metabolic rate (rate of glucose consumption) | No before measurement After: between 35 min to 1.05 h. post-intervention or resting | Pain intensity (visual analogue scale) (before and immediately after spinal manipulation, and before and after 20 min of resting, only for 9/21 subjects) |
Haavik-Taylor 2007aa [20] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: volunteers (unknown origin) with a history of recurring neck pain or stiffness and with evidence of cervical spinal dysfunction, pain free at the time of the study. | 13 (5 / 8) | -22-45 -31 | -HVLA cervical (X2 to 3 per subject) -Control 1: passive head movement without loading and thrust -Control 2: nothing | MEP amplitudes CSP durations | Before After: -within 0–10 min -within 10–20 min -within 20–30 min | None |
Haavik-Taylor 2010b [21] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: student and university staff members, volunteers, with reoccurring neck problems and evidence of cervical spine dysfunction, pain free at the time of the study. | 11 (4 / 7) | -22-40 -29 | -HVLA cervical (nb unknown, may be at several levels) + 20 min of typing task -Control: 20 min of typing task only | SEP MU/M + U peak ratios: -P14-N18 complex -Parietal N20 (N20-P25 complex) -Frontal N30 (P22-N30 complex) | Before After: exact time unclear (said to be immediately after HVLA+ 20 min of typing task or after 20 min typing task only, but also said to be within 25 min post interventions, i.e. possibly within 45 min after spinal manipulation) | None |
1st author Yr Ref | Design | Type of study subjects | Number of study subjects (males/females) | -Age (range) -Mean | -Type of spinal manipulation -Type of control | How was cerebral activity measured? | When was cerebral activity measured? | Clinical outcomes assessed (measurement tool and time of assessment) |
---|---|---|---|---|---|---|---|---|
Dishman 2008a [18] | Randomized controlled trial | Healthy: healthy chiropractic students, volunteers. | 72 (21 / 5) intervention group / (15 / 8) control 1 / (14 /9) control 2 | -? (3 groups, said to be between their 20s and 30s) -? (3 groups) | -HVLA L5-S1 (X1) -Control 1: L5-S1 preloading -Control 2: side posture positioning | MEP amplitudes | Before (10 MEP recorded during 100 s) After: immediately (10 MEP recorded during 100 s) | None |
Gay 2014 [22] | Randomized controlled trial | Symptomatic: volunteers from a previous clinical trial, recruited at the local university, hospital and surrounding community, who completed an exercise-injury protocol to induce myalgia in the low back. | 24 (1 / 5) manipulation group / (1 / 7) mobilization group / (5 / 5) therapeutic touch group (7 / 17) | -? /? /? (required to be between 18 and 44) −21 manipulation group / 21 mobilization group / 23 therapeutic touch group | -HVLA (X1, probably in the lumbar spine) -Control 1: grade III lumbar spinal mobilization -Control 2: therapeutic touch (light pressure with a contact to the sacroiliac joints) | Functional connectivity | Before After: immediately | Pain intensity (101-point numerical rating scale) (before and immediately after in each group) |
Haavik-Taylor 2007aa [20] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: volunteers (unknown origin) with a history of recurring neck pain or stiffness and with evidence of cervical spinal dysfunction, pain free at the time of the study. | 13 (5 / 8) | −22-45 −31 | -HVLA cervical (X2 to 3 per subject) -Control 1: passive head movement without loading and thrust -Control 2: nothing | MEP amplitudes CSP durations | Before After: -within 0–10 min -within 10–20 min -within 20–30 min | None |
Haavik-Taylor 2008 [23] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: adults (unknown origin) with a history of reoccurring neck pain or stiffness and with evidence of cervical spinal dysfunction, pain free at the time of the study. | 12 (7 / 5) | −19-45 −27 | -HVLA cervical (nb unknown, may be at several levels) -Control: passive head movement without loading and thrust | MEP amplitudes CSP durations SICI SICF | Before After: exact time unknown | None |
Haavik 2016 [24] | Crossover controlled trial | “Subclinical neck/spinal pain”: volunteers (unknown origin) with a history of spinal symptoms and with evidence of spinal and/or pelvic dysfunction but who did not yet sought treatment for this and pain free at the time of the study. | 12 (?) | -? −28 | -HVLA cervical (nb unknown, may be at several levels) -Control: passive head movement without loading and thrust | MEP amplitudes Slope of the steepest part of the curve (k) Stimulus intensity required to obtain a response that is 50% of the max (S50) | Before After: exact time unknown | None |
Haavik-Taylor 2007b [25] | Two groups “pseudo-randomized” trial | “Subclinical neck/spinal pain”: volunteers (origin unknown) with reoccurring neck problems and evidence of cervical spine dysfunction, pain free at the time of the study. | 24 (7 / 5) intervention group / (4 / 8) control group | −20-53 intervention group / 21–35 control group −30 intervention group / 27 control group | -HVLA cervical (X2 to 3 per subject) -Control: passive head movement without loading and thrust | SEP latencies and amplitudes: P14–18 complex, N20 (P14-N20 and N20-P27) and N30 (P22-N30) peaks | Before After: -within 0–10 min -within 10–20 min -within 20–30 min | None |
Haavik-Taylor 2010a [26] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: volunteers (origin unknown) with reoccurring neck problems and evidence of cervical spine dysfunction, pain free at the time of the study. | 13 (5 / 8) | −18-40 −28 | -HVLA cervical (nb unknown, may be at several levels) -Control: passive head movement without loading and thrust | SEP MU/M + U peak ratios: P14-N18 complex, N20-P25 complex, and P22-N30 complex ratios | Before After: within 25 min | None |
Niazi 2015 [27] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: volunteers (origin unknown) with recurring, intermittent low-grade spinal pain, ache, or tension, with evidence of spine dysfunction, but which did not sought treatment for this problem and are pain free at the time of the study. | 10 (10 / 0) | -? (required to be between 18 and 40) −28 | -HVLA (where needed, in any spine level or sacroiliac joints, nb unknown _ may be at several levels) -Control: passive and active movements of the subject’s head, spine, and body into the manipulation setup positions, without loading and thrust | V-wave amplitude | Before After: exact time unknown | None |
Christiansen 2018 [28] | Crossover controlled trial (order of interventions randomized) | “Subclinical neck/spinal pain”: elite Taekwondo athletes, from the Auckland area, with “subclinical spinal pain” and evidence or spine dysfunction, pain free at the time of the study. | 12 (6 / 6) | -? (required to be between 17 and 50) −25 | -HVLA (where needed, in any spine level or sacroiliac joints, nb unknown _ may be at several levels) -Control: passive and active movements of the subject’s head and spine into the manipulation setup positions, without loading and thrust | V-wave amplitude | Before After: -immediately −30 min −60 mins | None |
Data extraction (n = 18)
Data synthesis: methodological quality of the studies (n = 18)
1st Author Yr of publication Ref | -Were study subjects in sham controlled studies reported to be blind? (Yes / No / Unclear) -If yes / unclear, was the blinding tested for success? (Yes / No) -If yes, was it successful? (Yes / No) | -Were study subjects in studies with control group reported to be naive? (Yes / No / Unclear) -Was the origin of the subjects reported (Yes / No) -If yes, does it allow to exclude any interest? (Yes / No / Unclear) | Were study subjects reported to have been randomly allocated to study groups? (Yes / No / Unclear) | Were study groups comparable in relation to symptoms when studying symptomatic subjects (duration and pain intensity) (NA when cross-over study design)? (Yes / No) | Were the intervention and control(s) well described (at least where and how)? (Yes / No) | Was the assessor reported to be blind to group allocation? (Yes / No) | Were losses and exclusions of study subjects reported or obvious in result section (including in tables or graphs)? (Yes / No / Unclear) | Was the person who statistically analyzed the data reported to be blind to group allocation? (Yes / No) | Comments by the technical experts (i) on the statistical analysis, and (ii) in relation to the methodology and/or technical aspects |
---|---|---|---|---|---|---|---|---|---|
Quality score (risk of bias, also including an external validity criteria) and classification | |||||||||
Kelly 2000 [29] | -Yes (but in relation to the outcome) -Yes -No | 1: -The authors used a Student t tests to compare means instead of using a mixed-model ANOVA, followed by post-hoc tests if needed. -The authors did not study how RT (for correct answers) varied with angle, which is the main analysis conducted in the literature on such data. Therefore, without such a (usually linear) trend analysis it is not possible to understand if the overall mean effect observed by the authors is due to a change in slope (reflecting a change in processing speed) or in intercept (reflecting a change in stimulus encoding). 3: -Between-group difference pre-post significant only with one-sided t-test. -The between-group difference pre-post is not reported for the simple RT task but it seems that a contribution of the simple RT to the RT of the complex task cannot be excluded. -Unclear whether errors were also counted. | |||||||
3.5/6 (58%) medium | NA | = Unclear 0.5 pt | Yes 1 pt | NA (healthy subjects) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | Yes 1 pt | No 0 pt |
1st Author Yr of publication Ref | -Were study subjects in sham controlled studies reported to be blind? (Yes / No / Unclear) -If yes / unclear, was the blinding tested for success? (Yes / No) -If yes, was it successful? (Yes / No) | -Were study subjects in studies with control group reported to be naive? (Yes / No / Unclear) -Was the origin of the subjects reported (Yes / No) -If yes, does it allow to exclude any interest? (Yes / No / Unclear) | Were study subjects reported to have been randomly allocated to study groups? (Yes / No / Unclear) | Were study groups comparable in relation to symptoms when studying symptomatic subjects (duration and pain intensity) (NA when cross-over study design)? (Yes / No) | Were the intervention and control(s) well described (at least where and how)? (Yes / No) | Was the assessor reported to be blind to group allocation? (Yes / No) | Were losses and exclusions of study subjects reported or obvious in result section (including in tables or graphs)? (Yes / No / Unclear) | Was the person who statistically analyzed the data reported to be blind to group allocation? (Yes / No) | Comments by the technical experts (i) on the statistical analysis, and (ii) in relation to the methodology and/or technical aspects |
---|---|---|---|---|---|---|---|---|---|
Quality score (risk of bias, also including an external validity criteria) and classification | |||||||||
Dishman 2002 [17] | -No -Yes -No | 2: -MEP methodology does not correspond to standard: no motor threshold, no force control, and lack of random intervals between stimulus -The coil positioning seems not appropriate to lower leg MEPs. | |||||||
2.5/6 (42%) medium | NA | = No 0 pt | Unclear ("counterbalanced") 0.5 pt | NA (healthy subjects) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | Yes 1 pt | No 0 pt | |
Haavik-Taylor 2007a [20] | -No -No -NA | 1: -The authors stated having used planned comparisons instead of post hoc analysis in order to minimize Type 1 error. However,planned comparisons do not minimize Type1 error. -They mention running a one-way repeated measures ANOVA with the factor “intervention”. However, the degrees of freedom of the F for the result clearly show that authors treated “intervention” as a between-subjects factor, which is not correct. | |||||||
2/6 (33%) low | NA | = No 0 pt | Yes 1 pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | No 0 pt | No 0 pt | |
Dishman 2008 [18] | -No -Yes -No | 2: -MEP methodology is not standard: lack of precise motor threshold, and lack of random intervals between stimulus. -Fig. 1C indicates an inhibition in the time interval prior to SM, which may be responsible for significant differences and relative increase of amplitude after SM. | |||||||
2/6 (33%) low | NA | = No 0 pt | Yes 1 pt | NA (healthy subjects) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | No 0 pt | No 0 pt | |
Haavik-Taylor 2008 [23] | -No -No -NA | 1: -The authors mention running 2-way ANOVAs for repeated measures with the factors “muscle” and “intervention” were applied to compare the effects of SM on the two different upper limb muscles. However, the degrees of freedom of the F for the results clearly show that authors treated the two factors between-subjects, which is not correct. -They use t tests instead of post-hoc test for testing pairwise comparisons subsequent to the ANOVA. 2: The conclusions are farfetched as assumptions and deduction are made which cannot not be backed by the results. | |||||||
2/6 (33%) low | NA | = No 0 pt | Yes 1 pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | No 0 pt | No 0 pt | |
Fryer 2012 [16] | -No -Yes -No | 2: The coil positioning seems not appropriate to lower leg MEPs. | |||||||
2.5/6 (42%) medium | NA | = No 0 pt | Yes 1 pt | NA (healthy subjects) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | Unclear 0.5 pt | No 0 pt | |
Haavik 2016 [24] | -Unclear (most subjects were “novice to chiropractic”) -No -NA | 2: -The recruitment curves lack measure of variance. -Feedback from background EMG is lacking, which is a conceptual concern and could explain observed increased in amplitudes. | |||||||
2/6 (33%) low | NA | = No 0 pt | No 0 pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | Yes 1 pt | No 0 pt | |
Baarbé 2018 [15] | -Yes -No -NA | None in relation to statistics
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
3.5/6 (58%) medium | = Unclear 0.5 pt | NA | Yes 1 pt | NA (SCP subjects) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | Yes 1 pt | No 0 pt |
1st Author Yr of publication Ref | -Were study subjects in sham controlled studies reported to be blind? (Yes / No / Unclear) -If yes / unclear, was the blinding tested for success? (Yes / No) -If yes, was it successful? (Yes / No) | -Were study subjects in studies with control group reported to be naive? (Yes / No / Unclear) -Was the origin of the subjects reported (Yes / No) -If yes, does it allow to exclude any interest? (Yes / No / Unclear) | Were study subjects reported to have been randomly allocated to study groups? (Yes / No / Unclear) | Were study groups comparable in relation to symptoms when studying symptomatic subjects (duration and pain intensity) (NA when cross-over study design)? (Yes / No) | Were the intervention and control(s) well described (at least where and how)? (Yes / No) | Was the assessor reported to be blind to group allocation? (Yes / No) | Were losses and exclusions of study subjects reported or obvious in result section (including in tables or graphs)? (Yes / No / Unclear) | Was the person who statistically analyzed the data reported to be blind to group allocation? (Yes / No) | Comments by the technical experts (i) on the statistical analysis, and (ii) in relation to the methodology and/or technical aspects |
---|---|---|---|---|---|---|---|---|---|
Quality score (risk of bias, also including an external validity criteria) and classification | |||||||||
Haavik-Taylor 2007b [25] | -No -No -NA | 1: -No report of the testing of the normality of the data distribution. -To minimize Type 1 error, post hoc tests would be appropriate (instead of planned comparisons). -No between group comparison was performed.
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
2/6 (33%) low | NA | =No 0 pt | Unclear ("pseudorandomized") 0.5pt | NA (SCP subjects) | -Yes 0.5pt -Yes 0.5pt | No (but data were coded by an independent person to reduce any bias during analysis) 0.5pt | No 0 pt | No 0 pt | |
Haavik-Taylor 2010a [26] | -No -No -NA | 1: Both parametric and nonparametric results on the same data are reported. Usually, either data are normally distributed and parametric tests can be used or data are not normally distributed and non-parametric tests must be used.
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
2.5/6 (42%) medium | NA | = No 0 pt | Yes 1pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No (idem Haavik-Taylor 2007a) 0.5pt | No 0 pt | No 0 pt | |
Haavik-Taylor 2010b [21] | -No -Yes -Unclear (“students and staff population at the University of Auckland”) | 1: See comments in relation to Haavik-Taylor 2010a
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
2.5/6 (42%) medium | NA | = No 0 pt | Yes 1 pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No (idem Haavik-Taylor 2007b) 0.5 pt | No 0 pt | No 0 pt | |
Lelic 2016 [14] | -Unclear (said to be naïve) -Yes -No (sham intervention was discovered as such by most of the subjects) | 1: Unusual reporting of statistics: no report of which were the experimental factors and how they were treated (but probably pre/post was treated within-subjects and interventions as between-subjects), and of the detailed results for the F tests of the ANOVA.
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
2.5/6 (42%) medium | = No 0 pt | NA | Yes 1 pt | NA (SCP subjects / cross-over) | -Yes 0.5 pt -No 0 pt | No 0 pt | Yes 1 pt | No 0 pt |
1st Author Yr of publication Ref | -Were study subjects in sham controlled studies reported to be blind? (Yes / No / Unclear) -If yes / unclear, was the blinding tested for success? (Yes / No) -If yes, was it successful? (Yes / No) | -Were study subjects in studies with control group reported to be naive? (Yes / No / Unclear) -Was the origin of the subjects reported (Yes / No) -If yes, does it allow to exclude any interest? (Yes / No / Unclear) | Were study subjects reported to have been randomly allocated to study groups? (Yes / No / Unclear) | Were study groups comparable in relation to symptoms when studying symptomatic subjects (duration and pain intensity) (NA when crossover study design)? (Yes / No) | Were the intervention and control(s) well described (at least where and how)? (Yes / No) | Was the assessor reported to be blind to group allocation? (Yes / No) | Were losses and exclusions of study subjects reported or obvious in result section (including in tables or graphs)? (Yes / No / Unclear) | Was the person who statistically analyzed the data reported to be blind to group allocation? (Yes / No) | Comments by the technical experts (i) on the statistical analysis, and (ii) in relation to the methodology and/or technical aspects |
---|---|---|---|---|---|---|---|---|---|
Quality score (risk of bias, also including an external validity criteria) and classification | |||||||||
Ogura 2011 [19] | -No -Yes -Unclear (recruited on the campus of Tohoku University) | 1: The extent the threshold for the voxel cluster size was defined as “10 to 50 voxels minimum”. The purpose of this varying threshold is unclear. 3: Lenient statistical threshold: Z = 3, extent threshold; 10 voxels. | |||||||
2/6 (33%) low | NA | = No 0 pt | Unclear (“counterbalanced”) 0.5 pt | NA (cross-over) | -No 0 pt -Yes 0.5 pt | No 0 pt | Yes 1 pt | No 0 pt | |
Inami 2017 [8] | -No -No -NA | 1: The phrasing “(e.g., 10 voxels minimum)” suggests again (see the comment in relation to Ogura 2011) that this threshold was not fixed. | |||||||
2/6 (33%) low | NA | = No 0 pt | Yes 1 pt | NA (cross-over) | -Yes 0.5 pt -Yes 0.5 pt | No 0 pt | No 0 pt | No 0 pt | |
Gay 2014 [22] | -No -No -Unclear (recruited from the campus of the University of Florida and UF Health Hospital and the local community) | 1: -Authors “corrected for the number of separate RM-ANOVAs conducted across the 120 ROI-to-ROI pairs by using a p value less than .01 as significant.” (p.618). This threshold (p = 0.01) correction for multiple comparisons is not conservative enough. -There was neither between-groups statistical test at “pre”, nor at “post”. 3: Lenient statistical threshold: p = 0.01 with 120 comparisons. | |||||||
5/7 (71%) acceptable | NA | = No 0 pt | Yes 1 pt | Yes 1 pt | Yes 0.5 pt -Yes 0.5 pt | Yes 1 pt | Yes 1 pt | No 0 pt | |
Sparks 2017 [9] | -Yes -No -NA | 1: The authors used an alpha = 0.01 threshold for the fMRI analysis. It is not conservative enough in my opinion (as discussed by Eklund et al. 2015, and Lieberman & Cunningham 2009). 3: -Unclear whether statistical threshold applied across the whole brain or just for the region of interest. -It is unclear how the region of interest was defined | |||||||
5.5/7 (79%) acceptable | = Unclear 0.5 pt | NA | Yes 1 pt | Yes 1 pt | -Yes 0.5 pt -Yes 0.5 pt | Yes 1 pt | Yes 1 pt | No 0 pt |
1st Author Yr of publication Ref | -Were study subjects in sham controlled studies reported to be blind? (Yes / No / Unclear) -If yes / unclear, was the blinding tested for success? (Yes / No) -If yes, was it successful? (Yes / No) | -Were study subjects in studies with control group reported to be naive? (Yes / No / Unclear) -Was the origin of the subjects reported (Yes / No) -If yes, does it allow to exclude any interest? (Yes / No / Unclear) | Were study subjects reported to have been randomly allocated to study groups? (Yes / No / Unclear) | Were study groups comparable in relation to symptoms when studying symptomatic subjects (duration and pain intensity) (NA when cross-over study design)? (Yes / No) | Were the intervention and control(s) well described (at least where and how)? (Yes / No) | Was the assessor reported to be blind to group allocation? (Yes / No) | Were losses and exclusions of study subjects reported or obvious in result section (including in tables or graphs)? (Yes / No / Unclear) | Was the person who statistically analyzed the data reported to be blind to group allocation? (Yes / No) | Comments by the technical experts (i) on the statistical analysis, and (ii) in relation to the methodology and/or technical aspects |
---|---|---|---|---|---|---|---|---|---|
Quality score (risk of bias, also including an external validity criteria) and classification | |||||||||
Niazi 2015 [27] | -No -No -NA | None in relation to statistics
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
2.5/6 (42%) medium | NA | = No 0 pt | Yes 1 pt | NA (SCP subjects) | -Yes 0.5 pt -No 0 pt | No 0 pt | Yes 1 pt | No 0 pt | |
Christiansen 2018 [28] | -No -Yes -Yes | None in relation to statistics
No expert was available in relation to the technical aspects of this outcome measure
| |||||||
5/6 (83%) acceptable | NA | = Unclear 0.5 pt | Yes 1 pt | NA (SCP subjects) | -Yes 0.5 pt -No 0 pt | Yes 1 pt | Yes 1 pt | Yes 1 pt |
Data synthesis: answers to research questions (n = 13)
Sham controlled studies (Table 9), i.e. ‘effect’ studies (n = 3)
1st Author Year Ref | Type of study subjects | Outcome variable | Was a statistically significant difference between groups observed? | Was there a relationship between brain changes and any clinical outcome? | Time of assessment | Quality classification |
---|---|---|---|---|---|---|
Sparks 2017 [9] | Symptomatic (mechanical neck pain < of 6 weeks of duration) | Blood oxygenation-level dependent signal (in response to noxious stimuli) | Yes (p < .05) Statistically significant increase of activation in the insular and sensorimotor cortices post-SM compared to control; and in the anterior and posterior cingulate, supplementary motor area, and precentral gyrus post-control compared to SM | Pain intensity assessed but no relationship tested | Immediately after | Acceptable |
Lelic 2016 [14] | “Subclinical neck/spinal pain” | N30 somatosensory evoked potential peak amplitudes | Yes (significant post-intervention difference between-groups reported but without inclusion of the corresponding p-value and mention of the statistical threshold for significance) Statistically significant decrease post-SM (p = .02) but no statistically significant changes post-control (p = .4) | No clinical outcome included | Not reported | Medium |
Baarbé 2018 [15] | Cerebellar inhibition | Yes (p < .001) Statistically significant reduce post-SM compared to control | No clinical outcome included | Unclear (according to Fig. 1 immediately after the motor acquisition task, i.e. about 20 min after intervention) | Medium |
1st Author Year Ref | Type of study subjects | Outcome | Was a statistically significant difference between groups observed? | Time of assessment | Quality classification |
---|---|---|---|---|---|
Kelly 2000 [29] | Healthy | Reaction-time to a mental rotation task | Yes (p < .05) Statistically significant decrease post-SM compared to control | Unknown | Medium |
Dishman 2002 [17] | MEP amplitudes | Yes (p < .05) Statistically significant increase from 20 to 120 s. post-SM compared to control | -Immediately after (each 20s during 120 s after SM or control) − 5 min − 10 min | Medium | |
Dishman 2008 [18] | MEP amplitudes | Yes (p < .05) Statistically significant increase at 10 s. post-SM compared to control | Immediately after (each 10s during 100 s after SM or control) | Low | |
Fryer 2012 [16] | MEP amplitudes | Yes (p = .04) Statistically significant decrease post-SM compared to control | Unknown (approximately 10 min after one intervention or the other) | Medium | |
MEP latencies | No | ||||
CSP durations | No | ||||
Ogura 2011 [19] | Symptomatic (mechanical neck pain and shoulder stiffness) | Regional cerebral metabolic rate | Yes (p < .001) Statistically significant increase post-SM compared to control in the inferior prefrontal cortex, anterior cingulate cortex, and middle temporal gyrus; and statistically significant decrease post-SM compared to control in the cerebellar vermis and visual association cortex | Between 35 to 55 min | Low |
Inami 2017 [8] | Regional cerebral metabolic rate | Yes (p < .05) Statistically significant increase post-SM compared to control in the Broca’s area, anterior cingulate cortex, somatosensory association cortex, Wernike’s area, visual association cortex, cerebellar vermis, and visual cortex; and statistically significant decrease post-SM compared to control in the inferior parietal lobule, frontal pole, inferior frontal gyrus, pars triangularis, premotor area/supplementary motor area, primary motor cortex, frontal eye field, dorsolateral prefrontal cortex, angular gyrus, fusiform gyrus, inferior temporal gyrus, and temporal pole. | Between 35 to 65 min | Low | |
Haavik 2010b [21] | “Subclinical neck/spinal pain” | P14-N18 SEP peak ratio | No | Unclear (said to be within 25 min post-SM or control, possibly 45 min after one intervention or the other) | Medium |
N20-P25 SEP peak ratio | No | ||||
P22-N30 SEP peak ratio | Yes (p = .00005) Statistically significant decrease post-SM compared to control |
1st Author Year Ref | Type of study subjects | Outcome | Was a statistically significant difference between groups observed? | Time of assessment | Quality classification |
---|---|---|---|---|---|
Dishman 2008 [18] | Healthy | MEP amplitudes | Yes (p < 0.05) Statistically significant increase at 10 s. post-SM compared to control | Immediately after (each 10 s. during 100 s. after SM or control) | Low |
Haavik 2010a [26] | “Subclinical neck/spinal pain” | P14-N18 SEP peak ratio | No | Unclear (said to be within 25 min post-SM or control) | Medium |
N20-P25 SEP peak ratio | No | ||||
P22-N30 SEP peak ratio | Yes (p = .003) Statistically significant decrease post-SM compared to control | ||||
Haavik 2016 [24] | MEP amplitudes | Yes (p = .01) Statistically significant increase post-SM compared to control | Not-reported | Low | |
k (slope of the steepest part of the curve) | No | ||||
S50 (stimulus intensity to obtain a response 50% of the maximum) | No | ||||
Christiansen 2018 [28] | V-wave amplitudes | Yes (p < 0.01–0.03) Statistically significant increase at each time point post-SM compared to control | -Immediately after −30 min after − 60 min after | Acceptable |
Summary of finding in relation to the research questions 1–3
1 - Is there an effect of SM on ‘brain function’? (n = 3)
2 - If there is an effect, for how long does it last? (n = 3)
3- If there is an effect, is it associated with any clinical benefits? (n = 3)
4 - Is there a difference in ‘brain function’ after SM vs. an ‘inactive control’? (n = 7)
Type of study | First author / Year [ref] | Scorea (risk of bias and external validity) | Quality classification |
---|---|---|---|
Sham studies | Sparks, 2017 [9] | 5.5/7 (79%) | acceptable |
Baarbéé, 2018 [15] | 3.5/6 (58%) | medium | |
Lelic, 2016 [14] | 2.5/6 (42%) | medium | |
Comparison studies | Christiansen, 2018 [28] | 5/6 (83%) | acceptable |
Gay, 2014 [22] | 5/7 (71%) | acceptable | |
Kelly, 2000 [29] | 3.5/6 (58%) | medium | |
Dishman, 2002 [17] Haavik-Taylor, 2010a [26] Haavik-Taylor, 2010b [21] Fryer, 2012 [16] Niazi, 2015 [27] | 2.5/6 (42%) | medium | |
Haavik-Taylor, 2007a [20] Haavik-Taylor, 2007b [25] Haavik-Taylor, 2008 [28] Dishman, 2008 [23] Ogura, 2011 [19] HaaviK, 2016 [24] Inami, 2017 [8] | 2/6 (33%) | low |
5 - Is there a difference in ‘brain function’ after SM vs. ‘another physical stimulus’? (n = 4)
Discussion
Summary of findings and their interpretation
Methodological considerations of our own review
Methodological considerations of the included studies
Conceptual concerns
In relation to all the studies
In relation to the studies using “subclinical neck/spinal pain” subjects
Gap between scientific level of evidence and its implementation in clinical practice
Gap between scientific level of evidence and its modification for popularization
Recommendations
Recommendations for future research
-
The clinical relevance of any brain changes should be investigated using symptomatic study subjects. Also, clinical outcomes should be included and the correlation between these and brain changes should be tested to establish if there is some type of benefits.
-
To study the specific effect of SM, proper sham procedures must be adopted and checked for success after the intervention to control for any unspecific effects, including placebo responses.
-
Appropriate methodology in relation to randomized controlled trials, with appropriate attention to the potential sources of bias (e.g. blinding of study subjects, assessor, and statistician) should be respected.
-
In relation to the technical procedures, standard protocols should be employed to ensure reproducibility of the outcome measures.
-
Appropriate statistical methods and thresholds should be used.
-
Any conflict of interest should be reported.
-
Results should be replicated by independent research teams before their clinical acceptance.