The effects of spinal manipulative therapy on lower limb neurodynamic test outcomesin adults: a systematic reviewMaxwell, Christina Melanie; Lauchlan, Douglas Thomas; Dall, Philippa Margaret

Published in:Journal of Manual and Manipulative Therapy

DOI:10.1080/10669817.2019.1569300

Publication date:2020

Document VersionAuthor accepted manuscript

Link to publication in ResearchOnline

Citation for published version (Harvard):Maxwell, CM, Lauchlan, DT & Dall, PM 2020, 'The effects of spinal manipulative therapy on lower limbneurodynamic test outcomes in adults: a systematic review', Journal of Manual and Manipulative Therapy, vol.28, no. 1, pp. 4-14. https://doi.org/10.1080/10669817.2019.1569300

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Take down policyIf you believe that this document breaches copyright please view our takedown policy at https://edshare.gcu.ac.uk/id/eprint/5179 for detailsof how to contact us.

Download date: 11. Jun. 2022

1

The effects of spinal manipulative therapy on lower limb neurodynamic test outcomes 1

in adults: a systematic review. 2

3

Authors: 4

Christina Melanie Maxwell1, Douglas Thomas Lauchlan

1, Philippa Margaret Dall

1 5

6

1 School of Health & Life Sciences, Glasgow Caledonian University, Cowcaddens Road, 7

Glasgow, G4 0BA, United Kingdom. 8

9

Corresponding Author: 10

Philippa Dall 11

School of Health & Life Sciences 12

Glasgow Caledonian University 13

Cowcaddens Road, 14

Glasgow, G4 0BA 15

UK 16

philippa.dall@gcu.ac.uk 17

18

19

20

21

2

Abstract 1

Objective: Spinal Manipulative Therapy (SMT) is a routinely applied treatment modality 2

for various musculoskeletal conditions, including low back pain. The precise mechanisms 3

by which SMT elicits its effects are largely unknown, but recent research supports a multi-4

system explanation recognising both biomechanical and neurophysiological mechanisms. 5

Although the evaluation of changes in clinical presentation is complex, objective 6

neurophysiological measures of sensitivity to movement (e.g. neurodynamic tests) can be a 7

valuable clinical indicator in evaluating the effects of SMT. This review aimed to synthesise 8

current literature investigating the effects of SMT on lower limb neurodynamics. 9

Method: Eight electronic databases were systematically searched for randomised controlled 10

trials (RCT) that applied SMT (against any control) and evaluated lower limb 11

neurodynamics (Passive Straight Leg Raise or Slump Test). Selection and data extraction 12

was conducted by one researcher, reviewed by a second. Risk of bias (RoB) was assessed 13

using the Cochrane Back Review Group criteria. 14

Results: Eight RCTs were included, one with high RoB. SMT produced a clinically 15

meaningful (≥6⁰) difference in five of these studies compared with inert control, hamstring 16

stretching and as an adjunct to conventional physiotherapy, but not compared with standard 17

care, as an adjunct to home exercise and advice, or when comparing different SMT 18

techniques. Findings compared to sham were mixed. When reported, effects tentatively 19

lasted up to 6 weeks post intervention. 20

Conclusion: Limited evidence suggests SMT improved range of motion and was more 21

effective than some other interventions. Future research, using standardised Neurodynamic 22

tests, should explore technique types and evaluate longer-term effects. 23

Level of Evidence: 1a 24

25

3

Keywords 1

Low Back Pain; Passive Straight Leg Raise; Randomized Controlled Trials; Spinal 2

Mobilisation; Spinal Manipulation 3

4

4

Introduction 1

Non-specific low back pain (NSLBP) is the most common musculoskeletal condition 2

affecting the adult population, with a prevalence of up to 84%, costing the United Kingdom 3

approximately £500 million annually [1-2]. While there appears to be a consensus within 4

NSLBP clinical guidelines recommending treatment approaches such as exercise 5

prescription, there continues to be inconsistent recommendations relating to the treatment 6

efficacy of spinal manipulative therapy (SMT) [3]. SMT is a commonly applied treatment 7

for NSLBP used by a variety of health professionals, including physiotherapist, osteopaths 8

and chiropractors [4]. It is a ‘hands on’ treatment applied to the spinal column, 9

encompassing spinal manipulation and/or spinal mobilisation [5]. The continued 10

inconsistencies in guidelines and on-going debate concerning the efficacy of SMT largely 11

stems from the lack of understanding of perceived mechanisms involved in the application 12

of SMT and how exactly they mediate the experience of pain [6]. 13

14

Earlier research solely attributed the effects of SMT to be as a result of biomechanical 15

mechanisms alone [7-10]. In the last decade there has been a growing recognition of the 16

multi-systems effect of SMT, also including neurophysiological and supraspinal-mediated 17

mechanisms [11]. The effects of SMT have been shown to reduce levels of inflammatory 18

pain biomarkers [12], reduce spinal cord mediated responses [13-20]), and reduce activation 19

within brain regions associated with pain processing [21-22]. The beneficial effects of SMT 20

have also been attributed to supra-spinally mediated mechanisms such as placebo effects, 21

patient expectations, and psychosocial factors (e.g. fear) [23-25]. Therefore, in order to 22

provide further clarity to clinicians on the treatment efficacy of SMT, it is crucial that we 23

evaluate the effects of SMT on NSLBP recognising the multiple mechanisms involved. 24

25

5

To date there have been numerous systematic reviews evaluating the effects of SMT, 1

showing positive treatment effects in relation to improving pain and function in adults with 2

acute and chronic NSLBP [5,26-28] as well as asymptomatic populations with 3

experimentally induced LBP [29]. Research investigating the effects of SMT has 4

increasingly included neurophysiological outcome measures, with studies demonstrating 5

changes in skin conductance and temperature [30-33]. While these measures provide 6

valuable information, it has been suggested that there is a need to include more clinically 7

meaningful outcome measures, such as changes in neurophysiological sensitivity to 8

movement (i.e. neurodynamics [31,34]). 9

10

Heightened sensitivity to movement, is a common clinical finding within the NSLBP 11

population [35-36]. It is considered that this heightened sensitivity (due to adverse 12

neurodynamics) describes abnormal neural tissue mobility, involving both mechanical and 13

physiological mechanisms, resulting from excessive mechanical stress, and/or non-14

mechanical inputs from a peripherally and centrally up regulated nervous system [37-41]. 15

Studies evaluating the effects of SMT have demonstrated improvements in upper limb 16

neurodynamics following the application of cervical spine mobilisations in both 17

symptomatic (neck pain) and asymptomatic individuals [42]. To date, there have been no 18

reviews conducted evaluating the effects of SMT on lower limb neurodynamics. Given the 19

higher proportion of individuals reporting symptoms in the distribution of the L5 or S1 20

dermatomes [43], tests evaluating ‘posterior chain’ neurodynamics (i.e. passive straight leg 21

raise test (PSLR) or Slump tests) are of particular interest assessing the sciatic nerve and 22

associated nerve root (L4-S3) sensitivity [44-45]. Such tests form a routine component of 23

the neurological assessment of individuals with NSLBP and have been shown to have a 24

good level of reliability and validity [44,46-48]. 25

6

1

Therefore. this systematic review aims to investigate the limited literature available 2

evaluating the effect SMT on lower limb neurodynamics. Both symptomatic and 3

asymptomatic populations were included to determine if a similar magnitude of change was 4

evident between those with NSLBP and healthy population (as found in in relation to upper 5

limb neurodynamics [42]). The findings of this review will provide further insight into the 6

effects of SMT providing additional evidence to inform clinical guidelines for the treatment 7

of NSLBP. 8

9

Methods 10

The content and structure of the review were guided by the updated Cochrane Handbook for 11

Systematic Reviews of Interventions, PRISMA 2009 and PRISMA-P checklists [49-51]. All 12

screening, data extraction, and assessment of risk of bias were performed by a single 13

researcher (CM), and reviewed by a second researcher (DL), with consensus achieved 14

through discussion. Eight electronic databases were searched (until 3/2/2018) for 15

combinations of intervention and outcome measure (see Appendix A: SupplementalMaterial 16

Table 1) and reference lists were hand searched. The protocol of this review was not 17

registered a-priori. 18

19

RCTs that recruited adult participants (≥18 years, asymptomatic or symptomatic) from any 20

setting, applied SMT (mobilisation and/or manipulation) and assessed posterior chain lower 21

limb neurodynamics (passive straight leg raise (PSLR) test; Slump test) were included in the 22

review. The SMT intervention could be combined with an alternative intervention provided 23

the specific treatment effects of SMT on neurodynamics could be determined. The control 24

7

group could involve no treatment, sham, placebo or an alternative conventional treatment. 1

Articles not available with full-text in English were excluded. 2

3

Articles were screened for inclusion by title, abstract, and full-text. Study authors were 4

contacted via e-mail if missing data were identified or if the article was inaccessible in full-5

text. 6

7

The Cochrane Data Extraction Template for systematic reviews [52] was adapted following 8

pilot-testing on a randomly selected study [49] to include items in relation to SMT dosage 9

and implementation of the outcome measure(s). Risk of Bias (RoB) was determined using 10

the Updated Method Guidelines for Systematic Reviews produced by the Cochrane Back 11

Review Group [53], presented across studies in a table. Studies were rated as low RoB with 12

a score of ≥6/12, provided no serious flaws were detected. Outcomes were reported as the 13

mean difference and 95% confidence intervals of the difference between groups of the 14

change in ROM from baseline to follow-up assessment. When 95% confidence intervals 15

were not reported in the original study, these were calculated from reported mean, standard 16

deviation, and group size. When the original source only reported variability of group 17

means for baseline and follow-up separately (i.e. did not report on variability of the change), 18

pooled standard deviation was estimated assuming equal variance between groups. Between 19

group differences were interpreted as meaningful if the 95% confidence interval did not 20

include zero, and clinically meaningful if the difference in ROM was ≥6° [54-55]. 21

22

Results 23

Results of search 24

8

The search identified 1,039 articles (Figure 1), and eight RCTs were included in the review 1

[56-63]. One author provided additional detail from their thesis [64]. Due to the clinical 2

heterogeneity in the included studies a meta-analysis was not possible, and the results are 3

presented as a narrative synthesis. The included studies (Appendix B: Supplemental 4

Material Table 2) compared SMT with a range of different comparison groups (Table 1). 5

6

[Figure 1 near here] 7

[Table 1 near here] 8

9

Participants 10

Participants (aged 22 to 57 years) were mostly (n=7) recruited from private settings 11

(universities, private practices, research clinics), except one, that recruited from a secondary 12

care setting [59]. Two of the studies only recruited male participants, attempting to limit 13

variation due to gender [62-63]. Half of the studies recruited asymptomatic participants 14

[58,60-61,63]. The other studies recruited participants with back-related leg pain (90% 15

chronic [57]), degenerative lumbar disc disease at L5/S1 ± above knee radiating leg pain 16

(unknown duration [62]), intervertebral disc prolapse (L4/5 or L5/S1) with radiculopathy 17

(average >3 years, i.e. chronic [59]), or low back pain (3 weeks to 6 months, i.e. acute to 18

chronic) without signs of nerve root compression [56]. One study identified baseline 19

significant differences in PSLR ROM (SMT group higher than Sham SMT group [62]), 20

potentially overestimating treatment effects [65]. 21

22

SMT interventions 23

SMT was applied mostly by physiotherapists (n=5) [58-59,61-63], but also by chiropractors 24

(n=2) [57,60] and osteopaths (n=1) [56]. The type of SMT applied varied, including spinal 25

9

mobilisations (n=4) [58-59,61,63], spinal manipulation (n=2) [60,62] and a pragmatic 1

approach (technique and dosage chosen by practitioners, n=2 [56-57]). 2

3

Of those studies applying spinal mobilisations, two applied multilevel grade 3 posterior-4

anterior mobilisations to the L1-S1 facet joints [58,61], one applied a grade 3 posterior-5

anterior mobilisation to L4 or L5 spinous process [63], and one applied grade 1-4 rotational 6

mobilisations to L4/5 or L5/S1 [59]. When reported, oscillation rate ranged from 0.43Hz 7

[63] (below normal application rates of 1-2Hz [66]) to 2Hz [58,61]. Two studies applied 8

spinal manipulation, one involving a grade 5 pull-move technique applied to L5/S1 in a side 9

lying position [62], and the other involving either rotatory double index contact style 10

manipulation to C1 or lumbar roll position pisiform contact style manipulation to the 11

sacroiliac joint with the side of application chosen at random [60]. In the two studies 12

adopting a pragmatic approach, spinal manipulation was primarily applied. One changed to 13

mobilisations for individuals with severe LBP pain [57]. The other also applied various 14

manual therapy techniques including muscle energy, counterstrain, articulation and 15

myofascial release [56]. 16

17

Assessment of neurodynamics 18

All studies used the PSLR test to assess neurodynamics [67], but with variable protocols. 19

Only two studies [59,63] successfully met all three criteria required to determine a ‘positive’ 20

neurodynamic test [39,68-69]. There was variation in symptom reproduction (criterion 1) in 21

terms of end-point of measurement (onset of pain or resistance) and whether patient or 22

assessor determined these points. Structural differentiation, to distinguish between neural 23

and non-neural tissue dysfunction (criterion 2), was not attempted in four studies [56-24

57,60,62], and was adequately met in only two studies [59,63]. Unless this criterion has 25

10

been met it is impossible to specifically attribute the reproduction of symptoms to neural 1

tissue dysfunction, making this is the ‘key criterion’ to determine the validity of the 2

diagnostic test [68]. Lastly, inter-limb asymmetry (criterion 3) was reported as being 3

assessed in two studies [59,63], but its value was only reported in one of these [63]. It was 4

not reported on, but could be calculated from, data provided in a third study (baseline 5

between 1° and 2.4° [57]), but this was below that considered normal (i.e. <11° [70]). 6

7

Risk of bias 8

All the included studies scored above 6/12 (range 6/12 to 11/12; Table 2), suggesting an 9

overall low RoB. The main methodological weaknesses were lack of participant (n=7) and 10

care provider (n=8) blinding. However, one study was downgraded from low to high RoB 11

due to a large (41.5%) and poorly reported (no explanation, no separation by group) dropout 12

rate within the study (considered a serious flaw [59]), suggesting a strong likelihood of 13

attrition bias in this study. 14

15

[Table 2 near here] 16

17

Effects of SMT on neurodynamics 18

Between group differences (mean [95% confidence intervals]) in change in PSLR ROM 19

from baseline to follow-up assessment are presented in Table 3 (see Appendix C: 20

Supplemental Material Table 3 for within group changes). 21

22

[Table 3 near here] 23

24

SMT versus inert interventions 25

11

All of the four studies that compared SMT at a location in the lower thoracic and 1

lumbosacral region with an inert intervention (control) [58,60-61,63], indicated that change 2

in PSLR ROM from baseline to immediately post-test was higher in the SMT group than the 3

control group. Although the mean difference between groups was clinically relevant in two 4

of those studies [61,63], the lower bounds of the 95% confidence intervals were generally 5

less than 6, indicating that a clinically relevant difference might not always be expected. In 6

both studies with a longer-term follow-up, the between group difference was marginally 7

larger at follow up than immediately post-test (at 24-hours [58] and at 48 hours [63]). The 8

single study that compared SMT at the cervical spine to an inert control indicated no 9

between group difference immediately post intervention [60]. 10

11

SMT versus sham SMT 12

Two studies compared SMT to sham, with mixed findings [62-63]. One study found a 13

clinically relevant between-group improvement immediately post-test in favour of SMT 14

[62]. In contrast, the other study [63] found no difference between groups immediately post-15

test, or at 48 hour follow up for the neurodynamic test when performed with a structural 16

differentiation. There was a difference between groups at 48 hour follow-up, when the 17

PSRL test was performed without the structured differentiation. 18

19

SMT versus other interventions 20

There was a difference between groups in favour of SMT immediately post intervention in 21

one study comparing SMT to stretching [61]. The mean difference between groups was 22

clinically relevant, but the lower bound of the 95% CI was only 3. In contrast, a study 23

comparing SMT with standard care, there was no evidence of a difference between groups at 24

12 weeks post intervention [56]. 25

12

1

SMT as an adjunct to an intervention versus that intervention alone 2

In a study comparing Home Exercise and Advice (HEA) alone to HEA and SMT [57], there 3

were no between-group differences at 12 weeks post intervention. In contrast, a study with a 4

high RoB [59] reported a clinically relevant between-group difference in favour of adding 5

SMT to conventional physiotherapy compared with CP alone immediately post-intervention 6

and at 6-weeks post-intervention. 7

8

SMT versus another SMT technique 9

Immediately post-intervention, there was a small difference in lower limb (PSLR) ROM 10

between groups receiving SMT at different locations in favour of the inferior location 11

(sacroiliac joint compared to the cervical spine), however this difference was small, and not 12

clinically meaningful [60]. 13

14

15

Discussion 16

Compared with other groups, SMT was never found to be less effective at increasing PSLR 17

ROM, although it was not always demonstrated to be more effective. SMT was more 18

effective than inert interventions (e.g. prone lying [58,60-61,63]), static hamstring stretching 19

[61], and when used as an adjunct to conventional physiotherapy [59]. There were no 20

meaningful between group differences when SMT was compared to standard care [56] or 21

when used as an adjunct to home exercise and advice, both of which were assessed only at 22

12 weeks post-intervention [57]. Comparisons of SMT to sham interventions were mixed, 23

with one study reporting no difference [63] and one reporting SMT to be more effective 24

[62]. 25

13

1

Most of the six studies that evaluated ROM immediately post-SMT, demonstrated that SMT 2

performed better than each comparator group, although differences could be small (1.5° to 3

15°). However, it is the potential maintenance of meaningful therapeutic effects over a 4

longer period (≥24 hours) that is of greater clinical relevance. In the three studies that 5

assessed changes at two time-intervals (immediately post-test and up to 6 weeks) [58-59,63], 6

the difference between groups was larger at the longer-term follow up. In contrast, both of 7

the studies that assessed effect at 12 weeks showed no difference between groups [56-57]. 8

As neither of these studies assessed effect immediately post-intervention, it is impossible to 9

identify whether there was a trajectory of improvement that had subsided by 12 weeks post-10

intervention, or whether there was never a difference between groups. The hypoalgesic 11

effects of SMT have previously been shown to typically last 5 minutes or less, with limited 12

evidence showing effects lasting up to 24 hours [31,71]. From this review, it can be 13

suggested that the positive effects of SMT on PSLR ROM could be sustained in the longer-14

term (tentatively up to 6 weeks [59]). However, such inferences are limited by the small 15

number of studies at each follow-up time-point, with further research required to explore 16

how long after the application of SMT meaningful improvements in PSLR ROM last. 17

18

Half of the studies included in the review recruited asymptomatic participants [58,60-61,63]. 19

A review investigating the effects of SMT on upper limb neurodynamics observed a similar 20

magnitude of improvement between symptomatic and asymptomatic populations [42]. In the 21

current review, a clear trend could not be observed, due to how studies with asymptomatic 22

and symptomatic populations were grouped. All the studies comparing SMT to inert 23

interventions [58,60-61,63], and the study comparing types of SMT intervention [60], only 24

included asymptomatic populations. Whereas, all the studies comparing SMT as an adjunct 25

14

[57,59] only included symptomatic participants. Two studies compared SMT to a sham 1

intervention, with the study including symptomatic participants [62] demonstrating a larger 2

magnitude of change compared to that including asymptomatic participants [63]. However, 3

these treatment effects could be overestimated given significant baseline between-group 4

differences in PSLR ROM [62]. In contrast, the two studies comparing SMT to alternative 5

interventions demonstrated a larger magnitude of change in the study including 6

asymptomatic participants [61] compared with the study including symptomatic participants 7

[56]. Therefore, due to the heterogeneity of comparator groups, it is not possible to make 8

meaningful inferences about symptomatic and asymptomatic individuals. 9

10

There is mounting evidence advocating the need to classify patients into clinically relevant 11

subgroups using prediction rules (i.e. criteria that may predict a more successful outcome) 12

[72-73]. Within this review, symptomatic participants were widely heterogeneous in 13

diagnosis and considered mechanism for pain. Given the lack of consensus and 14

standardisation in the methodologies used, it is impossible to conclude to what extent, 15

criteria such as symptom duration or location of symptoms impact on the magnitude and 16

duration of changes noted in PSLR ROM following SMT. Future exploration of SMT 17

should recruit more clinically relevant subgroups of participants to develop understanding of 18

what criteria may optimise the therapeutic effects of SMT. 19

20

Previous research studies have demonstrated that mobilisation technique [74] and oscillation 21

rates [75] have little impact on treatment efficacy, and that a non-specific level mobilisation 22

is preferable in the lumbar spine [76]. Within the current review, there was considerable 23

variability in technique and location of spinal mobilisation applied and comparator groups 24

included. Consequently, it is impossible to make any specific recommendations on the 25

15

superiority of any particular spinal mobilisation approach [58-59,61,63]. Further 1

investigation is clearly required to determine whether optimal parameters exist in relation to 2

the application of this style of technique, location of application and dosage (i.e. grade, 3

speed etc.) of spinal mobilisations and what impact this has, if any, on the magnitude and/or 4

duration of treatment effects. In relation to spinal manipulation, research studies have 5

documented superior treatment effects associated side-specific treatment responses in the 6

lumbar spine [77]. In contrast, both side-specific [78] and bilateral responses [79-80] have 7

been identified following cervical spine manipulation. Only two of the included studies 8

provided information on the spinal manipulation applied, and as these were compared to 9

different comparator groups it was impossible to draw any meaningful conclusions from this 10

to contribute to the current knowledge base. 11

12

None of the selected studies used the Slump test, despite it being shown to have a higher 13

sensitivity than the PSLR test [81]. PSLR protocols varied between studies with only two of 14

the included studies adhering to all three PSLR test criteria [59,63]. Consequently, one 15

cannot confidently attribute the improvements in ROM observed within this review 16

specifically to neurodynamics. These findings highlight a definite need for consensus on the 17

implementation and interpretation of such tests for application in future research. 18

19

Of the studies included in a recent review evaluating the effects of SMT on upper limb 20

neurodynamics [42], statistically significant superiority was observed when SMT was 21

compared to sham, placebo and control [42]. In contrast, within the current review, although 22

a superior magnitude of change was observed when SMT was compared to all comparator 23

groups, only three studies demonstrated 95% confidence intervals with a lower boundary 24

that exceeded the minimum threshold for clinical significance (i.e. 6°) [59,62-63], 25

16

questioning whether positive trends such as these could be reproduced or generalised to the 1

wider population. The review by Chu and colleagues [42] shares some of the current 2

review’s limitations, including lack of application of positive test criteria (i.e. none applied 3

structural differentiation), and variability in SMT application (i.e. different location, 4

duration, side of application). 5

6

Although the overall quality of studies was high, given the methodological limitations and 7

diversity of the included literature, a meta-analysis could not be conducted [82]. Small 8

sample sizes and unreported measures of variance increased the likelihood of sampling bias 9

in non-random/convenience samples [58,60,63]. Standard care comparator groups [56,59] 10

did not reflect current best practice [83]. Adverse events of treatment were only documented 11

in one study [57]. However, small sample sizes and lack of follow up make adverse events 12

difficult to estimate accurately [84]. Without an evaluation of associated benefits and 13

adverse effects, it is difficult to draw balanced conclusions about the clinical utility of SMT. 14

Recruitment of participants from mostly private settings, and studies including only male 15

participants [62-63], limit the generalisability of the review to the wider health care context. 16

Although thorough, it is possible that relevant literature was not identified during the search. 17

Three studies were excluded as they were unavailable in English that may have led to bias 18

[85-87]. 19

20

It is recommended that future studies provide full and transparent documentation of results, 21

especially with regard to within- and between-group statistical calculations, calculations of 22

data variance and also reporting of adverse events. A full description of the SMT technique 23

should be provided, including type, technique and dosage, allowing replication and 24

appropriate comparison between studies. Furthermore, neurodynamic testing procedures 25

17

need to be standardised, with the key test criteria routinely and accurately applied. 1

Furthermore, the longer-term effects of SMT need to be routinely evaluated, given that the 2

potential for such therapeutic effects to be sustained longer term is of particular interest. 3

4

Conclusion 5

SMT was more effective at increasing PSLR ROM when compared to all comparator 6

groups, however these were only clinically meaningful in five of the eight included studies 7

[58-59,61-63]. The beneficial effects of SMT persisted at a clinically significant level up to 8

6 weeks [59], suggesting that SMT could have the potential to produce lasting changes in 9

terms of sensitivity to movement. The experience of non-specific low back pain is complex 10

and multifactorial. Research evaluating treatment for back pain needs to reflect it’s 11

multidimensional nature by incorporating a variety of subjective and objective outcome 12

measures. Neurodynamic tests provide valuable objective and subjective information that 13

can be used as a clinical indicator to monitor treatment response and subsequently make 14

adaptions as required. It is important to note that due to the methodological weaknesses of 15

the included studies the improvements observed in ROM cannot be solely attributed to 16

changes in neurodynamics. Although the conclusions drawn from this review are neither 17

robust nor definitive, they highlight a real need to improve transparency and accountability 18

within the research community in terms of reporting interventions such as SMT more 19

thoroughly and applying outcome measures such as neurodynamic tests in a more 20

standardized and criteria-focused format. 21

18

Funding: 1

This research did not receive any specific grant from funding agencies in the public, 2

commercial, or not-for-profit sectors. 3

4

disclosure of interest: 5

No potential conflict of interest was reported by the authors. 6

7

Biographical Notes: 8

Christina Maxwell is an experienced Physiotherapist who specialises in the management of 9

musculoskeletal injuries and conditions. She graduated from Queen Margaret University in 10

2011 (with a BSc in Physiotherapy) and Glasgow Caledonian University in 2016 with an 11

MSc in Musculoskeletal Physiotherapy. She has worked as a research assistant for the 12

University of Limerick and has worked clinically across a range of community and hospital 13

settings. This article reports on work conducted as her Master’s dissertation. 14

15

Dougie Lauchlan is a Senior Lecturer in Physiotherapy at Glasgow Caledonian University. 16

He graduated from both Queen’s College, Glasgow in 1990 (with a BSc Physiotherapy) and 17

Glasgow Caledonian University in 2002 with a MSc Physiotherapy. He has been teaching at 18

GCU since 2005 and became a Senior Fellow of the Higher Education Academy in 2016. He 19

has worked in a variety of healthcare and sporting environments during this time and 20

continues to practice within the GCUClinic and with the Scottish Football Association. 21

22

Dr Philippa Dall is a senior research fellow at Glasgow Caledonian University. She 23

graduated from the Universities of Newcastle-upon-Tyne (with BSc in physics with medial 24

applications) and Strathclyde (with a PhD in bioengineering). She has extensively published 25

19

in peer review journals, and her current research interest is focused on the objective 1

measurement of physical activity and sedentary behaviour in a free-living environment. She 2

is an expert working group member for the 2018 update of the UK CMO physical activity 3

guidelines. 4

5

6

Figure captions: 7

Figure 1: Study selection flow chart (PRISMA, 2009) 8

9

10

References 11

[1] Balagué F, Mannion AF, Pellisé F, et al. Non-specific low back pain. Lancet. 12

2012;379(9814):482-91. 13

[2] National Institute for Health and Clinical Excellence. Low back pain: costing report, 14

implementing NICE guidelines [online]. London: NICE; 2009. 15

[3] Oliveira CB, Maher CG, Pinto RZ, et al. Clinical practice guidelines for the management 16

of non-specific low back pain in primary care: an updated overview. Eur Spine J. 17

2018;27(11):2791—803. 18

[4] Rubinstein SM, van Middelkoop M, Assendelft WJ, et al. Spinal manipulative therapy 19

for chronic low-back pain. Cochrane Db Syst Rev. 2011;2011(2):CD008112. 20

[5] Rubinstein SM, Terwee CB, Assendelft WJ, et al. Spinal manipulative therapy for acute 21

low-back pain. Cochrane Db Syst Rev. 2012;2012(9):CD008880. 22

[6] Bialosky JE, Bishop MD, Robinson ME, et al. The influence of expectation on spinal 23

manipulation induced hypoalgesia: An experimental study in normal subjects. BMC 24

Musculoskel Dis. 2008;9(19):1-9. 25

20

[7] Schekelle PG. Spine update: spinal manipulation. Spine. 1994;19(7):858–61. 1

[8] Herzog W, Scheele D, Conway PJ. Electromyographic responses of back and limb 2

muscles associated with spinal manipulative therapy. Spine. 1999; 24(2): 146–52. 3

[9] Maigne JY, Guillon JF. Highlighting of intervertebral movements and variations of 4

intradiscal pressure during lumbar spinal manipulation. A feasibility study. J Manip Physiol 5

Ther. 2000; 23(8): 531–5. 6

[10] Bogduk N, Jull G. The theoretical pathology of acute locked back: a basis for 7

manipulative therapy. Manual Med. 1985; 1:78–82. 8

[11] Bialosky JE, Bishop MD, Price DD, et al. The Mechanisms of Manual Therapy in the 9

treatment of musculoskeletal pain: a comprehensive model. Manual Ther. 2009; 14(5): 531-10

538. 11

[12] Degenhardt BF, Darmani NA, Johnson JC, et al. Role of osteopathic manipulative 12

treatment in altering pain biomarkers: a pilot study. J Am Osteopath Assoc. 2007; 107(9): 13

387-400. 14

[13] Colloca CJ, Keller TS, Gunzburg R. Neuromechanical characterization of in vivo 15

lumbar spinal manipulation. Part II. Neurophysiological response. J Manip Physiol Ther. 16

2003; 26(9): 579-591. 17

[14] Pickar JG, Wheeler JD. Response of muscle proprioceptors to spinal manipulative-like 18

loads in the anesthetized cat. J Manip Physiol Ther. 2001; 24(1): 2-11. 19

[15] Mohammadian P, Gonsalves A, Tsai C, et al. Area of capaicin-induced secondary 20

hyperalgesia and allodynia are reduced by a single chiropractic adjustment: a preliminary 21

study. J Manip Physiol Ther. 2004; 27(6): 381-387. 22

21

[16] Vicenzino B, Paungmali A, Buratowski S, et al. Specific manipulative therapy 1

treatment for chronic lateral epicondylalgia produces uniquely characteristic hypoalgesia. 2

Manual Ther. 2001; 6(4): 205-212. 3

[17] Malisza KL, Stroman PW, Turner A, et al. A Functional MRI of the rat lumbar spinal 4

cord involving painful stimulation and the effect of peripheral joint mobilization. J Magn 5

Reson Imaging. 2003; 18(2): 152-159. 6

[18] Dishman JD, Burke J. Spinal reflex changes after cervical and lumbar spinal 7

manipulation: a comparative study. Spine J. 2003; 3(3): 204-212. 8

[19] Grindstaff TL, Hertel J, Beazell JR, et al. Effects of lumbopelvic joint manipulation on 9

quadriceps activation and strength in healthy individuals. Manual Ther. 2009; 14(4): 415-20. 10

[20] Dunning J, Rushton A. The effects of cervical high-velocity low-amplitude thrust 11

manipulation on resting electromyographic activity of the biceps brachii muscle. Manual 12

Ther. 2009; 14(5): 508-13. 13

[21] Malisza KL, Gregorash L, Turner A, et al. Functional MRI involving painful 14

stimulation of the ankle and the effect of physiotherapy joint mobilization. Magn Reson 15

Imaging. 2003; 21(5): 489-96. 16

[22] Savva C, Giakas G, Efstathiou M. The role of descending inhibitory pain mechanism in 17

musculoskeletal pain following high velocity, low amplitude thrust manipulation: A review 18

of the literature. J Back Musculoskel Rehabil. 2014; 27(4): 377-82. 19

[23] Kaptchuk TJ. The placebo effect in alternative medicine: can the performance of a 20

healing ritual have clinic significance? Ann Intern Med. 2002; 143(11): 817-825. 21

[24] Kalauokalani D, Cherkins DC, Sherman KJ, et al. Lessons from a trial of acupuncture 22

and massage for low back pain: patient expectations and treatment effects. Spine. 2001; 23

26(13): 1418-1424. 24

22

[25] Ernst E. Does spinal manipulation have specific treatment effects? Fam Pract. 2000; 1

17(6): 554-6. 2

[26] Rubinstein SM, Terwee CB, Assendelft WJ, et al. Spinal manipulative therapy for acute 3

low back pain: an update of the Cochrane review. Spine. 2013; 38 (3): E158-77. 4

[27] Paige NM, Miake-Lye IM, Booth MS, et al. Association of Spinal Manipulative 5

Therapy with Clinical benefit and harm for Acute Low Back Pain. JAMA. 2017; 317(14): 6

1451-1460. 7

[28] Akindele-Agbeja O, Mbada CE, Egwu MO. Does the inclusion of spinal manipulative 8

therapy in multimodal treatment regimens result in better outcomes in chronic low back 9

pain? A Systematic Review. Proc Singapore Healthcare. 2017; 26 (2): 114-120. 10

[29] Millan M, Leboeuf-Yde C, Budgell B, et al. The effect of spinal manipulative therapy 11

on experimentally induced pain: a systematic literature review. Chiropr Manual Ther. 2012; 12

20, pp.1-22. 13

[30] Pickar JG, Bolton PS. Spinal manipulative therapy and somatosensory activation. J 14

Electromyog Kinesiol. 2012;22(5):785-94. 15

[31] Hegedus EJ, Goode A, Butler RJ, et al. The neurophysiological effects of a single 16

session of spinal joint mobilisation: does the effect last? J Manual Manip Ther. 2011; 19(3): 17

143-151. 18

[32] Coronado RA, Bialosky JE, Cook CE. The temporal effects of a single session of high 19

velocity, low amplitude thrust manipulation on subjects with spinal pain. Phys Ther Rev. 20

2010;15(1):29-35. 21

[33] Schmid A, Brunner F, Wright A, et al. Paradigm shift in manual therapy? Evidence for 22

a central nervous system component in the response to passive cervical joint mobilisation. 23

Manual Ther. 2008;13(5):38796. 24

23

[34] Younger J, McCue R, Mackey S. Pain outcomes: A brief review of instruments and 1

techniques. Curr Pain Headache R. 2009; 13(1): 39-43. 2

[35] Boyle KL, Demske JR. Management of a female with chronic sciatica and low back 3

pain: A case report. Physiother Theory Pract. 2009; 25(1): 44-54. 4

[36] Cleland JA, Fritz JM, Whitman JM, et al. The use of a lumbar spine manipulation 5

technique by physical therapists in patients who satisfy a clinical prediction rule: a case 6

series. J Orthop Sports Phys Ther. 2006;36(4):209-14. 7

[37] Ellis RF, Hing WA. Neural Mobilization: A Systematic Review of Randomized 8

Controlled trials with an analysis of Therapeutic Efficacy. J Manual Manip Ther. 9

2008;16(1):8-22. 10

[38] Butler DS. Letter to Editor: Neurodynamics in a broader perspective. Manual Ther. 11

2007;12(1)Le7-e8. 12

[39] Shacklock M. Clinical Neurodynamics: A new system of musculoskeletal treatment. 13

Amsterdam: Elsevier; 2005. 14

[40] Butler DS. The sensitive nervous system. Adelaide: Noigroup publications; 2000. 15

[41] Shacklock M. Neurodynamics. Physiotherapy. 1995;81(1):9-16. 16

[42] Chu J, Allen DD, Pawlowsky S, et al. Peripheral response to cervical or thoracic spinal 17

manual therapy: an evidence-based review with meta analysis. J Manual Manip Ther. 2014; 18

22(4): 220-229. 19

[43] Gore S, Nadkarni S. Sciatica: Detection and confirmation by new method. Int J Spine 20

Surg. 2014;8(15):1-47. 21

[44] Walsh J, Hall T. Agreement and Correlation between the straight leg raise and slump 22

tests in subjects with leg pain. J Manip Phys Therapeut. 2009;32(3):184-92. 23

[45] Kamath SU, Kamath SS. Lasègue’s Sign. J Clin Diagnostic Res. 2017;11(5):RG01-24

RG02. 25

24

[46] Van der Windt DAWM, Simons E, Riphagen II, et al. Physical examination for lumber 1

radiculopathy due to disc herniation in patients with low-back pain. Cochrane Db Syst Rev 2

2010;2010(2):CD007431. 3

[47] Herrington L, Bendix K, Cornwell C, et al. What is the normal response to structural 4

differentiation within the slump and straight leg raise tests? Manual Ther. 2008;13(4):289-5

94. 6

[48] Majlesi J, Togay H, Unalan H, et al. The sensitivity and specificity of the Slump and 7

the straight leg raising tests in patients with lumbar disc herniations. J Clin 8

Rheumatol.2008;14(2):87-91. 9

[49] Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions. 10

The Cochrane Collaboration; 2011. 11

[50] Moher D, Shamseer L, Clarke M, et al. Preferred reporting items for systematic review 12

and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015; 4(1): 1-9. 13

[51] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for Systematic Reviews 14

and Meta-Analyses: The PRISMA Statement. PLoS Med. 2009; 6(7): 1-6. 15

[52] Cochrane Consumers and Communication. Data Extraction Template for included 16

studies. The Cochrane Collaboration; 2015. 17

[53] Furlan AD, Pennick V, Bombardier C, et al. 2009 Updated Method Guidelines for 18

Systematic Reviews in the Cochrane Back Review Group. Spine. 2009; 34(18), pp.1929-19

1941. 20

[54] Cejudo A, Sainz de Baranda P, Ayala F, et al. Test-retest reliability of seven common 21

clinical tests for assessing lower extremity muscle flexibility in futsal and handball players. 22

Phys Ther Sport. 2015; 16(2): 107-113. 23

25

[55] Ekedahl H, Jonsson B, Frobell RB. Finger-to-floor test and straight leg raising test: 1

validity, responsiveness, and predictive value in patients with acute/subacute low back pain. 2

Arch Phys Med Rehabil. 2012; 93(12), pp.210-2215. 3

[56] Andersson GGJ, Lucente T, Davis AM, et al. A comparison of osteopathic spinal 4

manipulation with standard care for patients with low back pain. New Engl J Med. 1999; 5

341(19): 1426-1431. 6

[57] Bronfort G, Hondras MA, Schulz CA, et al. Spinal manipulation and home exercise 7

with advice for subacute and chronic back-related leg pain – A trial with adaptive allocation. 8

Ann Intern Med. 2014; 161(6): 381-391. 9

[58] Ganesh GS, Mohanty P, Pattnaik SS. The immediate and 24-hour follow up effect of 10

unilateral lumbar Z-joint mobilisation on posterior chain neurodynamics. J Bodywork Move 11

Ther. 2015; 19(2): 226-231. 12

[59] Kumar SP, Cherian PJ. Efficacy of spinal mobilization in the treatment of patients with 13

lumbar radiculopathy due to disc herniation: A randomized clinical trial. Int J Neurolo 14

Neurosurg. 20113(2): 65-76. 15

[60] Pollard HP, Ward G. The effect of upper cervical or sacroiliac manipulation on hip 16

flexion range of motion. J Manip Phys Therapeut. 1998; 21(9): 611-616. 17

[61] Szlezak AM, Georgilopoulos P, Bullock-Saxton JE, et al. The immediate effect of 18

unilateral lumbar Z-joint mobilisation on posterior chain neurodynamics: A randomised 19

controlled study. Manual Ther. 2011; 16(6): 609-613. 20

[62] Vieira-Pellenz F, Olivia-Pascal-Vaca A, Rodriguez-Blanco C, et al. Short-term 21

manipulation on pain perception, spinal mobility, and full height recovery in male subjects 22

with degenerative disk disease: a randomized controlled trial. Arch Phys Med Rehabil. 23

2014; 95(9): 1613-9. 24

26

[63] Wood L, Moran L. Investigation into the effects of a posterior-to-anterior lumbar 1

mobilisation technique on neurodynamic mobility in the lower limb. SportEX Dynamics. 2

2011; 27:8-12. 3

[64] Wood L. An investigation into the effects of a posterior-to-anterior lumbar mobilisation 4

technique on neurodynamic mobility in the lower limb [master’s thesis]. Aukland: Unitec 5

Institute of Technology; 2008. 6

[65] Roberts C, Torgerson DJ. Baseline imbalance in randomised controlled trials. Brit Med 7

J. 1999; 319(7203): 185. 8

[66] Souvlis T, Vicenzino B, Wright A. Neurophysiological effect of spinal manual therapy. 9

Grieve’s Modern Manual Therapy: The Vertebral Column. 3d ed. New York (NY): 10

Churchill Livingstone; 2004. P367-380. 11

[67] Boyd BS, Wanek L, Gray AT, et al. Mechanosensitivity of the lower extremity nervous 12

system during straight-leg raise neurodynamic testing in healthy individuals. J Orthop Sport 13

PhysTher. 2009; 39(11), pp.780-790. 14

[68] Nee RJ, Butler D. Management of peripheral neuropathic pain: Integrating 15

neurobiology, neurodynamics and clinical evidence. Phys Ther Sport. 2006; 7(1): 36-49. 16

[69] Kostopoulos D. Treatment of carpal tunnel syndrome: a review of the non-surgical 17

approaches with emphasis in neural mobilization. J Bodywork Mov Ther. 2004;8(1), pp.2-8. 18

[70] Boyd BS, Villa PS. Normal inter-limb differences during the straight leg raise 19

neurodynamic test: a cross sectional study. BMC Musculoskel Dis. 2012; 13(245): 1-9. 20

[71] Coronado RA, Bialosky JE, Cook CE. The temporal effects of a single session of high 21

velocity, low amplitude thrust manipulation on subjects with spinal pain. Phys Ther R. 2010; 22

15(1): 29-35. 23

27

[72] Herbert JJ, Stomski N, French S, et al. Serious Adverse events and spinal manipulative 1

Therapy of the Low Back region: A systematic review of cases. J Manip Physiol Therapeut. 2

2013; 38(9): 677-91. 3

[73] Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with 4

low back pain who demonstrate short-term improvement with spinal manipulation. Spine. 5

2002; 27(24): 2835-43. 6

[74] Chiradejnant A, Maher CG, Latimer J, et al. Efficacy of “therapist-selected” versus 7

“randomly selected” mobilisation techniques for the treatment of low back pain: a 8

randomised controlled trial. Aus J Physiother. 2003; 49(4): 233-41. 9

[75] Willett E, Hebron C, Krouwel O. The initial effects of different rates of lumbar 10

mobilisations on pressure pain thresholds in asymptomatic subjects. Manual Ther. 2010; 11

15(2): 173-178. 12

[76] Slaven EJ, Goode AP, Coronado RA, et al. The relative effectiveness of segment 13

specific level and non-specific level spinal joint mobilization on pain and range of motion: 14

results of a systematic review and meta-analysis. J Manual ManipTher. 2013; 21(1): 7-17. 15

[77] Perry J, Green A. An investigation into the effects of a unilateral applied lumbar 16

mobilisation technique on peripheral sympathetic nervous system activity in the lower limb. 17

Manual Ther. 2008; 13(6): 492-499 18

[78] Pikula JR. The effect of spinal manipulative therapy (SMT) on pain reduction and 19

range of motion in patients with acute unilateral neck pain: a pilot study. J Can Chiropr 20

Assoc. 1999; 43(2): 111-119. 21

[79] De Camargo VM, Alburquerque-Sendín F, Bérzin F, et al. Immediate effects on 22

electromyographical activity and pressure pain thresholds after a cervical manipulation in 23

mechanical neck pain: a randomized controlled trial. J Manip Physiol Therapeut. 2011; 24

34(4): 211-20. 25

28

[80] Fernàndez-Camero J, Fernández-de-las-Peñas C, Cleland J. Immediate hypoalgesic and 1

motor effects after a single cervical spine manipulation in subjects with lateral 2

epicondylalgia. J Manip Physiol Therapeut. 2008; 31(9): 675-681. 3

[81] Majlesi J, Togay H, Unalan H, et al. The sensitivity and specificity of the Slump and 4

the straight leg raising tests in patients with lumbar disc herniations. J Clin Rheumat. 2008; 5

14(2): 87-91. 6

[82] Garg AX, Hackam D, Tonelli M. Systematic Review and meta-analysis: When one 7

study is just not enough. ClinJ Am Soc Nephrol. 2008; 3(1): 243-260. 8

[83] National Institute for Health and Care Excellence. Low back pain in adults: early 9

management [online]. London: NICE; 2009. 10

[84] Zorzela L, Loke YK, Ioannidis JP, et al. PRISMA harms checklist: improving harms 11

reporting in systematic reviews. Brit Med J. 2016;352(i157):1-17. 12

[85] Lu LJ, Wang XD, Lu SW, et al. Case-control study on supine rotation manipulation for 13

the treatment of lumbar disc herniation, Zhongguo Gu Shang. 2012;25(8):674-7. 14

[86] Lu LJ, Jin YD, Zheng RY, et al. The observe of clinical effect of treating lumbar 15

intervertebral disc herniation by bone setting manipulation of different directions. Zhongguo 16

Gushang, 2009;22(4):255. 17

[87] Schlager S, Scöttker-Königer T. Effect of longitudinal- Cranial manipulation of the 18

Occiput/C1 on straight leg raise. Manueulle Therapie, 2012;16(5):227-35. 19

20

29

1 Types of

Comparators

Study Details of Comparator Groups

Inert

Intervention

Ganesh et al.,

2015 [58]

(2 arms)

Inert intervention: Prone lying for 3 minutes

Sham SMT Vieira-Pellenz

et al., 2014

[62]

(2 arms)

Sham: Side lying position with hips and knees

flexed for same time as intervention

Wood and

Moran, 2011

[63]

(3 arms)

Sham: Prone lying with the application of low

amplitude oscillation to lower lumbar (L4/5) skin

and soft-tissue in lateral direction.

Control: Supine lying for 3 minutes and 40

seconds

Other

interventions

Szlezak et al.,

2011 [61]

(3 arms)

Other intervention: Static Hamstring stretching

as per the PSLR protocol (ipsilateral side to

tested leg) for 3 minutes at R1

Control group: Supine lying for 3 minutes

Andersson et

al., 1999 [56]

(2 arm)

Other intervention: 8 Visits of ‘Standard Care’

involving a combination of analgesics, anti-

inflammatories, Active Physiotherapy,

Ultrasound, Diathermy, Hot/cold packs, back

corset, TENS, 10 minute educational lower

back pain video.

Adjunct

intervention

alone

Bronfort et al.,

2014 [57]

(2 arms)

Adjunct Intervention: Home Exercise and

Advice – 4 visits (1 hour each) involving

positioning, stabilization exercises with printed

instructions, postural advice, pain management

techniques, Modified back in action book.

Kumar &

Cherian, 2011

[59]

Adjunct Intervention: Conventional

Physiotherapy - 5 minutes of intermittent

lumbar traction, 15 minutes of heat, home

30

(2 arms) exercise program handouts.

Different SMT

techniques to

one another and

to a control

Pollard &

Ward, 1998

[60]

(3 arms)

Other interventions:

(1) Upper Cervical Spine Manipulation involving

a rotatory double index contact style to C1 with

the side of application chosen at random,

(2) Sacroiliac Joint Manipulation involving a

lumbar pisiform contact to the sacroiliac joint

with the side of application chosen at random.

Control: Digital pressure on mastoid process

bilaterally for 30s x 3

Table 1. Comparison Groups used in included studies. 1

2 3

31

1This study was downgraded to a high RoB due to a large and poorly reported dropout rate within the study despite scoring ≥6/12 [59] 1

2Additional information obtained from masters dissertation, resulted in the study going from high (5/12) to low RoB (11/12) [63] 2

Potential Risk of Bias sources Pollard & Ward (1998) [60]

Andersson et al. (1999) [56]

Kumar & Cherian (2011) [59]

1

Wood & Moran (2011) [63]

2

Szlezak et al. (2011) [61]

Vieira-Pellenz et al. (2014) [62]

Bronfort et al. (2014) [57]

Ganesh et al. (2015) [58]

Selection Bias 1) Random Sequence Generation

? + + + + + + +

2) Allocation concealment - + + + + - + - Performance Bias

3) Blinding of participants ? - - + - + - ?

4) Blinding of personnel ? - - - - - - -

Detection Bias

5) Blinding of outcome assessment

? + ? + + + + +

Attrition Bias Incomplete outcome data

6) Dropout rate described and acceptable

+ + - + + + + +

7) Participants analysed in allocated groups

+ + ? + + + + +

Reporting Bias

8) Selective outcome reporting - + + + + + + +

Other Bias Other Bias 9) Similarity of groups at baseline

+ + + + + + + +

10) Co-interventions avoided or similar between groups

+ + + + + + + +

11) Acceptable compliance in all groups

+ + + + + + + +

12) Similar timing of outcome assessment in all groups

+ + - + + + + +

Overall Score 7/12 10/12 6/12 11/12 10/12 10/12 10/12 9/12

32

Note: Each criterion is scored as either yes or +, denoting a low RoB, no or -, denoting a high RoB, and unclear or “?”, denoting an unclear result due to insufficient 1 information, with a total score of ≥6/12 indicating an overall low RoB [53]. 2

Table 2 – Assessment of risk of bias of included studies 3 4 5

33

Study

Intervention Comparator

Test Variation

Assessment time period

post test 24 hours 48 hours 6 weeks 12 weeks

Ganesh et al. (2015) [58]

1

SMT Inert Control 4.4

[0.1 to 8.7] 6.1

[1.8 to 10.4]

Wood & Moran (2011) [63]

2

SMT Inert Control PSLR 15

[7.2 to 22.8]

18.5 [13.1 to 23.8]

Wood & Moran (2011) [63]

2

SMT Inert Control PSLR+NF 12.9

[3.7 to 22.2]

15.6 [10.3 to 20.8]

Szlezak et al. (2011) [61]

1

SMT Inert Control 8.4

[4.4 to 12.3]

Pollard & Ward (1998) [60]

3,4

SIJ SMT Inert Control 3.4

[1.5 to 5.4]

Pollard & Ward (1998) [60]

3,4

Csp SMT Inert Control 1.5

[-0.7 to 3.6]

Vieira-Pellenz et al. (2014) [62]

2

SMT Sham 14.1

[11.4 to 16.8]

Wood & Moran (2011) [63]

2

SMT Sham PSLR 3.0

[-5.0 to 11.0]

9.4 [3.4 to 15.4]

Wood & Moran (2011) [63]

2

SMT Sham PSLR+NF 1.7

[-7.7 to 11.2]

4.6 [-2.2 to 11.4]

Szlezak et al. (2011) [61]

1

SMT Stretching 7.0

[2.9 to 11.0]

Andersson et al. (1999) [56]

SMT SC supine 1.5

[-1.5 to 4.5]

Andersson et al. (1999) [56]

SMT SC sitting 1.4

[-2.4 to 5.1]

Bronfort et al. (2014) [57]

SMT+HEA HEA alone left leg 3.4

[-0.1 to 6.8]

Bronfort et al. (2014) [57]

SMT+HEA HEA alone right leg 3.6

[-0.02 to 7.2]

Kumar & Cherian (2011) [59]

5

SMT+CP CP alone 12.5

[10.2 to 14.8]

17.5 [15.2 to 19.8]

Pollard & Ward (1998) [60]

3,4

SIJ SMT Csp SMT 2.0

[0.1 to 3.9]

Data reported as mean difference [95% confidence intervals (CI)] between groups of change from baseline. 1mean difference and 95%CIs calculated from reported 1

group means and standard deviation at each time point. 2mean difference and 95%CIs calculated from reported group mean and standard deviation of change pre 2

to post 395%CIs calculated from standard deviation of between group differences of change pre to post.

4Values originally reported as relative to vertical, direction 3

34

of difference inverted so that a positive difference is an increase in range of motion.. When conducted, pooled standard deviation was calculated assuming equal 1 variance of groups. Mean differences between groups that are greater than clinical significance (6) are marked bold. 95%CIs of the difference between groups are 2 only positive, indicating potential improvement in favour of SMT, are shaded in grey. 3 CP: Conventional Physiotherapy, Csp: Cervical spine, HEA: Home Exercise and Advice, SC: Standard Care, SMT: Spinal manipulative Therapy, SIJ: Sacroiliac 4 Joint 5 6 Table 3. Between group differences: comparing mean change in PSLR ROM at each time interval relative to baseline 7

8 9

35

Figure 1 1 2 3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20 21 22 23

24

1

Appendix A: Supplemental Material Table 1. Systematic Review Search Strategy 25

26

Databases MEDLINE, Allied and Complementary Databases (AMED),

Cumulative Index to Nursing and Allied Health Literature (CINAHL)

and Education Resources Information Center (ERIC) Via

EBSCOHost

Date 03/02/18

Strategy Step 1.

S1 (All ‘Intervention’ Key Search

Terms with Boolen Operator

“OR” searching in “TX ALL

TEXT”)

Step 2. S2 (All ‘Outcome’ Key Search

Terms with Boolen Operator

“OR” searching in “TX ALL

TEXT”)

Step 3. S1 + S2 (Apply Boolen Operator

‘AND’)

Apply Limiters: Age (All adult),

Source Type (Academic Journal,

Journals)

Search ID# Search Terms Search Options Results

S1

(Intervention

Key Search

Terms)

“Spinal Manipulative Therapy” OR “Manual Therapy”

OR “Mobilisation” OR “Mobilization” OR

“Manipulative Therapy” OR “Manipulation” OR “High

Velocity Thrust” OR “High Velocity Technique” OR

“High Velocity Low Amplitude Thrust” OR “”

207,717

S2

(Outcome

Key Search

Terms)

“Neurodynamics” OR “Neural Tension” OR “Nerve

Tension” OR “Straight Leg Raise” OR “Straight-Leg

Raise” OR “Straight leg raising” OR “Straight-leg

raising” OR “Lasègue sign” OR “Slump test” OR

“Neural Provocation Test” OR “Neuromechanic” OR

“Neurophysiology” OR “Nerve Provocation Test” OR

120,987

2

“Leg Pain” OR “Sciatica” OR “Lumbago” OR “Lower

limb pain”

S1 + S2

With

Limiters

Apply Limiters: Age (All adult), Source Type

(Academic Journal, Journals)

542

3

Appendix B: Supplemental Material Table 2. Characteristics of included studies 27 28

Author Study Design Sample Size

Characteristics of Subjects Outcome Measure(s) Intervention/Control Groups

Main Results

Ganesh et al. (2015) [58] India Physio Journal Impact factor (2014): 0.00

RCT - Prospective trial- pre, post and follow-up design Two arms: 1) Mob group (n=30) 2) Control group (n=30)

60

Asymptomatic Physiotherapist students Age: 20-24 (mean 22.03 yrs) Sex: 15 F/15 M each group Matched baseline measures (Age, Height, Weight, PSLR)

PSLR - Right limb Ax Time points measured: Pre, immediately post & 24 hrs End-point measured: P1 or R1 SD manoeuvres: None – plantargrade maintained at ankle

Intervention: Spinal mobilizations - Prone lying, PA direction - Gr 3, Frequency 2 Hz, applied to T12-S1 Z-joints - 30s per joint (total 3 mins) Provider: Unknown experience Control: Prone lying- 3 mins

- SMT group produced SS (p=0.000) differences within both groups at both post-intervention Ax intervals -CS effect at 24 hrs post-intervention for SMT group alone (7.4°) - No SS (p=0.367) between-groups effect. CS effect in favour of SMT at 24 hrs post-intervention (6°). No loss to follow-up

Bronfort et al. (2014) [57]

U.S.A.

Chiro Journal impact factor (2014): 17.810

RCT- Pragmatic trial – parallel design with allocation by minimization Two arms: 1) SMT+HEA (N=96) 2) HEA (N=96)

192 Symptomatic - Adults with BRLP for a min. of 4 wks Median duration leg pain: 104 weeks Age: ≥ 21 SMT+HEA (57.1yrs) HEA (57.7yrs) Sex: SMT + HEA–59% F HEA–68% F

PSLR -Both limbs Ax Time points measured: Baseline & 12 wks post-intervention End point measured: P1 SD manoeuvres: None

Intervention: Spinal manipulative Therapy + HEA -Pragmatic- manipulation or mobilisations Provider: Chiro min. 5 yrs -Max. 20 sessions (10-20 mins) Control: HEA – 4 visits (1hr each) –positioning, stabilization ex’s- printed, postural advice, pain-mgt techniques, modified back in action book

- SMT+HEA is a safe and effective conservative Rx resulting in better ST outcomes at 12 wks compared to HEA alone –borderline SS: (L) p=0.053, (R) p=0.051 - No within or between-group CS findings Reported 191 (99%) provided follow up data at 12 weeks

4

Vieira-Pellenz et al. (2014) [62] Brazil Physical Therapy Journal Impact factor (2014): 2.565

RCT – Double blind controlled trial Two arms: 1) Spinal Manip (n=20)

2) Sham SMT (n=20)

40

Symptomatic – Lumbar DDD- L5/S1 Mean duration of sx: unknown Age: 18-25 Control: 37yrs SMT group: 39yrs Sex: All Male

PSLR –LL with radiating pain Ax Time points measured: 3 mins pre+ 3 mins post-intervention End point measured: P1 SD manoeuvres: None

Intervention: Spinal manipulation - Gr 5- Side lying - Rotational pull-move technique to L5/S1 Provider: 8 yrs of clinical experience in MT Control: Side lying - hip + knee flexed for same time as Intervention

- SMT group produced immediate SS (p=<0.001) + CS improvements in PSLR ROM of 14.11° compared to Sham group - A SS (p=<0.001) + CS (13.65°) improvement was observed in the SMT group alone reaching CS unlike Sham group No loss to follow-up

Szlezak et al (2011) [61] Australia Physio Journal Impact factor (2014): 1.714

RCT – Single blinded random controlled design Three arms: 1) Mobs group (n=12)

2) Stretching group (n=12) 3) Control group (n=12)

36 Asymptomatic – Adult volunteers Age: 18-65 (mean 37.28yrs) Sex: 19 male/17 female Matched baseline measures (weight, height + PSLR) **Control group younger at baseline

PSLR - Unknown which LL Ax Time points measured: Immediate pre + post-intervention - average of 3 measurements documented End point measured: R1 – determined by researcher 1 SD manoeuvres: None - Device held ankle in plantargrade

Intervention: Spinal mobilisation - Prone lying applied to unilateral side to tested leg - PA to T12-S1 Z-joints - Gr 3, oscillatory 2Hz - 30s per joint (3 mins total Rx) Provider: Researcher 3 (unknown experience) Control: Supine lying for 3 mins Control: Static hamstring stretching as per the PSLR protocol (ipsilateral side to tested leg) for 3 minutes at R1

- SMT resulted in immediate CS (6.95°) superior improvement on PSLR ROM on the ipsilateral side to the intervention compared to stretching. - No between-group SS calculations provided. - CS (8.5°) + SS improvements in PSLR ROM in SMT group alone (p=0.000) No loss to follow up

5

Wood & Moran (2011) [63] New Zealand Osteopath Journal Impact factor (2014): Unknown

RCT – blinded controlled experimental design Three arms: 1) SMT group (n=12)

2) Sham group (n=11) 3) Control group (n=12)

35 Asymptomatic – Healthy adults Age: 18-40 SMT group: 28.8yrs Sham group: 28.8yrs Control: 28yrs Sex: Male only Matched baseline measures (age, height, weight, least mobile segment, LL with neurodynamic restriction)

PSLR +/- NF - LL with neurodynamic restriction Time points measured: Immediately pre + post-intervention + 48 hrs End-point measured: R1 - perceived by participant - Average 3 measurements - NF angles –0-60° SD manoeuvres: - Ankle in plantargrade – re-Ax with NF.

Intervention: Spinal mobilisation - Prone lying applied to least mobile spinous process – L4 or L5 - Grade 3 – 77 Newtons, PA - 3 x 60s (20s rest) - 26bpm speed Provider: Osteopath – unknown experience Sham: Prone lying – application low amplitude oscillation to lower lumbar (L4/5) skin + soft-tissue in lateral direction Control: Prone lying - 3 mins 40s

- SMT was not SS superior to Sham, but produced a CS superior effect compared to Sham at 48hrs post-intervention on PSLR alone (9.5°) - SMT produced a CS superior improvement compared to control at both time intervals for PSLR 15°, 18.5°) and PSLR+NF (12.9°, 15.6°) - No with-group SS calculations provided. - Both SMT and Sham groups produced CS improvements in PSLR ROM for PSLR + PSLR+NF at both post-intervention intervals No loss to follow up

Kumar & Cherian (2011) [59] India Physio Journal impact factor (2014): 1.127

RCT – Observer blinded Two arms: 1) SMT + CP (n=12)

2) CP (n=12)

24 Symptomatic – L4/5 or L5/S1 IVD prolapse with radiculopathy Duration of sx: 3.38yrs Age: 20-50 (51.45yrs) Sex: Overall: 13 male/ 11 female Matched baseline measures

PSLR - LL with symptoms Time points measured: Pre-and post (3 days later) + 6 wks post-intervention End-point measured: P1 SD manoeuvres: at P1 - add NF and ankle DF to confirm the increase in pain

Intervention: Spinal mobilisation+ CP - position unclear - Rotational transverse pressure on the spinous process from pain free to painful side - 3 sessions on 3 days including day of Ax - Grade 1-4 – 5 reps x 2 sets – re-Ax between sets CP: 5 mins intermittent lumbar traction, 15 mins heat, home program handouts

- SS (p=0.029) + CS (12.5°) superior immediate improvement in PSLR ROM compared to CP - No SS (p=0.312) difference at 6 weeks post-intervention, however, it did produce a CS (17.5°) superior effect - No within-group SS calculations provided. - CS improvements were observed within both groups at both post-intervention Ax intervals

Andersson et al. (1999) [56]

U.S.A.

Osteopath Journal Impact factor (2014): 55.873

RCT Two arms: 1) SMT group (n=83) 2) SC group (n=72)

155 Symptomatic – LBP Duration of symptoms: ≥ 3 weeks but ≤ 6 months Age: 20-59 Osteopathic: 28.5yrs SC: 37yrs Sex: 66 Male/89 Female Matched baseline measures

PSLR - unclear which LL Ax - Ax in supine and sitting Time points measured: Baseline + 12 wks post-intervention End point measured: Not documented SD manoeuvres: None

Intervention: Spinal manipulation -Pragmatic approach - 8 visits (4 wkly visits, then 4 at 2 wk intervals) Provider: 1 of 3 osteopaths (unknown experience) SC: 8 visits - Analgesics, Anti-inflammatories, Active PT, Ultrasound, Diathermy, Hot/cold packs, back corset, TENS, 10 min. educational LBP video

- No between-group SS or CS differences in PSLR ROM found measured in sitting (p=0.94) or supine (p=0.40) - No within-group calculations conducted - A CS (6.6°) improvement in PSLR ROM was found 12 wks post-intervention when measuring PSLR in sitting alone

6

Pollard & Ward (1998) [60] Australia Chiro Journal impact factor (2014): 1.647

RCT – Clinical cohort study Three arms: 1) Upper Csp Manip group (n=18) 2) SIJ Manip Group (n=18) 3) Control Group (n=16)

52 Asymptomatic - Volunteer Chiro students Age: 18-34 Sex: Not documented No calculations provided on baseline comparability

PSLR Time points measured: Pre + immediately (30s) post- intervention End-point measured: - Force transducer - application of 5% of the subject’s body weight SD manoeuvres: None documented

Intervention: Spinal manipulation Upper Csp Manip: - rotatory double index contact style to C1 - side chosen at random SIJ Manip: - lumbar roll pisiform contact to SIJ - side chosen at random Control: Digital pressure on mastoid process bilaterally for 30s x 3

- No SS or CS differences were found between groups - The application of the Csp manip produced an immediate SS (p=0.001) improvement in PSLR ROM, but this did not reach CS (4.12°).

Note: Ax: Assessment; BRLP: Back Related Leg Pain; Chiro: Chiropractor; CP: Conventional Physiotherapy; CS: Clinically Significant; Csp: Cervical spine; DDD: Degenerative Disc Disease; DF: 29 Dorsiflexion; EOR: End of Range; ex’s: exercises; F: Female; Gr: Grade; HEA: Home Exercise and Advice; hrs: Hours; IVD: Intervertebral Disc; (L): Left; LBP: Low Back Pain; LL: Lower Limb; M: 30 Male; Manip: Manipulation; Max: maximum; Mgt: Management; Min: Minimum; Mins: Minutes; Mob: Mobilisation; MT: Manual Therapy; NF: Neck Flexion; P1: Onset of Pain; Physio: 31 Physiotherapist; PSLR: Passive Straight Leg Raise; PT: Physical Therapy; PA: Posterior-anterior; (R): Right; R1: Onset of Resistance; Rx: Treatment; RCT: Randomised Controlled Trial; S: 32 seconds; SS: Statistical Significance; SC: Standard Care; SD: Standard Deviation; SIJ: Sacroiliac Joint; SMT: Spinal Manipulative Therapy; ST: Short Term; Sx: Symptoms; TENS: Transcutaneous 33 Electrical Nerve Stimulation; Wks: weeks; Yrs: years 34

7

Appendix C: Supplemental Material Table 3. Summary of within-group mean differences: comparing mean PSLR ROM at each time interval relative to baseline

Study Comparator PSRL test Variation

Assessment Time Point (change from baseline)

post test 24 hrs 48 hrs 6 wks 12 wks

Ganesh et al. (2015) [58]

1

SMT 5.7°

[1.6 to 9.8] 7.4°

[3.4 to 11.5]

Control 1.3°

[-3.2 to 5.8] 1.4°

[-3.2 to 5.9]

Wood & Moran (2011)

[63]2

SMT PSLR

14.7° [9.7 to 19.7]

17.0°

[10.6 to 23.3]

PSLR+NF 13.2°

[7.5 to 19.0]

14.7° [7.6 to 21.8]

Sham PSLR

11.7° [6.5 to 16.9]

7.5°

[2.0 to 13.0]

PSLR+NF 11.5°

[5.1 to 17.9]

10.0° [3.6 to 16.5]

Control PSLR

-0.3° [-3.3 to 2.7]

-1.5°

[-5.6 to 2.6]

PSLR+NF 0.3°

[-3.2 to 3.7]

-0.9° [-3.2 to 1.4]

Szlezak et al. (2011) [61]

1

SMT 8.5°

[3.8 to 13.2]

Stretching 1.55°

[1.8 to 4.9]

Control 0.11

[-3.0 to 3.2]

Pollard & Ward (1998)

[60]2,3

SIJ SMT 4.1°

[2.4 to 5.8]

Csp SMT 2.2°

[0.1 to 4.2]

Control 0.7°

[-1.5 to 2.9]

Vieira-Pellenz et al. (2014)

SMT 13.7°

[9.6 to 17.7]

8

[62] Sham

-0.5° [-0.6 to -0.1]

Andersson et al. (1999) [56]

2

SMT supine

2.8° [0.7 to 4.9]

sitting 6.6°

[3.9 to 9.3]

SC supine

1.3° [-0.8 to 3.4]

sitting 5.2°

[2.8 to 7.6]

Bronfort et al. (2014) [57]

1

SMT+HEA left leg

4.7° [2.0 to 7.4]

right leg 5.2°

[2.5 to 7.9]

HEA alone left leg

0.8° [-1.6 to 3.2]

right leg -0.4°

[-3.2 to 2.4]

Kumar & Cherian

(2011) [59]1

SMT+CP 22.5°

[18.7 to 26.3]

32.5° [28.7 to 36.3]

CP 10°

[8.0 to 12.2]

15° [13.0 to 17.0]

Data reported as mean change [95% confidence interval (95%CI)] from baseline to assessment period. 1mean difference and 95%CIs calculated from

reported group means and standard deviation at each time point. 295%CIs calculated from standard deviation of between group differences of change pre to

post. 3Values originally reported as relative to vertical, direction of difference inverted so that a positive difference is an increase in range of motion. When

conducted, pooled standard deviation of was calculated assuming equal variance of groups. Mean differences from baseline to assessment period between

groups that are greater than clinical significance (6) are marked bold. CP: Conventional Physiotherapy, Csp: Cervical spine, HEA: Home Exercise and

Advice, SC: Standard Care, SMT: Spinal manipulative Therapy, SIJ: Sacroiliac Joint