Effectiveness of Thoracic Spine Manipulation on the Management of Neck Pain: A Systematic Review and Meta-Analysis of Randomized Control Trials

Introduction

Neck pain is a common condition worldwide which is increasing in the general population and most people experience at some point in their lifetime.¹ The movement pattern and daily activities can be affected with ultimate impact on the social participation and economy of individuals.^2,3 The one-year prevalence for neck pain has been reported to range between 16.7% and 75.1%.⁴ According to the Global Burden of Disease 2010 study, neck pain ranked 4th highest in terms of disability as measured by years lived with disability (YLDs), and 21st in terms of overall burden, and in 2015 neck pain reported as the leading cause of disability in most countries.^5,6

The common classification of neck pain is mechanical neck pain.⁷ Although the definition varies among different research studies, mechanical neck pain is most commonly defined as pain located in the cervical spine or cervicothoracic junction that is elicited and/or exacerbated by cervical motion and/or palpation of cervical musculature.^7–10 Recently, the Cochrane Collaboration (2015) suggested categorizing patients based on symptom duration: acute (<6 weeks), subacute (6–12 weeks), and chronic (>12 weeks).¹¹ Acute, subacute, and chronic stages are time-based stages helpful in classifying patient conditions. A clinical practice guideline on neck pain stated that time-based stages help make treatment decisions based on the irritability of tissues on neck conditions.¹²

Neck pain management may be applied using a different treatment approach’s like; medications, manual therapy, and exercises. Manual therapies contribute to improve pain and function in adults with neck pain.^13,14 The intrinsic biomechanical linkage of the thoracic with the cervical spine, disturbances in the biomechanics of the thoracic spine could be a primary contributor to neck pain.¹⁵ Thoracic manipulation alone or in combination with other physiotherapy treatments was used to reduce cervical spine pain, improve function and cervical range of motion (CROM).^10,16

Evidence has begun to emerge in support of thoracic spine thrust manipulation as an intervention in the management of non-specific neck pain and acute mechanical neck pain.^7,10,17–19 The effect of these interventions on patients with chronic neck pain has been inconsistent; for instance, pain relief was obtained in the studies that compare thoracic manipulation alone or in combination with the control or comparison group.^20–24 However, no statistically significant differences were found in the intragroup or intergroup analyses of the experimental and placebo groups regarding the intensity of neck pain at rest in the immediate or short-term post-intervention evaluations. Similarly, the study that compares cervical manipulation to thoracic manipulation did not indicate a statistically significant group-by-time interaction for chronic neck pain.^25–27 Hence, the main objective of this review is to evaluate the current evidence on the effectiveness of thoracic manipulation on decreasing pain intensity and neck disability on subjects with chronic mechanical neck pain.

Materials and Methods

Protocol and Registration

This systematic review and meta-analysis were registered in the PROSPERO database as (CRD42020204983) accessed at https://www.crd.york.ac.uk/PROSPERO and conducted according to the PRISMA guidelines.²⁸

Eligibility Criteria

Inclusion Criteria

Randomized control trials available in full text and written in the English languages that were recently conducted within the last 10 years (2010–2020). It was decided to use the year 2010 as a cut off because a preliminary search conducted by the authors did not reveal any substantial literature on this topic prior to the year 2010. Studies looking at participants with only chronic mechanical neck pain and age ≥18 years that compared the effect of thoracic manipulation with other managements were included.

Exclusion Criteria

Studies that were not matched to the objective of this review, with low quality of PEDro score <4/10 and studies that compared two thoracic manipulation techniques, not specifically explained the participants’ symptom duration or with acute neck pain, and studies without the outcome of interests were excluded from this review.

Study Type

Any published randomized controlled trial (RCT) in a full-text form that investigated the effect of thoracic manipulation on subjects with chronic mechanical neck pain was included.

Participants

The participants were adults with age ≥18 years and both sexes (male and female) with a chronic mechanical type of neck pain.

Interventions

The intervention of interest was thoracic manipulation (TM). The technique could be provided multiple times to a single spinal region or various spinal regions during a single session or over multiple sessions. Co-interventions could also be included within the treatment session if these were also included in the comparison group. This allows for differences in treatment effect to be attributed to the addition of TM in the experimental group.

Comparison

The studies need to compare thoracic spine thrust manipulation to any other non-thrust manipulation, mobilization, exercise, and no intervention or placebo.

Outcomes

The outcomes of interest for this systematic review were pain and disability. Across all studies, the pain was measured using either the Numeric Pain Rating Scale (NPRS; 0–10pts) or the Visual Analog Scale (VAS; 0–100mm), and Disability was assessed using either the Neck Disability Index (NDI; 0–100%) or the Northwick Park Pain Questionnaire (NPQ; 0–100%).

Search Strategy

Beginning in June 2020, the authors comprehensively searched articles relevant to this meta-analysis using the electronic databases of PubMed, CINAHL, Cochrane Library, and PEDro data bases. The search was restricted to RCT, written only in English language, and conducted from 2010 to 2020. We used the following terms or keywords independently or in combination: (thoracic manipulation, thoracic thrust manipulation, neck pain, manipulation, thrust, cervical, and chronic neck pain filtered with RCTs, full text, human participants, and duration <10 years) (Table 1).

Table 1 The Literature Search Strategy

The goal behind the search strategy was to identify all potential RCTs that assessed the role of thoracic spine manipulation in the management of individuals with CMNP. The authors (GS, GG, and TT) comprehensively searched articles relevant to this meta-analysis. After the computerized search was completed, reference lists of all selected articles were searched by the authors (GS and GG) to identify additional related articles. Each author (GS, GG, and TT) examined all titles and abstracts to determine initial study eligibility. Full-text articles were then re-evaluated for specific inclusion criterion by (GS and GG).

Selection of Studies

We employed the predetermined inclusion criteria to screen for relevant full-text trials. An article was included for appraisal, data extraction and analysis based on the eligibility criteria.

Quality Assessment

The methodological qualities of the selected articles were critically appraised with 11-item Physiotherapy Evidence Database (PEDro) scale (Table 2).²⁹ PEDro scale is a validated tool to assess the methodological quality of physical therapy and rehabilitation trials. Two authors (GS and GG) who were physiotherapy professionals and more familiar with the scale independently scored each of the included articles, and then the discrepancies resolved through further discussion until consensus was met. A study was considered to be of “high” quality if the PEDro score of 7 or above, “fair” quality if the score of 5–6, and “poor” quality if the score was 4 or below.³⁰

Table 2 Quality of Articles (PEDro Scale)

To assess the overall body of evidence, the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach was also utilized for all included studies. There are five domains used in the GRADE system, including the risk of bias, the inconsistency of results, indirectness of evidence, imprecision, and reporting bias.^11,31 Randomized controlled trials begin with a high-quality evidence classification but may be downgraded if one or more of the above-described domains are present.³¹

Data Extraction

Two investigators (GS and GB) extracted data using a standardized data extraction form. We extracted data like sample size, age, outcome measures, interventions performed (including manipulation techniques or approach and treatment frequency), and treatment given to the comparative group, follow-up time, and results. Mean difference (MD) with standard deviation (SD) was extracted for the identified outcomes. Similarity and differences of the findings were putted and discussed.

Analysis

Data analysis was conducted using RevMan 5.3. The post-intervention scores and standard deviations (SD) were inputted into the meta-analysis. For all meta-analyses, a random-effects model with inverse variance method was used to calculate weighted mean differences and 95% CI. When there were multi-group data, we selected the common and clinically relevant comparison. Heterogeneity between trials was assessed using the I² statistic. The I² statistic describes the percentage of variability in effect estimates that was due to heterogeneity rather than sampling error (chance): less than 40% suggests that heterogeneity was not important, and above 75% suggests considerable heterogeneity. Values between 40% and 75% suggest moderate-to-substantial heterogeneity.³² The pooled data for each outcome were reported as weighted mean differences with a 95% confidence interval (CI).

Result

Study Selection

Searching of literature was done from different data bases. During the first search, a total of 745 articles were found from an electronic database. After removing the duplications, 312 articles were left. Of the 24 full-text articles 16 were excluded using the eligibility criteria. Finally, eight RCTs that matched the review objective were included in the meta-analysis. PRISMA flow diagram was used to show the searching and identifying the relevant articles (Figure 1).

Figure 1 Preferred Reporting Items for Systematic reviews and Meta-analysis (PRISMA) flow diagram of study selection.

Description of Studies

The studies chosen for inclusion in this meta-analysis were all randomized clinical trials investigating the effects of TSM on pain and neck disability among patients with CMNP compared to various other interventions. A total of 385 participants with a range from 30 to 120 were included across all 8 studies.^20–26 All of the studies included both male and female participants with the age range 18–60 except one article²⁷ in which all the participants were young women with chronic mechanical neck pain. All studies included participants with CMNP that were randomly assigned to either the TSM group or a comparison group. The follow-up times varied between studies from immediate to 6 months post-intervention. All studies investigated pain (NPRS or VAS), and three studies assessed disability (NDI). Comparison of thoracic manipulation was done with placebo^22,25,27 thoracic spine mobilization,²¹ cervical manipulation,²⁶ and co-interventions like infrared, craniocervical deep flexor exercise, and cervical stability training.^20,23,24 The result of the articles were synthesized, analyzed, and discussed for the conclusion of the review, and a summary of these included articles is presented in Table 3.

Table 3 Characteristics of the Included Studies

Methodological Quality Assessment

These articles were then assessed for quality utilizing a PEDro scale and presented in Table 2. The score of the included articles ranged from 4 to 8 with a mean score of 6.63/10 (SD=±1.3), indicating that the average quality of the included studies is fair or moderate. Five studies^{20,21,25–27} were found to be of high quality, two articles^22,23 of “fair” quality, and one article²⁴ of “poor” quality. None of the studies blinded the treating clinicians or therapists, and only one article²⁷ met the criterion regarding blinding of subjects as it is known that blinding is difficult to the patient and impossible to blind the care provider in manual treatments.

A formal grading of evidence was conducted using the GRADEpro software to provide an overall level of evidence for thoracic manipulation in the management of CMNP. The individual level ranged from low to high, and overall level of evidence (certainty) of the body of this study ranged from low to moderate. It is presented in Table 4 and Table 5.

Table 4 GRADE Evidence Profile of Individual Articles

Table 5 GRADE Overall Evidence Profile

Synthesis of Results

This review provides evidence on the effectiveness of TSM to reduce pain intensity and decrease neck disability on subjects with chronic mechanical neck pain. Some of the articles compare thoracic thrust manipulation with non-thrust, cervical thrust manipulation, and with a group out of intervention. The result of these articles are summarized and discussed. Statistical significance was determined by a p-value of <0.05 or 95% confidence interval.

Pain Intensity

Pain intensity was accessed in all eight studies with outcome measures of (VAS and NPRS).^20–27 Pain relief was attained in most of the studies.^20–24 Between-group differences for pain achieved by the thoracic spine thrust manipulation group were not only statistically significant but also clinically meaningful decrease in NPRS ranged from 34.4% to 40.6% in the TM group in immediate post-treatment, which was maintained for the 3 and 6-month follow-up similarly, patients receiving thoracic spine thrust manipulation experienced a greater decrease in neck pain between-group mean difference: 1.4; 95% CI (0.8–2.1) than did those receiving thoracic spine non–thrust mobilization within 10-minute post-intervention.^20,21

The study that investigated the acute effects of single-level thoracic manipulation (STM) and multiple-level thoracic manipulations (MTM) on chronic mechanical neck pain result showed pain level at rest significantly decreased after MTM1.96; 95% CI (1.04–2.95) and 2.35; 95% CI (1.44–3.71) at 24-hour and 1-week follow-up, respectively.²² A study that assessed the effect of thoracic manipulation (TM) and deep craniocervical flexor (DCF) training on the pain intensity of the patients revealed that the improvements in pain intensity were more significant in group TM combined with DCF training (MD and SD =−3.8±0.6) than in the group that received DCF training alone after 10 weeks’ follow-up time.²³ According to the post-intervention comparison of the upper thoracic manipulation and cervical stability training versus cervical stability training groups, there were significant differences in the pain values.²⁴

However, the result of the study that compared right and left cervical manipulation to thoracic manipulation did not indicate a statistically significant group-by-time interaction for neck pain. All three groups experiencing a similar decrease in neck pain after the intervention.²⁶ Similarly, a study conducted to investigate the effects of thoracic manipulation versus mobilization on chronic neck pain results at the 24-hour follow-up demonstrated VAS ratings significantly decreased for both treatment groups. However, there was no statistically significant difference in pain at rest between the three groups 24 hours after the intervention.²⁵ In a study conducted to assess the immediate and short-term effects of upper thoracic spine manipulation on pain intensity and myoelectric activity of the sternocleidomastoid muscles in young women with chronic neck pain, no statistically significant differences were found in the intragroup or intergroup analyses of the experimental and placebo groups regarding the intensity of neck pain at rest in the immediate or short-term post-intervention evaluations.²⁷

In this systematic review and meta-analysis, the immediate or short-term effect of thoracic manipulation on pain intensity in patients with chronic mechanical pain was examined by pooling data from eight studies that used self-reported NPRS and VAS. Meta-analysis of the five studies^22–25,27 with a total 151 participants that measures pain by VAS (0–100mm) revealed a significant effect (MD −12.46; 95% CI: −17.29, −7.64; I² = 52%, p < 0.001) (Figure 2), and a meta-analysis of three studies^20,21,26 that measures pain by PNRS (0–10 points) revealed a significant effect (MD −0.85; 95% CI: −1.60, −0.10; I² = 61%, p = 0.03) at immediate follow-up (Figure 3).

Figure 2 A meta-analysis on the effect of thoracic spine manipulation (TSM) for pain using visual analog scale (VAS).

Figure 3 A meta-analysis on the effect of thoracic spine manipulation (TSM) for pain using numeric pain rating scale (NPRS).

Neck Disability

The neck disability was accessed in three studies^20,22,23 with 183 participants. The result of a study that compares experimental groups that received TM and a control group without the manipulative procedure revealed the TM group significantly decrease in neck disability compared to the control group with a between-group difference of 6–8.9 points.²⁰ Similarly, in the study that investigated the acute effects of single-level (STM) and multiple-level thoracic manipulations (MTM) on CMNP the neck disability in the STM and MTM groups was significantly lowered by 8–12 points than the control group at 24-hour and 1-week follow-up.²² The NDI of a group that receives thoracic manipulation combined with deep craniocervical flexion training was reduced by a change of 5.9 points than that of a group with deep craniocervical flexion training alone.²³

In this systematic review and meta-analysis, the immediate or short-term effect of thoracic manipulation on neck disability in patients with chronic mechanical pain was examined by pooling data from these three studies. Meta-analysis of the three studies^20,22,23 with a total 183 participant measure neck disability revealed a significant effect (MD −6.46; 95% CI: −10.43–2.50; I² = 57%, p = 0.001) at immediate follow-up (Figure 4).

Figure 4 A meta-analysis on the effect of thoracic spine manipulation (TSM) on neck disability.

Discussion

This systematic review and meta-analysis evaluated the effect of TSM in patients with chronic mechanical neck pain. To the limit of our knowledge, this is a preliminary systematic review study to evaluate the effect of TSM among subjects with chronic mechanical neck pain. This systematic review includes eight recent RCTs conducted within 10 years with a mean PEDro score of fair quality of evidence and the GRADE level of evidence found to be low to high for individual studies, and low-to-moderate overall evidence of the review.

The review provides evidence that thoracic manipulation alone or in combination with other treatments has produced an immediate and short-term effect to improve pain and neck disability among patients with chronic mechanical neck pain. The summary pooled effect of this meta-analysis favors the experimental or thoracic manipulation group with the effect size of −0.85 and-12.46 for the studies that assessed pain using PNRS and VAS measurements, respectively. Similarly, the summary effect on neck disability was −6.46 in favor of the thoracic manipulation group. This finding is consistent with the results of previous studies investigating thoracic manipulation for neck pain.^17–19 TSM is more beneficial than thoracic mobilization, cervical mobilization, and standard care in the short-term, but no better than cervical manipulation or placebo thoracic spine manipulation to improve pain and disability among patients with mechanical neck pain (MNP)¹⁷ and nonspecific neck pain.¹⁸ Similarly, a systematic review that compares thoracic manipulation versus mobilization found that there is a significant amount of evidence for the short-term benefits of thoracic manipulation in treating patients with mechanical neck pain.¹⁹ This may support the concept of the intrinsic biomechanical linkage of the thoracic with the cervical spine, and disturbances in the biomechanics of the thoracic spine could be a primary contributor to neck pain.¹⁵

The current review used only participants with chronic mechanical neck pain. The included studies compare the intervention or thoracic manipulation with sham techniques or placebo,^22,25,27 thoracic spine mobilization,²¹ and cervical manipulation.²⁶ Co-interventions like infrared and educational material, craniocervical deep flexor exercise, and cervical stability training^20,23,24 were included within the treatment session as these were also included in the comparison group which shows the differences in treatment effect to be attributed to the addition of TM in the experimental group. However, regardless of the comparative interventions, the results of the review suggested that treatment programs which incorporate thoracic spine thrust manipulation have larger treatment effects on the outcome measures. Including TM in the experiment group with co-interventions in both groups can bring clinically important differences in decreasing the pain and neck disability among the participants.

There is moderate heterogeneity among the included studies with I² = 52%, 57%, and 61%. This might be because most of the included studies investigated different control interventions, different doses, and techniques of experimental intervention, another possible source of heterogeneity might be a large variance in sample size, from 26 in the Suvamnato study²⁵ to 120 in the Lau study.²⁰ Though this heterogeneity questions the overall validity of conducting meta-analysis, the included articles were fair in methodological quality and moderate level of evidence the authors consider meta-analysis valuable in providing an overall treatment effect of TSM as it is relevant and may aide in the clinical decision-making process in the management of individuals with MNP.

The studies described different thoracic spine thrust manipulation techniques. The specific site to which the thrust manipulation was applied also varied among studies, with four RCTs attempting to provide the thrust manipulation at one or more generic locations of the mid-thoracic and upper thoracic spine with a patient in supine line elbow flexed anteroposterior thrust manipulation was done.^20,21,23,26 A study extrapolated on the decision for the location for thoracic spine thrust manipulation, recognizing that the technique does not target specific vertebral segments.³³ In contrast, to suggested clinical practice, Suvernato et al attempted to specifically manipulate a single segment that was determined to be hypo-mobile during joint mobility testing.²⁵ Puntamutakul et al investigated both single and multiple thoracic manipulations by comparing with the control group in which the therapist’s hand was placed without any treatment.²² Nevertheless, outcomes among the current studies did not appear to be influenced by the specific technique or rationale for application.

The follow-up intervals for all studies included in this review were relatively short. The current literature has individual reports for treatment effectiveness immediately following the treatment within 24 hours. Only one study assessed the post-intervention outcome immediately, 3-months, and at a 6-month follow-up time.²⁰ Although various follow-up intervals have been included, long-term outcomes have not been reported.

Implications for Clinical Practice

This review and meta-analysis found that thoracic manipulation provide the immediate and short-term effect to improve pain and neck disability among patients with chronic mechanical neck pain. So clinicians or physiotherapists can apply the techniques of high-velocity low amplitude (HVLA) thoracic thrust manipulation as a choice of treatment to decrease pain and neck disability for patients with chronic mechanical neck pain.

Implications for Future Research

Future studies are needed to identify the most effective technique of thoracic thrust manipulation for each specific type of neck pain (acute, subacute, chronic, mechanical, and no mechanical). Its long-term effect with a longer follow-up time and to conduct a review of studies focused on the effect of thoracic spine manipulation as a single modality.

Strengths and Limitation

This review had strengths like it followed the PRISMA guidelines, registered in the PROSPERO, evidence quality of the studies assessed by PEDRo and GRADE, and all the included articles are recently published RCTs. Despite the strengths, this review had potential limitations. Inclusion of fair to low quality evidence articles, potential limitation of search, and inclusion of only English language articles as this might cause a chance of missing articles published in non-English languages. Lack of homogeneity among the included studies was also one of the main limitations, which might be due to the review was not specific to the follow-up time of the articles.

Conclusion

This systematic review and meta-analysis indicate that thoracic spine thrust manipulation reduces pain and neck disability in all adults with chronic mechanical neck pain compared with other interventions. While the findings of this review support the clinical benefit of TSM for reducing pain, clinicians should interpret these findings carefully as the overall quality of evidence ranges from low to moderate.