Comparative Effectiveness of 2 Manual Therapy Techniques in the Management of Lumbar Radiculopathy: A Randomized Clinical Trial

Abstract

Objective

The purpose of this study was to compare the effectiveness of Dowling's and Mulligan's manual therapy techniques on pain and disability in the management of lumbar disc herniation with radiculopathy (LDHR).

Methods

A total of 40 individuals with LDHR were randomly allocated into 2 groups, 20 participants each in PINS and SMWLM groups. Each participant was assessed at baseline, 4 weeks, and 8 weeks postintervention. The primary outcomes measured were pain (visual analog scale) and disability (Roland-Morris Disability Questionnaire). Secondary variables were quality of life (Short-Form 36 Health Survey), sciatica bothersomeness (Sciatica Bothersomeness Index), sciatica frequency (Sciatica Frequency Index), and general perception of recovery (Global Rating of Change Scale). Repeated-measures analysis of variance was used to compute within-group and between-groups interactions.

Results

No significant differences were observed in the baseline characteristics of participants in both groups. The results indicate that there were significant time effects for all outcomes in the study (P < .001) within each group. However, there was no significant difference between the 2 groups on any outcome variable (P > .05).

Conclusion

The findings indicate that there was no difference in pain or disability between the 2 manual therapy techniques in the management of LDHR.

Introduction

The 2016 Global Burden of Disease Survey identified low back pain (LBP) as a major health problem worldwide that is getting worse.¹ LBP is a common reason patients visit a physician or other musculoskeletal health care provider² and is among the most common symptoms of lumbar disc herniation (LDH).³ LDH is believed to be a major contributor to the estimated 60% to 80% lifetime prevalence of LBP in the general population and is among the most common causes of sciatica.³ Sciatica, also known as radiculopathy, accompanies approximately 10% of cases of LBP, with a lifetime prevalence ranging from 13% to 40%.⁴

The importance of identifying effective treatment strategies for sciatica has been emphasized in research indicating that the presence of sciatica is associated with delayed recovery, persistent disability, and increased health care system utilization and costs.⁴ Various treatment strategies have been tried for lumbar radiculopathy with varying degrees of success. In the last few years, studies have shown that manual therapy is beneficial in the management of many musculoskeletal disorders, including lumbar radiculopathy.^{5, 6, 7, 8} However, there are no standardized guidelines for appropriate manual therapy programs, which suggests that more treatment options are urgently needed to ameliorate lumbar radiculopathy.⁹

Manual therapy is a conservative management that uses different skilled hands-on techniques on a person's body (spine and extremities) for the purpose of assessing, diagnosing, and treating a variety of symptoms and conditions.^{5, 6, 7, 8} Dowling¹⁰ developed a manual therapy technique that involves progressive inhibition of neuromuscular structures (PINS) for the management of recalcitrant somatic dysfunctions. PINS has been hypothesized to resolve radiating neuromuscular pain and normalize reflex activity through progressive deactivation of myofascial trigger points (MTrPs). It can be used as a sole treatment or in combination with any other modality, and has been recognized by some authors^11,12 in the management of many neuromuscular dysfunctions. However, only a few case studies have demonstrated the effectiveness of PINS in the management of sciatica.^10,13,14 Further randomized controlled trials are needed to establish the efficacy of PINS in the management of sciatica.

Mulligan¹⁵ also developed a manual therapy technique using spinal mobilization with leg movement (SMWLM) for improvement of lumbar lesions resulting in pain and other signs below the knee. Mulligan's technique is postulated to free nerve compression through increased vertebral rotation and intervertebral disc space gapping. Thus, this technique has the ability to correct positional faults, which takes pressure off the structures that compress the nerve, and may reduce the extent of pain by centralization.¹⁶ However, available studies on the efficacy of Mulligan's technique are mainly case reports,^{15, 16, 17, 18} with only a few randomized controlled trials.^{19, 20, 21} In addition, these studies were based on limited outcomes and durations, which suggests that more studies on the efficacy of the technique are still needed. Therefore, this study was carried out to compare the effectiveness of Dowling's and Mulligan's manual therapy techniques in the management of LDHR. It was hypothesized that there would be significant differences between the 2 techniques with regard to pain intensity and functional disability in individuals with LDHR after 8 weeks of intervention.

Methods

Study Design and Setting

This study was a single-blind randomized clinical trial, approved by the Health Research Ethics Committee of the Federal Medical Centre, Nguru, Yobe State, Nigeria in accordance with the Declaration of Helsinki (FMC/N/CL.SERV/355/VOL iii/197). The trial was registered with the Pan African Clinical Trial Registry (trial number PACTR201810520651000).

Eligibility Criteria

Individuals with LBP diagnosed with LDHR via magnetic resonance imaging were included in the study. Criteria for inclusion were age between 18 and 65 years, unilateral radiculopathy, and pain in the distribution of the sciatic nerve. Exclusion criteria were any sign or symptom of dementia or other cognitive impairments, diagnosis of claudication, inflammatory or other specific disorders of the spine such as ankylosing spondylitis, vertebral collapse, rheumatoid arthritis, stenosis, spondylolisthesis, osteoporosis, previous spinal surgery, a known pregnancy, bilateral radiculopathy, and presence of any of the spinal red flags.

Sample Size and Sampling Technique

The required sample size was calculated using G*Power version 3.1.²¹ The effect size (ES) used for calculating the sample size was obtained from the previous study²² using pain as the primary outcome:

$$ES(\Delta )={\mu }_0-{\mu }_1/\sigma ,$$

where μ₀ = mean effect for the experimental group, μ₁ = mean effect for the control/comparison group, and σ = standard deviation for the control group.²³

Based on the previous study,²² the means and standard deviations for pain in the Mulligan Mobilization group (experimental group) and the Neural Mobilization group (control group) were 0.57 (0.14) and 0.30 (0.21), respectively.

$$ES=0.57-0.30/0.21.$$

Therefore, ES = 1.29.

The probability level (α), the power (p), and ES used for the calculation were then set at 0.05, 0.95, and 1.29, respectively, which yielded a sample size of 17 participants per group (a total of 34 participants), estimated by independent t test. The sample size was adjusted to 40 participants (20 participants per group) to cover up for potential drop outs.²⁴

Participant Recruitment

Participants were first screened for eligibility by the research assistant, and then those who met the inclusion criteria were contacted by the principal investigator to arrange for baseline assessment. Participants were provided with written informed consent on arrival at the baseline assessment. Each participant who was willing to participate and signed the consent form was recruited into the study.

Randomization and Concealment

Participants were allocated to groups using a simple random process. Forty (40) opaque envelopes with pieces of paper, on which were written either “Dowling” or “Mulligan,” were prepared by an assistant therapist who did not have any other input into the study. The envelopes were shuffled every time a new participant picked one. Each sealed envelope was opened by the participant as the participant was recruited. Neither the researcher nor the participants were aware of the contents of any specific envelope until it was opened. See Figure 1 for participant screening and group allocation.

Fig 1.

Flowchart of the study.

Outcome Measures

In lumbosacral radicular symptoms, the most common complaints are pain and inability to perform normal daily activities.^25,26 Outcome measures were therefore classified into primary and secondary.

Primary Outcome Measures

The Roland-Morris Disability Questionnaire for Sciatica (RMDQ) is a 23-item illness-specific functional assessment questionnaire that is frequently used for low back pain and sciatica. Scores range from 0 to 23, reflecting a simple unweighted sum of items endorsed by the respondent. Individuals with high scores at baseline have a severe disabling symptom. This tool has been reported to have a high level of reliability and validity and is responsive to change.^27,28

A visual analog scale (VAS) measured the intensity of pain in the back and leg. Pain was assessed on a horizontal 100-mm scale ranging from 0 mm, “no pain in the leg,” to 100 mm, “the worst pain ever.” The VAS has been shown to have a high interobserver reliability coefficient (r = 0.88).²⁹

Secondary Outcome Measures

The secondary outcome measures are the Short-Form 36 Health Survey (SF-36), which measures quality of life, the Sciatica Bothersomeness Index (SBI) for sciatica bothersomeness, the Sciatica Frequency Index (SFI) to measure frequency of sciatica, and the Global Rating of Change Scale (GROC) to assess general perception of recovery.

Study Procedures

Initial assessment of each participant was carried out by a coresearcher who was not involved in the administration of interventions. This was to determine back and leg pain intensity, functional limitation, sciatica bothersomeness, sciatica frequency, general perception of recovery, and quality of life, using VAS, RMDQ, SBI, SFI, GROC, and SF-36, respectively. The participants were asked to complete all questionnaires at the beginning of the study. In addition, outcomes were assessed 4 and 8 weeks after intervention.

The Dowling group received PINS, whereas the Mulligan group received SMWLM. However, all participants received therapeutic exercises in the form of lumbar stabilization and stretching exercises as adjunct therapies. Each participant had 2 treatment sessions per week for the 8 weeks of the study.

PINS Group

Participants in the PINS group received the PINS program using Dowling's protocol.¹⁰ Two related points, termed primary and secondary or endpoints, were palpated. The points were areas of most and least sensitivity, respectively, found along the continuum of a neuromuscular structure. Once the area of least sensitivity (endpoint) was found, a moderate ischemic compression was steadily maintained with the index finger of 1 hand without relieving pressure until completion of the protocol. The index finger of the other hand was used to apply pressure to the primary point (area of most sensitivity) for 30 seconds, after which another sensitive point was located with the middle finger of the hand proximal to the endpoint without relieving the pressure of the index finger. If the participant indicated that the latter point was more sensitive than the former, then pressure was maintained on the second point and relieved on the first point without relieving the endpoint pressure. This was maintained for 30 seconds before the third point was identified. The same pattern was followed progressively along the dysfunctional neuromuscular structure until the last point, approximately 2 cm proximal to the endpoint, was found. Pressure was maintained for 30 seconds simultaneously on the 2 points (the last and the endpoint) and then relieved.

SMWLM Group

Participants in the SMWLM group received SMWLM using Mulligan's protocol.³⁰ The participant was laid on his or her side, facing the therapist, with the affected leg uppermost. An assistant therapist supported the affected leg. The therapist then leaned over the subject and placed 1 thumb, reinforced with the other, on the spinous process of the herniated vertebra as palpated with reference to the posterior superior iliac crest. The therapist then pushed down on the chosen spinous process. This pressure was sustained while the participant actively performed the straight leg raise for the leg supported by the assistant therapist, provided this did not cause pain. This position was maintained for 30 seconds, after which the therapist released the pressure on the spinous process and the participant lowered the supported leg to the couch. During the first visit, 3 repetitions were applied. On subsequent visits, as the participant improved, the assistant therapist applied overpressure on the supported leg as the participant performed the straight leg raise. This was also sustained for 30 seconds, after which the leg was lowered to the starting position. This procedure was repeated 6 times on subsequent visits.

Therapeutic Exercises

Therapeutic exercises (lumbar stabilization and stretching exercises) were performed by all participants in both groups.

Lumbar Stabilization Exercises

McGill's protocol³¹ of lumbar stabilization exercises (curl-ups, horizontal side bridge, and bird dog) was given as adjunct treatment to each group. These exercises were performed for 10 minutes: 6 repetitions for 6 seconds each followed by a rest period of 30 seconds before successive repetitions. Each participant had 2 sessions a week for 8 weeks.

Curl-ups

The participant lay supine with hands supporting the lumbar spine, both knees bent at 90° and the hips bent at 45°. The participant then lifted up the thoracic and cervical spine as 1 unit, maintaining a rigid block position with no cervical motions (chin poking or chin tucking), and held the position for 8 counts. This exercise trains the rectus abdominis muscle.

Horizontal Side Bridge

The participant assumed a side-lying position and supported his or her body weight using the ipsilateral elbow. The participant then crossed the contralateral arm against the chest and the contralateral foot in front of the ipsilateral foot, and bridged by lifting the hip up while maintaining the trunk straight and supporting the whole body on the elbow and feet. This position was held for a count of 8, after which the participant returned to the starting position before carrying out another repetition. This exercise trains the quadratus lumborum, lateral obliques, and transversus abdominis.

Bird Dog

The participant went on hands and knees (quadruped position), then simultaneously raised the contralateral arm and ipsilateral leg and stretched them out completely. After a count of 8, the participant switched limbs and repeated the exercise for the same duration. This exercise trains the back extensors, including longissimus, iliocostalis, and multifidi.

Stretching Exercises

Myers's protocol³² of stretching exercises (plantar stretching, calf stretching, and hamstring stretching) was performed by each group. These exercises were performed for 10 minutes at 2 sessions a week for 8 weeks.

Plantar Stretching

The participant sat on a chair with feet on the floor. A tennis ball was placed under the foot and the participant put weight on various parts of the plantar surface, rolling the ball from the front of the heel out to the ball of the foot, to find places that hurt or feel tight. Enough weight was then applied until the point between pleasure and pain was reached. Pressure was sustained on each point for 10 seconds.

Calf Stretching

The participant stood erect and leaned forward with forearms resting on the wall, then stretched the lower leg section by putting 1 foot back and resting on the heel. If the heel reached the floor easily, then the knee was flexed forward toward the wall to increase the stretch on the soleus.

Hamstring Stretching

The forward bend described for calf stretching was used to lengthen the hamstring group. The upper body was swung left and right during bends to ensure the entire hamstring muscle group was activated and stretched.

Data Analysis

Data obtained from this study were analyzed using SPSS 20.0 (SPSS Inc., Chicago, Illinois). Descriptive statistics were used to summarize the sociodemographic and clinical parameters of the participants. The Shapiro-Wilk test was used to assess normality of the data, whereas Levene's test was used to assess homogeneity of variances between groups. A repeated-measures analysis of variance was used to analyze treatment effects, with intervention (SMWLM, PINS) as the between-participants variable and time (baseline, week 4, week 8) as the within-participant variable. The dependent variables analyzed were RMDQ, VAS leg, VAS back, SBI, SFI, SF-36, and GROC. When significant intragroup differences were detected by the analysis of variance, Bonferroni post hoc analysis of variance was used to assess differences across baseline and weeks 4 and 8. Differences between the means were considered at a 5% probability level (P < .05) with a 95% confidence interval.

Results

A total of 40 participants completed the study. No significant differences were observed in the baseline characteristics between participants in either group (Table 1). The analysis of effects of time and the intervention × time interaction on the variables is presented in Table 2. The results indicate that there were significant time effects for all outcomes (all P values <.001) with large effect sizes. However, the effect of the intervention × time interaction was not significant for all outcomes—RMDQ: P = .961; VAS leg: P = .924; VAS back: P = .445; SBI: P = .212; SFI: P = .098; SF-36: P = .135; GROC: P = .107—indicating that the intervention was not time dependent.

Table 1.

Baseline Demographics and Clinical Parameters of the Participants

Variable	SMWLM Group (n = 20)	PINS Group (n = 20)
Age (y)	45.44 (8.84)	49.51 (10.06)
BMI (kg/m2)	24.59 (1.83)	25.35 (2.52)
Duration of symptoms, y	2.25 (1.07)	2.10 (0.97)
RMDQ	11.40 (3.83)	11.25 (3.14)
VASL	5.75 (2.05)	5.50 (2.04)
VASB	5.55 (1.54)	6.00 (1.95)
SBI	10.35 (4.30)	10.35 (4.30)
SFI	10.95 (4.08)	13.15 (4.10)
SF-36	11.86 (3.99)	12.35 (3.36)
GROC	−3.20 (1.58)	−3.90 (1.02)

Data are presented as mean (SD).

BMI, body mass index; GROC, Global Rating of Change Scale; PINS, progressive inhibition of neuromuscular structures; RMDQ, Roland-Morris Disability Questionnaire for Sciatica; SBI, Sciatica Bothersomeness Index; SF-36, Short-Form 36 Health Survey; SFI, Sciatica Frequency Index; SMWLM, spinal mobilization with leg movement; VASB, visual analog scale back; VASL, visual analog scale le; y, years.

Table 2.

Analysis of Effects of Time and Intervention × Time Interaction

Variable	Effect	F	P	ηp2
RMDQ	Time	220.471	<.001	0.853
	Intervention × Time	0.022	.961	0.001
VASL	Time	51.670	<.001	0.576
	Intervention × Time	0.079	.924	0.002
VASB	Time	81.534	<.001	0.682
	Intervention × Time	0.799	.445	0.021
SBI	Time	80.933	<.001	0.680
	Intervention × Time	1.605	.212	0.041
SFI	Time	107.780	<.001	0.739
	Intervention × Time	2.430	.098	0.060
SF-36	Time	63.199	<.001	0.625
	Intervention × Time	2.201	.135	0.055
GROC	Time	360.693	<.001	0.905
	Intervention × Time	2.339	.107	0.058

GROC, Global Rating of Change Scale; RMDQ, Roland-Morris Disability Questionnaire for Sciatica; SBI, Sciatica Bothersomeness Index; SF-36, Short-Form 36 Health Survey; SFI, Sciatica Frequency Index; VASB, visual analog scale back; VASL, visual analog scale leg.

The Bonferroni post hoc analysis of time effects (Table 3) indicates that there were significant increases for all outcomes (except GROC) at all time points of intervention (P values <.001, except P = 1.000 for GROC). Between-groups comparisons (Table 4) did not reveal significant difference between the 2 groups on any measure—RMDQ: P = .819; VAS leg: P = .689; VAS back: P = .241; SBI: P = .082; SFI: P = .301; SF-36: P = .107; GROC: P = .806.

Table 3.

Bonferroni Post Hoc Analysis of Time Effect

Variable	Time Points (wk)	Mean Difference (95% CI)	P
RMDQ	0 vs 4	2.250 (−0.736, 1.236)	<.001
	0 vs 8	0.100 (−0.737, 0.937)	<.001
VASL	0 vs 4	0.250 (−0.399, 0.899)	<.001
	0 vs 8	0.500 (−0.734, 0.834)	<.001
VASB	0 vs 4	0.400 (−1.225, 0.425)	<.001
	0 vs 8	−0.400 (−1.748, 0.152)	<.001
SBI	0 vs 4	−1.750 (−3.038, −0.462)	<.001
	0 vs 8	−0.350 (−1.453, 0.753)	<.001
SFI	0 vs 4	−0.400 (−1.662, 0.862)	<.001
	0 vs 8	0.300 (0.655, 1.255)	<.001
SF-36	0 vs 4	−0.850 (−2.002, 0.302)	<.001
	0 vs 8	−2.500 (−3.735, −1.265)	<.001
GROC	0 vs 4	−0.300 (−0.872, 0.272)	<.001
	0 vs 8	−0.550 (−0.087, −0.013)	.000

GROC, Global Rating of Change Scale; RMDQ, Roland-Morris Disability Questionnaire for Sciatica; SBI, Sciatica Bothersomeness Index; SF-36, Short-Form 36 Health Survey; SFI, Sciatica Frequency Index; VASB, visual analog scale back; VASL, visual analog scale leg.

Table 4.

Between-Groups Comparison of Outcomes

Variable	F	P	ηp2
RMDQ	0.053	.819	0.001
VAS L	1.417	.241	0.036
VASB	0.163	.689	0.004
SBI	3.200	.082	0.078
SFI	1.098	.301	0.028
SF-36	2.731	.107	0.067
GROC	0.061	.806	0.002

GROC, Global Rating of Change Scale; RMDQ, Roland-Morris Disability Questionnaire for Sciatica; SBI, Sciatica Bothersomeness Index; SF-36, Short-Form 36 Health Survey; SFI, Sciatica Frequency Index; VASB, visual analog scale back; VASL, visual analog scale leg.

Discussion

This study investigated the effectiveness of Dowling and Mulligan's manual therapy techniques in the management of lumbar disk herniation with radiculopathy. The findings show that both techniques, alongside recommended back care exercises, have great impact on pain, functional disability, quality of life, and sciatica outcomes. The findings indicate that there were significant time effects for all outcomes (with the exception of GROC at week 8). This indicates that all measures significantly improved over time from baseline to week 8. The failure of GROC to show significant increases at week 8 could be due to inappropriate item selection from the 15-point GROC Likert scale or to recall bias by participants.

However, the effect of the intervention × time interaction was not significant for any outcome, indicating that the intervention was not time dependent. This means that individuals may improve significantly irrespective of number of treatment sessions. In addition, the minimal time required to observe significant changes in participants with the 2 manual therapy techniques is 4 weeks, and this is expected to be maintained if the treatment sessions are increased.

The findings also revealed that there was no significant difference between the 2 groups on any outcome. This indicates that neither of the 2 manual treatment techniques was better than the other in the management of LDHR. Individuals with lumbosacral radiculopathy usually report leg pain and sciatica as the factors most limiting functional ability.²⁵ The lack of a significant difference between the treatment effects of the 2 techniques could be due to the fact that they individually targeted all the structures that bring about leg pain and sciatica. The mechanism through which PINS reduces leg pain and sciatica may be due to its direct effects on MTrPs. Following MTrP deactivation, the resultant hyperemia may reduce the capacity of the nociceptive receptors to process through washing away the metabolites in the neuromuscular structures.¹⁰ In addition, ischemic compression applied on points near the greatest sensitivity or at the location of the symptoms may act as counterirritant. Large, fast-conducting afferents gate transmission in the dorsal horn and collateral fibers in the substantia gelatinosa or adjacent interneurons, then inhibit the transmission of pain via the spinothalamic tract. In this manner, pressure acts as a stimulant to neighboring tissues, reducing the sensitivity of the original tender point, and thus pain may resolve.¹⁰

On the other hand, SMWLM has the ability to free nerve compression through increased vertebral rotation and intervertebral disc space gapping that can lead to nucleus deformation and simultaneous approximation in the alternate layers of the annulus, thereby producing favorable therapeutic effects on the intervertebral disc.¹⁵ Thus, this technique has the ability to correct positional faults, which takes the pressure off structures that compress the nerve and may also reduce the extent of leg pain and disability by centralization.¹⁵

Participants in SMWLM group had more improvement in all outcomes compared to the PINS group. This is in agreement with findings from previous studies.^19,20,22 The current study also used more outcomes that previous studies did not explore, which strengthened the internal validity of the study.

Although both techniques were equally effective, there may be preferences in the choice of treatments in terms of suitability and ease of administration. It was observed in this study that participants in the PINS group complained of increased pain postintervention, which gradually declined 24 hours after treatment. This indicates that participants are not likely to prefer this approach if given the choice to select a treatment. This is common to all neuromuscular and trigger point release techniques.^10,11 In addition, administration of PINS was more stressful to the researcher, as it involves localizing a painful spot continuously, which is likely to cause fatigue in the hands of the administrator. By contrast, the SMWLM involves 2 physiotherapists (PTs) and is administered in a pain-free position, which makes it more likely to be preferred by both participants and therapists, as there was no pain during administration and no sign of fatigue from the PTs. However, this technique is resource intensive, as there must be 2 PTs to administer it, which put it at disadvantage in settings where the number of the PTs is very limited.

Limitations and Future Studies

The absence of a true control group makes it difficult to differentiate between the treatment effect and the natural course of the disorder, thus threatening the internal validity of the study. This study used a relatively small sample size of only 40 patients, which is known to affect the validity and generalizability of the results. Second, long-term outcomes were not assessed, and it is not known whether the differences observed after treatment could be maintained over longer periods of time.

Future studies should focus on the effect of these techniques on unilateral and bilateral low back pain with radiculopathy. Because both techniques produced clinically meaningful results, future studies should be done to see the combined effects of both techniques in individuals with LDHR.

Conclusion

The outcome of this study revealed that there was no difference in pain and disability between the 2 manual therapy techniques in the management of LDHR.

Funding Sources and Conflicts of Interest

No funding sources or conflicts of interest were reported for this study.

Contributorship Information

Concept development (provided idea for the research): B.B., M.S.D.

Design (planned the methods to generate the results): B.B., M.S.D.

Supervision (provided oversight, responsible for organization and implementation, writing of the manuscript): B.B., B.K.

Data collection/processing (responsible for experiments, patient management, organization, or reporting data): B.B., M.S.D.

Analysis/interpretation (responsible for statistical analysis, evaluation, and presentation of the results): B.K., M.S.D.

Literature search (performed the literature search): B.B., M.S.D.

Writing (responsible for writing a substantive part of the manuscript): B.B., M.S.D., B.K.

Critical review (revised manuscript for intellectual content, this does not relate to spelling and grammar checking): B.B., B.K.

Practical Applications.

• •The study highlights the use of specific protocols to ameliorate problems related to lumbar radiculopathy.
• •The study highlights the effectiveness of a nongadget therapy to manage symptoms of lumbar radiculopathy.
• •The study also showed that a combined nonoperative therapy is likely to give better improvements in people with lumbar radiculopathy than individual protocols, as highlighted by previous studies.

Alt-text: Unlabelled box