Randomized controlled trial of interferential therapy and manipulativetherapy for acute low back pain.

Hurley, D. A., McDonough, S. M., Dempster, M., Moore, A. P., & Baxter, G. D. (2004). Randomized controlledtrial of interferential therapy and manipulative therapy for acute low back pain. Spine, 29, 2207-2216.

Published in:Spine

Document Version:Early version, also known as pre-print

Queen's University Belfast - Research Portal:Link to publication record in Queen's University Belfast Research Portal

General rightsCopyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or othercopyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associatedwith these rights.

Take down policyThe Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made toensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in theResearch Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk.

Download date:08. Oct. 2020

1

Title Page

Randomized Controlled Trial of Interferential Therapy and Manipulative Therapy for Acute

Low Back Pain

Authors List

1. Deirdre A. Hurley, MISCP, MCSP, PhD, College Lecturer, School of Physiotherapy,

University College Dublin, Mater Misericordiae Hospital, Eccles St, Dublin 7, Rep.

Ireland.

2. Suzanne M. McDonough, MCSP, PhD, Lecturer in Rehabilitation Sciences,

Rehabilitation Sciences Research Group, Faculty of Social and Health Sciences and

Education, University of Ulster at Jordanstown, Co. Antrim, BT37 OQB, N.Ireland,

3. Martin Dempster, PhD, Lecturer, School of Psychology, Faculty of Science and

Agriculture, Queen’s University, Belfast, N. Ireland,

4. Ann P. Moore, FCSP, PhD, Professor of Physiotherapy and Director of Clinical

Research Unit for Healthcare Professions, University of Brighton, Robert Dodd

Building, 49 Darley Road, Eastbourne, BN20 7UR, England,

5. G. David Baxter, MCSP, DPhil, Professor of Rehabilitation Sciences, Rehabilitation

Sciences Research Group, Faculty of Social and Health Sciences and Education,

University of Ulster at Jordanstown, Co. Antrim, BT37 OQB, N.Ireland.

Author for correspondence (and to who requests for reprints should be addressed):

Dr Deirdre A. Hurley, PhD, School of Physiotherapy, University College Dublin, Mater

Misericordiae Hospital, Eccles St, Dublin 7, Rep. Ireland.

Tel: +353 1 8034310; Fax: +353 1 8303550; E-mail: deirdre.hurleyosing@ucd.ie

This work was completed as part of a PhD thesis at the University of Ulster Rehabilitation

Sciences Research Group.

2

Supported by the Society of Orthopaedic Medicine (UK and Republic of Ireland) Project

Grant, Manipulation Association of Chartered Physiotherapists Churchill Livingstone Award

and Research Presentation Award, Tenscare Ltd for loan of interferential therapy Omega

Inter 4150 portable units.

3

Abstract

Study Design. A multi-center assessor-blinded randomized clinical trial was conducted.

Objectives. To investigate the relative effectiveness of interferential therapy and manipulative

therapy for patients with acute low back pain when used as sole treatments and in

combination.

Summary of Background Data. Both manipulative therapy and interferential therapy are

commonly used treatments for low back pain. Evidence for the effectiveness of manipulative

therapy is available only for the short term. There is no evidence for interferential therapy

and no study has investigated the effectiveness of interferential therapy combined with

manipulative therapy.

Methods. Consenting subjects (n=240) were randomly assigned to receive a copy of the Back

Book and either manipulative therapy (MT; n=80), interferential therapy (IFT; n=80) or

combined manipulative therapy and interferential therapy (CT; n=80). Follow-up outcome

questionnaires were posted at discharge, 6 and 12 months.

Results. The groups were balanced at baseline for low back pain and demographic

characteristics. All interventions were found to significantly reduce functional disability and

pain and increase quality of life at discharge and to maintain these improvements at 6 and 12

months. No significant differences were found between groups for reported LBP recurrence,

work absenteeism, medication consumption, exercise participation and healthcare use at 12

months.

Conclusions. For acute low back pain, interferential therapy whether used in isolation or in

combination with manipulative therapy was as effective as manipulative therapy alone (in

addition to the Back Book).

Key Words: physiotherapy; effectiveness; low back pain; manipulation; interferential

therapy; primary care; randomized clinical trial.

4

Key Points

For acute low back pain, interferential therapy whether used alone or in combination

with manipulative therapy was as effective as manipulative therapy alone.

Physiotherapists should question the usage of the combination of interferential therapy

and manipulative therapy for patients with acute low back pain.

Physiotherapists with postgraduate qualifications in manipulative therapy provided all

treatments.

Future randomised controlled trials that compare the relative effectiveness of

treatments should include a no treatment and placebo group to account for the

potential effect of natural history and placebo.

5

Mini Abstract

A multi-center assessor-blinded randomized clinical trial was conducted to investigate the

relative effectiveness of interferential therapy and manipulative therapy for patients with acute

low back pain when used as sole treatments and in combination. At 12 month follow-up all

interventions were found to significantly reduce functional disability, pain, work absenteeism

and medication consumption and to increase quality of life and exercise participation.

6

Text

Introduction

Simple backache (low back pain with no associated nerve root or serious spinal

pathology) affects the majority of the population at some stage in their adult lives and

constitutes a major public health problem in Western industrialised societies (Deyo 1998).

Despite substantial advances in our understanding of low back pain (LBP) sources and pain

mechanisms, it is claimed that up to 90% of cases do not have a demonstrable pathologic

basis (Spitzer et al. 1987) and are therefore termed nonspecific LBP. Patients are typically

classified as having ‘acute’ (i.e. current attack of less than three months) or ‘chronic’ LBP

(i.e. current attack of more than three months duration, Waddell 1998).

Physiotherapists treat an estimated 1.6 million people with LBP each year in the UK

and are more likely to treat LBP patients with poor prognostic indicators, for example

moderate to high levels of self-reported disability due to LBP and pain radiating below the

knee (Mielenz et al. 1997). In 1998, physiotherapists, chiropractors and osteopaths accounted

for 37% of the Stg£1.6 billion direct healthcare costs of LBP, followed by the hospital sector

(31%) and primary care (14%) (Maniadakis and Gray 2000). The results of several recent

surveys of the physiotherapeutic management of LBP in Britain and Ireland (Foster et al.

1999; Gracey et al. 2002) found that a range of treatment strategies is being utilised: advice,

Maitland mobilisations, McKenzie, abdominal exercises, and interferential therapy.

Interferential therapy (IFT) has the highest ownership and usage of all

electrotherapeutic modalities in the United Kingdom and the Republic of Ireland (Pope et al.

1995; Cooney et al. 2000) being predominantly employed for its hypoalgesic effects, ease of

application and time efficiency (Lindsay et al. 1990; Noble 1998). Despite being used by up

to 44% of therapists for LBP management (Moore 1998; Foster et al. 1999; Gracey et al.

2002), the complete absence of any evidence of effectiveness necessitated its omission from

7

recent LBP clinical guideline documents (Waddell et al. 1999). A previous randomised

controlled trial (RCT) involving patients with acute and chronic LBP by Werners and

colleagues (1999) reported that IFT alone was no more (or less) effective than lumbar traction

with massage in reducing LBP-related functional disability or pain up to 3 month follow-up.

Interferential therapy is usually used in combination with other forms of treatment (typically

advice and manipulative therapy) for LBP management (Moore 1998; van Tulder 1999b;

Gracey et al. 2002) and consequently investigations into the effectiveness of IFT within a

multimodal (physiotherapeutic) treatment regime have been advocated (van Tulder 1999b).

Therefore, a randomised controlled trial that compared the effectiveness of IFT when used in

isolation and, as part of a multimodal package for patients with acute LBP would contribute

significantly to its limited evidence base.

Clinical guidelines advocate the use of manipulative therapy (MT) in acute LBP

conditions (Waddell et al. 1999) which incorporates both mobilisation (low velocity manual

force that does not involve a thrust) and manipulation (high velocity range-expanding thrust)

techniques. Within current physiotherapy practice in Britain and Ireland, the most well

recognised and used approaches are those of Geoffrey Maitland (Maitland 1986) and Dr

James Cyriax (Cyriax 1984; Foster et al. 1999; Gracey et al. 2002), being preferred by up to

62% of physiotherapists for their hypoalgesic and mobilising effects. However, despite a

total of 38 RCTs investigating the effectiveness of various types of MT, systematic reviews

have not consistently detected positive effects for spinal manipulation in people with acute

LBP compared to placebo and a range of other active treatments (Shekelle et al. 1992; Bigos

et al. 1994; Evans and Richards 1996; Koes et al. 1996; van Tulder et al. 1997; Mohseni-

Bandpei et al. 1998; Bronfort 1999). Further large-scale rigorous RCTs of patients with

clinically homogeneous LBP syndromes, who receive well-defined MT interventions and who

8

are assessed for response with valid outcome measures have been recommended (Shekelle et

al. 1992).

Furthermore, the recommendations of the UK Clinical Guidelines regarding the

promotion of early return to normal activities and the avoidance of bed rest were

encompassed into the Back Book (1996), an evidence-based patient education booklet. It has

been successfully piloted in primary care and shown to be readily acceptable and

understandable to individuals with LBP, and to create a positive shift in beliefs about LBP

(Burton et al. 1996). The developers believe it is more likely to have an impact as part of a

combined treatment package (Burton et al. 1999), and thus it was an appropriate standardised

co-intervention for the proposed RCT.

The aim of this RCT was to investigate the relative effectiveness of interferential

therapy and manipulative therapy when used as sole treatments and in combination (in

addition to the Back Book advocated by current clinical guidelines) for patients with acute low

back pain.

9

Materials and Methods

Selection of Patients. Patients were recruited within the British government-funded National

Health Service (NHS) in Northern Ireland in four of the United Hospitals Health and Social

Services Trust Hospitals (Antrim, Whiteabbey, Mid Ulster and Waveney) physiotherapy

departments. All patients aged 18 to 65 years referred by general practitioners (GPs) for

treatment of LBP with or without pain radiation into the buttock and/or one or both lower

limbs, of between 1 to 3 months duration were invited to participate. The first author, who

was principal investigator, was responsible for verifying eligibility, providing detailed written

and verbal explanations regarding the nature of the study and obtaining written consent. All

subjects signed a consent form before admission to the study. The Research Ethical

Committee of the University of Ulster approved the study protocol. Consenting subjects were

screened for any exclusion criteria as detailed in Table 1.

Randomization. Subjects were randomly allocated to one of three groups (manipulative

therapy, interferential therapy, combined therapy) using an allocation schedule generated from

a random numbers table. This was drawn up by a member of the research team not involved

in the day-to-day running of the trial (S.Mc.D.). Based on this schedule the group allocation

of each consenting, numbered subject was communicated to the relevant treating

physiotherapist by telephone contact with the research group secretary. The principal

investigator was not involved in any aspect of randomization and the allocation schedule was

concealed from her until all interventions were assigned.

Interventions. Only chartered physiotherapists who had successfully completed the Society

of Orthopaedic Medicine (SOM) membership examination, a postgraduate qualification in

manipulative therapy, recognised by the Chartered Society of Physiotherapy (CSP) and the

International Forum of Orthopaedic Manipulative Therapists (IFOMT), were eligible to

administer the treatment protocol. Sixteen physiotherapists met the criteria and were willing

10

to participate in this study (100% participation rate); i.e. 4 physiotherapists within each of the

4 participating hospitals. All treatments were provided to an individual patient by the same

therapist.

The Back Book. Following assessment, all subjects received the Back Book, from their treating

physiotherapist, who reinforced its positive messages during the first visit, by encouraging

early return to normal activities and participation in low impact activities such as walking,

swimming and cycling.

Manipulative Therapy (MT) Group. Subjects assigned to this group were treated by the MT

protocol, which was defined as techniques for the spine described by Maitland (1986) and

Cyriax (1984). Each physiotherapist was limited to using only these techniques but had free

choice of which to use and when, and the spinal regions to which they were applied.

Interferential Therapy (IFT) Group. Subjects assigned to this group were treated by the IFT

protocol based upon the results of a previous study by the researchers (Hurley et al. 2001a).

Omega Inter 4150 portable IFT units were utilised to deliver standardised IFT stimulation

parameters: i.e. carrier frequency 3.85 kHz; beat frequency 140 Hz constant; pulse duration

130 µs; treatment time 30 minutes; using the spinal nerve root electrode placement method via

two Reply 658 carbon silicone self-adhesive electrodes (50x100mm) (Figure 1).

Combined Therapy (CT) Group. Both the MT and IFT protocols were provided to subjects

assigned to the CT group with the MT protocol preceeding the IFT protocol.

Apart from the designated protocol, the participating physiotherapists were not

permitted to administer any other forms of MT, electrotherapy or other techniques (spinal

traction, heel raises, corsets, acupuncture, injection therapy, taping, McKenzie) during the

intervention period of the RCT. Subjects were requested to continue normal activities and to

avoid other treatments for the duration of the RCT apart from routine GP management,

TensCare Ltd, 89 Robin Hood Way, London, SW15 3PW.

11

analgesics and the advice contained in the Back Book. Subjects were requested to attend twice

per week, to receive a minimum of 6 and a maximum of 10 treatments over a period of 8

weeks (Hurley et al. 2000). Discharge was at the discretion of the treating therapist. Patients

who failed to attend for three successive appointments (or who requested it), were considered

noncompliant and withdrawn from the study and treated ‘as necessary’ by the same

physiotherapist, but were included in all subsequent follow-ups for the purposes of intention-

to-treat analysis. Given the nature of the treatments it was not possible to blind subjects or

therapists with respect to the content of the interventions, but the physiotherapists were

equally positive in the delivery of each protocol as recommended by Koes and Hoving (1998).

Outcome Measurement and Follow-up Procedures. A range of valid and reliable outcome

measure questionnaires and a patient-centered questionnaire (Hurley et al. 2001b) were used

to collect outcome data at baseline, discharge, 6 months and 12 months (Table 2). All follow-

ups were conducted by post (using self-addressed envelopes) and administered by the

principal investigator. Non-respondents were sent a postcard reminder after two weeks, and a

second copy of the questionnaires and self-addressed envelopes after four weeks. The

principal investigator, who assessed all the outcomes, was blind to group allocation until

completion of data analyses. The treating physiotherapists were not involved in any aspect of

outcome assessment.

Data Analysis. All data were analysed using the Statistical Package for the Social Sciences

(Windows 9.0). Analyses were performed according to the ‘intention-to-treat’ principle. An

alternative analysis that accounted for dropouts (i.e. nonrespondents) at follow-up was

conducted whereby missing values were replaced with imputed values generated by a series

of linear regression equations (Sim and Wright 2000). The comparability of groups at

12

baseline and the patient-centered questionnaire data were assessed using Chi-Square tests for

discrete nominal variables and one-way analysis of variance for continuous interval variables

if the assumptions for parametric statistics held; otherwise the Kruskal-Wallis H test was

used. Differences between two groups (i.e. respondents and non-respondents; compliers and

noncompliers) were assessed using Chi-Square tests for discrete nominal variables and

unrelated t tests for continuous interval variables; otherwise the nonparametric equivalent, the

Mann-Whitney U test was employed. For the functional disability, pain and quality of life

continuous variables change scores were calculated for the differences from baseline at

discharge, 6 months and 12 months. The effects of the interventions on these outcome

measures were estimated using univariate analyses of covariance with the change score as the

dependent variable and the baseline score as the covariate. Why not a mixed Ancova? For all

comparisons, a probability of <0.05 was considered to be statistically significant (two-tailed).

Where multiple comparisons of variables were conducted, significance levels for each

individual variable were determined, and the Bonferroni correction for multiple tests of

significance was then performed. Exploratory analysis of Roland Morris Disability

Questionnaire change scores determined if subjects had achieved the relevant minimal

clinically important difference (MCID) value according to their baseline score: 0-8 (MCID =

2), 9-16 (MCID = 4), 17-24 (MCID = 8) (Stratford et al. 1998). Don’t understand this.

Power Analysis. Sample size was determined by the statistician (M.D.) in accordance with

the procedures described by Buchner et al. (1997). Estimates of variability for the primary

endpoint were obtained from a previous RCT conducted by the researchers (Hurley et al.

2001a) that also concurred with the recommendations of the original developers of the Roland

Morris Disability Questionnaire for a change score of 2 or 3 points for sample size

calculations for RCTs (Roland and Fairbanks 2000). Calculations determined that a

13

minimum of 50 subjects would be required for each intervention group to provide at least

90% probability, at an alpha of 0.05, of detecting the MCID of 2 points in the mean change of

the Roland Morris Disability Questionnaire (in either direction) if such an effect existed.

Allowing for 15% attrition at three follow-up points increased the minimum sample size for

each group to 76 subjects; total sample size was a minimum of 228 subjects.

14

Results

Ninety-seven GPs from 36 primary care practices in Northern Ireland referred patients to the

RCT. From May 1999 to May 2000 a total of 240 subjects who met the inclusion criteria were

randomly assigned to one of three groups: (1) MT (n=80; 35 male, 45 female; mean age=39.6

years, ±SD=11.6 years) (2) IFT (n=80; 30 male, 50 female; mean age=40.2 years, ±SD=12.1

years) and (3) CT (n=80; 32 male, 48 female; mean age=40.5 years, ±SD=11.3 years). The

trial flow diagram is shown in Figure 2; five subjects deemed ineligible after randomisation

were excluded from the study and subsequent analysis as recommended by Pocock (1983).

One patient randomly assigned to the IFT group was mistakenly treated according to the MT

protocol and consequently 234 subjects received treatment as allocated. Subjects received an

average of five physiotherapy treatments (±SD=2.5; range = 1 to 10), over a period of five

weeks (±SD=2.3; range = 1 to 10), and there were no significant differences between groups

for the number of treatments (P = 0.62) or the number of weeks of treatment (P = 0.84)

received.

No adverse effects of treatment were reported in any of the study groups. Follow-up

data were obtained from 194 (83%) subjects at discharge, 166 (71%) at 6 months and 158

(67%) at 12 months. Respondents and nonrespondents were comparable for all baseline

variables apart from smoking status with a higher percentage of current smokers within the

nonrespondent (NR) group compared to respondents (R) at 6 months (NR:50.7%, n=35;

R:31.3%, n=52; P = 0.02) and 12 months (NR:50.6%, n=41; R:29.8%, n=46; P = 0.007). One

patient died between the discharge and 6 month follow-up points due to a cause unrelated to

LBP or physiotherapy.

15

Subject characteristics

Sociodemographic, clinical characteristics and outcome measure scores at baseline

were well balanced for the three arms of the trial (Table 3).

Treatment compliance

A total of 35 subjects (15%) were considered noncompliant with the study protocol

and there was no difference between groups for the level of noncompliance (2

= 1.91, df = 1,

Pp = 0.39) as illustrated in Figure 2. Noncompliers were significantly younger (mean=33.31

years, ±SD=10.53 years) than those who adhered to the protocol (mean =41.01 years,

±SD=11.2 years; t (df) = -3.78, P = 0.001; 95% CI of difference = –11.71 to –3.69).

Outcomes

Table 4 shows the mean change in outcome measures over time for each group from

randomization to follow-up at discharge, 6 months and 12 months. The values of the mean

adjusted change scores suggested that subjects in all groups had experienced clinically

meaningful improvements in functional disability, pain and quality of life at discharge, which

were largely maintained at 6 and 12 month follow-up (Figure 3). The results of univariate

analysis of covariance found no significant differences between groups in the magnitude of the

change scores at discharge, 6 months and 12 months apart from SF-36 Physical Functioning (P

= 0.03), Bodily Pain (P=0.04), and Mental Health (P = 0.03) at 12 months. Pairwise

comparisons found significantly greater differences and 95% confidence intervals in favour of

CT over MT for SF-36 Physical Functioning (mean difference = -12; 95% CI of difference = -

23.59 to –0.41; P = 0.04; Figure 4) and Bodily Pain (-13; -24.55 to –0.62; P = 0.036) scales,

and for CT over IFT in the Mental Health (-9.5; -18 to –0.97; P = 0.023) scale.

16

While there was no significant difference between groups for the primary outcome, the

Roland Morris Disability Questionnaire, the majority of subjects in all groups displayed the

minimal clinically important difference value of at least 3 points for this measure at each

follow-up point (Discharge: 67% MT, 60% IFT, 67% CT; 6 months: 64% MT, 71% IFT, 67%

CT; 12 months: 77% MT, 74% IFT, 92% CT). Table 5 shows that the mean and 95%

confidence interval of the RMDQ change scores for each group were in accordance with the

recommended minimal clinically important difference values for each baseline score category;

i.e. 0-8, 9-16, 17-24 (Stratford et al. 1998).

Approximately 70% of subjects in each group reported recurrences at 12 months (77%

MT, 69% IFT, 64% CT) and there was no significant difference between groups (2= 2.06; df

= 2; pP = 0.36). The rate of work absenteeism for employed subjects was similar among the

groups at 12 months (none: 79% MT, 78% IFT, 82% CT, < 30 days: 10% MT, 14% IFT, 6%

CT, > 30 days: 12% MT, 8% IFT, 12% CT; 2

= 2.08; df = 4; P = 0.72). The percentage of

subjects reporting participation in exercise was similar (73% MT, 77% IFT, 72% CT; 2

=

0.36; df = 2; P = 0.84), as were the percentages that reported additional treatment for LBP

(33%, 29%, 24%, respectively; 2

= 3.04; df = 4; P = 0.55). The most commonly sought

treatments were visits to general practitioners (72%; n=33), NHS physiotherapy (21%; n=10),

private chiropractic (14%; n=6), and hospital consultants (9%; n=4). There were no

significant differences between groups for analgesic medication usage at discharge (56%,

45%, 48%, respectively; 2

= 1.57; df = 2; P = 0.47) or 12-month follow-up (46%, 42%, 32%,

respectively; 2

= 2.26; df = 2; P = 0.32).

Alternative analyses

An intention-to-treat analysis is considered most valid when the dropout rate and

number of missing values is low and there is no contamination bias (Koes et al. 1992). In this

17

study the results of the intention-to-treat analysis may be biased because of dropouts and

missing values. Therefore, an alternative analysis was conducted in which missing values

were replaced by values generated from a series of linear regression equations. The main

underlying assumption for this analysis is that subjects’ previous scores are used to determine

a predicted value that reduces the variance of the value for each variable (Sim and Wright

2000). The results of the alternative analysis were very similar to the intention-to-treat

analysis. Univariate analysis of covariance found no significant differences between groups in

the magnitude of the change scores at discharge, 6 months and 12 months apart from SF-36

Physical Functioning (P = 0.013), Bodily Pain (P=0.018), and Mental Health (P = 0.003) at 12

months. Pairwise comparisons found significantly greater differences and 95% confidence

intervals for SF-36 Physical Functioning in favour of CT over MT (mean difference = -10.53;

95% CI of difference = -20.86 to –0.19; P = 0.04) and CT over IFT (-11.38; -21.58 to –1.19; P

= 0.02). Similarly for SF-36 Bodily Pain the differences were in favour of CT over MT (-

11.16; -22.24 to –0.08; P = 0.04) and CT over IFT (-11.41; -22.31 to –0.51; P = 0.04). For SF-

36 Mental Health the difference was in favour of CT over IFT (-10.30; -17.56 to –3.05; P =

0.02).

18

Discussion

The main finding of this randomized clinical trial, was that MT and IFT whether used

in isolation or in combination (in addition to the Back Book), were equally effective for

patients with acute LBP. While, there were some significant differences on the SF-36 scales

at 12 month follow-up in favour of the combined therapy group, the weight of evidence

suggests that the positive short-term effects for functional disability, pain and quality of life

were well maintained at 6 month and 12 month follow-up in all groups, providing evidence

that all treatments were successful in the medium and long term. The RCT design allowed for

the control of confounders such as age, sex, duration of LBP, past history and treatment,

which could have distorted the results. Additionally, the comparable homogeneous patient

groups, blinded randomization procedure, standardized interventions and follow-up

procedures, the expertise of the physiotherapists and use of the intention-to-treat principle

should render these results valid.

Despite dropouts due to nonrespondents, a retrospective power analysis on the basis of

the 12 month follow-up data showed the study achieved power of at least 90% to detect MCID

values for the RMDQ, VAS, MPQ, SF-36 Physical Functioning, Bodily Pain, General Health,

Vitality, Social Functioning, Role Emotional, and Mental Health variables. While the sample

size exceeded the minimum established from power calculations, the large number of dropouts

(particularly at 12 months) may have represented those who did not respond well to treatment.

Thus an alternative analysis that replaced missing values with imputed values was conducted,

which largely concurred with the intention-to-treat analysis findings thus strengthening their

reliability.

The patient sample exhibited characteristics similar to those of the archetypal patient

with acute LBP who presents to the health service for treatment of this common, disabling

condition, i.e. a preponderance among the female population, mean age of 40 years, high

19

recurrence rate, and utilisation of previous treatment for LBP (Blomberg et al. 1992; Cherkin

et al. 1998a; Gracey et al. 2002) and the outcome measure baseline values were within the

limits of those described previously (Hsieh et al. 1992; Herman et al. 1994; Pope et al. 1994;

Dettori et al. 1995; Hides et al. 1996; Cherkin et al. 1998; Kind et al. 1998; Seferlis et al.

1998; Burton et al. 1999; Chok et al. 1999; Werners et al. 1999; Hurley et al. 2001a).

Given the number of variables examined the significant differences on the SF-36

Physical Functioning, Bodily Pain and Mental Health scales at 12 month follow-up in favour

of the combined therapy group may be statistical artefacts. An alternative explanation may be

related to variations between groups in the natural history during the follow-up period.

Although not statistically significant, the combined therapy group had a lower recurrence rate

compared to the manipulative therapy in particular and this may suggest some mutability of

the LBP symptoms between treatment groups (Waxman et al. 2000). Furthermore, patients in

the combined therapy group inevitably had a longer treatment time than the MT or IFT groups

and this dose effect may have contributed to the differences recorded at 12 months. Finally,

there may be uncontrolled factors unknown to the investigator that were related to these

findings. Nonetheless, the weight of evidence suggests that there was little difference

between the outcomes of the three treatment groups at 12 months.

While there is already strong evidence for the positive short-term effects of MT for

functional disability (Postacchini et al. 1988; MacDonald and Bell 1990; Delitto et al. 1993;

Blomberg et al. 1994; Erhard et al. 1994; Morton 1999) and pain (Hoehler et al. 1981; Farrell

and Twomey 1982; Arkuszewski 1986; Postacchini et al. 1988; Blomberg et al. 1992; Morton

1999), including the recommendations of clinical guidelines (Waddell et al. 1999; Koes et al.

2001), interferential therapy lacks any evidence of positive short-term effectiveness compared

to other active treatments (Werners et al. 1999; Hurley et al. 2001a). Fewer RCTs have

evaluated the effectiveness of MT at long-term follow-up of 6 months or longer as advocated

20

by the Cochrane Collaboration Back Review Group for Spinal Disorders (vanTulder et al.

1997) and as a result, clinical guidelines and consensus reports do not discuss the change of

effect over time (Skargren et al. 1998). Previous RCTs found equivalent (Berqquist-Ullman

and Larsson 1977; Arkuszewski 1986; Postacchini et al. 1988; Cherkin et al. 1998; Seferlis et

al. 1998; Skargren et al. 1998), rather than positive (Meade et al. 1990; Blomberg et al. 1994)

long-term effects for MT compared to other active treatments for reductions in disability,

pain, work absenteeism, analgesic consumption and additional health care utilisation. Given

the lack of any previous investigations of the long-term effectiveness of IFT, the findings of

this RCT contribute substantially to the evidence base by demonstrating for the first time

comparative short and long-term effectiveness for these two commonly used treatments for

adult patients with non-specific acute low back pain whether used as sole treatments or in

combination in addition to the Back Book.

Both MT and IFT are proposed to activate the large diameter, low threshold A beta

afferent fibres causing pain relief through the ‘gate control’ theory of Melzack and Wall

(1965). Similarly, activation of the descending pain suppression pathway, at segmental or

supraspinal levels is attributed to both treatments, but while MT has been shown to produce

concomitant physiological responses consistent with noradrenergic mediated

sympathoexcitation, IFT is believed to cause simultaneous opioid-mediated

sympathoinhibition (Wright and Sluka 2001). The placebo effect, a highly variable and

complex phenomenon that is influenced by a myriad of factors other than the treatment has

been reported to account for 5% to 72% of the treatment effect (Simmonds 2000) and may

partly explain the findings. While, the participating GPs and physiotherapists were trained to

be equally positive about each intervention (Koes and Hoving 1998), it was impossible to

control for the potential effect of patients’, family, friends and colleagues. However the

absence of a placebo group in this and the majority of previous RCTs of manipulative therapy

21

render this purely speculative and future researchers should endeavour to address this

problem. Additionally, all patients received a copy of the Back Book, which may have had a

potentially confounding effect on the observed outcomes, such that all patients benefited to a

certain degree.

Consistent with the recommendations of acute LBP clinical guidelines, all the

participating physiotherapists had a recognised qualification in MT and thus the

generalisability of these results are limited to therapists with similar qualifications.

Postgraduate MT education is extremely popular amongst physiotherapists in Britain and

Ireland (Foster et al. 1999; Gracey et al 2002), and these findings support continuation of this

form of training. As interferential therapy is the most widely owned and used

electrotherapeutic modality in Britain and Ireland for the physiotherapeutic management of

patients with LBP, these findings provide for the first time an evidence-based protocol upon

which to base the treatment of similar LBP patients that should benefit physiotherapy

education programmes and clinical practice alike.

Several factors need to be considered in terms of the future clinical usage of

manipulative therapy and interferential therapy including patient preference, the adverse

effects of each intervention and therapist preference. In this RCT, patients did not have a bias

towards one of the interventions, probably due to a lack of previous experience. However, in

some clinical settings patients with frequent treatment episodes may prefer a manual or

electrophysical intervention. While both MT and IFT treatments have been associated with

unwanted side effects, they are considered rare for both interventions and were absent in this

RCT. The risk of inducing cauda equina due to manipulation of the lumbar spine, estimated at

between 10 and 100 million to one (Shekelle et al. 1992) has made therapists reluctant to

utilise these techniques compared to mobilisation itself (Adams and Sim 1998; Foster et al.

1999). In Britain, IFT accounts for the highest percentage (approximately 33%) of adverse

22

reactions (e.g. burns, blisters, nausea) albeit in a small number of patients (Partridge and

Kitchen 1999; Kitchen 2000a; Kitchen 2000b). Contraindications to IFT are predominantly

based on ‘common sense’ due to the lack of evidence for current stimulation parameters

(Johnson 1999) and the likely chance of adverse effects has not been established. Anecdotally

many therapists report the use of IFT primarily to minimise treatment soreness due to MT and

its provision as a sole treatment is perceived as being unacceptable to both clinician and

patient. Further investigation of the factors and barriers that influence therapists’ clinical

reasoning in the use of these interventions is warranted. Notwithstanding the results of such

research, clinicians are challenged to consider the known advantages and disadvantages of

each intervention before determining the most optimal treatment for individual patients.

There are several limitations to the current investigation, which should be considered

and addressed where practicable in future RCTs of patients with acute LBP. Blinding of

patients was hampered, as they knew which treatment they received but every effort was made

to minimise any expectation bias about the effectiveness of each treatment protocol. The

therapists delivering the treatment were not blinded to the content of each protocol, not

unusual in RCTs comparing the effectiveness of physiotherapeutic interventions (van Tulder

et al. 1997) but detailed staff training aimed to standardise delivery of the treatment protocols

as already discussed. There was a high dropout rate despite use of numerous recommended

strategies i.e. preliminary notification letter of scheduled follow-ups, repeated postal

reminders at follow-up time points, and inclusion of a pre-paid return envelope with follow-up

questionnaires (Yammarino et al. 1991). Future RCTs may consider using repeated phone call

reminders or monetary incentives (Hsieh et al. 2002). The 15% level of non-attendance for

treatment by participants in this clinical trial was better than the 21% rate for routine patients

attending physiotherapy in the participating hospitals during the same time period. In view of

the significantly younger age of noncompliers, extra emphasis needs to be placed on educating

23

younger patients about the purpose of a RCT, and the importance of keeping treatment

appointments. While difficult within a clinical context, the potential effect of natural history

should be investigated in future RCTs involving LBP patients by the inclusion of a no

treatment group. Finally, the lack of a valid, reliable and sensitive measure of patient

satisfaction precluded its measurement in this RCT and as new tools become available (Hudak

and Wright 2000) should be routinely included to complement the existing outcomes package.

Numerous systematic reviews have already highlighted the evidence base for MT in

contrast to the lack of evidence for the effectiveness of IFT (van Tulder 1999b). The current

results contribute to this evidence base by demonstrating for the first time the comparative

short and long-term effectiveness of these two commonly used treatments for adult patients

with non-specific acute LBP whether used as sole treatments or in combination (in addition to

the Back Book). Future research should establish the cost-effectiveness of these interventions

for patients with LBP, information that will be of particular significance to commissioners of

healthcare.

24

Acknowledgements

The authors gratefully acknowledge the physiotherapists in the United Hospitals Health and

Social Services Trust who took part in this study: Yvonne Kirkpatrick, Maureen Campbell,

Stephen Doherty, Joe Fegan, Dot Gaston, Nuala Grant, Mary Halferty, Aine Hasson, Shona

Magill, Florence Mawhinney, Elizabeth McGarry, Mark McGladdery, Sarah McKay, Sean

Moran, Debbie Ross, Mary Scullion, Sandra Taggart and Margaret Walls, the general

practitioners in the Northern Health and Social Services Board Northern Ireland and Ms

Jennifer Simpson who was responsible for trial administration.

25

References

1. Adams G, Sim J.A survey of UK manual therapists’ practice and attitudes towards

manipulation and its complications. Phys Res Int 1998;3:206-227.

2. Anon. The Back Book. The Stationery Office, London.

3. Arkuszewski Z. The efficacy of manual treatment in low back pain: a clinical trial. Man

Med 1986;2:68-71.

4. Bergquist-Ullman M, Larsson U. Acute low back pain in industry. Acta Orthopaediac

Scandinavia 1977;170(suppl):1-117.

5. Blomberg S, Svardsudd K, Mildenberger F. A controlled multicenter trial of manual

therapy in low-back pain. Scandinavian Journal of Primary Health Care 1992;10:170-

178.

6. Blomberg S, Hallin G, Grann K, et al. Manual therapy with steroid injections-A new

approach to treatment with low-back pain. Spine 1994;19:569-577.

7. Brazier JE, Harper R, Jones NMB, et al. Validating the SF-36 health survey

questionnaire: new outcome measure for primary care. BMJ 1992;305:160-164.

8. Brazier J, Jones N, Kind P. Testing the validity of the EuroQol and comparing it with

the SF-36 health survey questionnaire. Qual Life Res 1993;2:169-180.

9. Bronfort G. Spinal manipulation: current state if research and its indications. Neuro Clin

North Am 1999;17:91-111.

10. Buchner A, Faul F, Erdfelder E. G power: A priori, post-hoc and compromise power

analyses for the Macintosh (Version 2.1.2), University of Trier, Germany, 1997.

11. Burton AK, Waddell G, Burtt R, et al. Patient educational material in the management

of low back pain in primary care. Hospital for Joint Diseases 1996;55:138-141.

12. Burton AK, Waddell G, Tillotson M, et al. Information and advice to patients with back

pain can have a positive effect; A randomised controlled trial of a novel educational

booklet in primary care. Spine 1999;24:2484-2491.

13. Cherkin DC, Deyo RA, Battie M, et al. A comparison of physical therapy, chiropractic

manipulation, and provision of an educational booklet for treatment of patients with low

back pain. N Engl Med J 1998;339:1021-1029.

14. Chok B, Lee R, Latimer J, et al. Endurance training of the trunk extensor muscles in

people with subacute low back pain. Phys Ther 1999;79:1032-1042.

15. Cooney M, Gallen C, Mullins G. A survey of ownership and use of electrotherapeutic

modalities in public out-patient departments and private practice in the Republic of

Ireland. Physiotherapy Ireland 2000;21 3-11.

26

16. Cyriax J. Textbook of Orthopaedic Medicine. 11th

ed. London: Balliere Tindall, 1984.

17. Delitto A, Cibulka MT, Erhard RE, et al. Evidence for use of an extension-mobilisation

category in acute low back syndrome: a prescriptive validation pilot study. Phys Ther

1993;73:216-228.

18. Dettori JR, Bullock SH, Sutlive TG, et al. The effects of spinal flexion and extension

exercises and their associated postures in patients with acute low back pain. Spine

1995;20:2303-2312.

19. Deyo RA. Low-back pain. Science America 1998:49-50.

20. Erhard RE, Delitto A, Cibulka MT. Relative effectiveness of an extension program and

combined program of manipulation and flexion and extension exercises in patients with

acute low back syndrome. Phys Ther 1994;74:1093-1100.

21. EuroQol Group EuroQol: a new facility for the measurement of health related quality of

life. Health Policy 1991;16:199-208.

22. Evans G, Richards SH. Low back pain: an evaluation of therapeutic interventions.

University of Bristol, Bristol, 1996.

23. Farrell JP, Twomey LT. Acute low back pain: comparison of two conservative

treatment approaches. Med J Aus 1982;1:160-164.

24. Foster NE, Thompson KA, Baxter GD, et al. Management of non-specific low back

pain by physiotherapists in Britain and Ireland: A descriptive questionnaire of current

clinical practice. Spine 1999;24:1332-1342.

25. Garratt AM, Ruta DA, Abdalla MI, et al. The SF 36 health survey questionnaire: an

outcome measure suitable for routine use within the NHS. BMJ 1993;306:1440-1444.

26. Gracey JH, McDonough SM, Baxter GD. Physiotherapy management of low back pain:

A survey of current practice in Northern Ireland. Spine 2002;27:406-411.

27. Herman E, Williams RW, Stratford P, et al. A randomised controlled trial of

transcutaneous electrical nerve stimulation (Codetron) to determine its benefits in a

rehabilitation programme for acute occupational low back pain. Spine 1994;19:561-568.

28. Hides JA, Richardson CA, Jull GA. Multifidus muscle recovery is not automatic

following resolution of acute first episode low back pain. Spine 1996;21:2763-2769.

29. Hoehler FK, Tobias JS, Buerger AA. Spinal manipulation for low back pain. JAMA

1981; 245:1835-1838.

30. Hsieh CJ, Phillips RB, Adams AH, et al. Functional outcomes of low back pain:

comparison of four treatment groups in a randomised controlled trial. J Manipulative

Physiol Ther 1992;15:4-9.

27

31. Hsieh CJ, Adams AH, Tobis J. et al. Effectiveness of four conservative treatments for

subacute low back pain. A randomized clinical trial. Spine 2002;27:1142-1148.

32. Hurley DA, Dusoir TE, McDonough SM, et al. Biopsychosocial screening

questionnaire for patients with low back pain: preliminary report of utility in

physiotherapy practice in Northern Ireland. Clin J Pain 2000;16:214-228.

33. Hurley DA, Minder PM, McDonough SM, et al. Interferential therapy electrode

placement technique in acute low back pain: a preliminary investigation. Arch Phys

Med Rehabil 2001;82:485-493.

34. Hurley DA, Dusoir TE, McDonough SM, et al. How Effective is the Acute Low Back

Pain Screening Questionnaire for Predicting Outcome at One-Year Follow-up in

Patients with Low Back Pain? Clin J Pain 2001;17:256-263.

35. Johnson MI. The mystique of interferential currents when used to manage pain.

Physiotherapy 1999;85:294-297.

36. Kind P, Dolan P, Gudex G, et al. Variations in population health status: results from a

United Kingdom national questionnaire survey. BMJ 1998;316:736-741.

37. Kitchen S. Audit of the unexpected effects of electrophysical agents, Interim report:

responses to December 1999. Physiotherapy 2000;86:152-155.

38. Kitchen S. Audit of the unexpected effects of electrophysical agents, Interim report:

responses January to June 2000. Physiotherapy 2000;86:511-509.

39. Koes BW, Bouter LM, van Mameren H, et al. Randomised clinical trial of manipulative

therapy and physiotherapy for persistent back and neck complaints: results of a one year

follow up. BMJ 1992;304:601-605.

40. Koes BW, Assendelft WJJ, van der Heijden GJMG, et al. Spinal manipulation for low

back pain: An updated systematic review of randomised clinical trials. Spine

1996;21:2860-2873.

41. Koes BW, Hoving JL. The value of the randomised clinical trial in the field of

physiotherapy. Man Ther 1998;3:179-186.

42. Koes BW, vanTulder MW, Ostelo R, et al. Clinical guidelines for the management of

low back pain in primary care: an international comparison. Spine 2002;15:2504-13.

43. Lindsay DM, Dearness J, Richardson C, et al. A survey of electromodality usage in

private physiotherapy practices. Aus J Physio 1990;36:249-256.

44. MacDonald RS, Bell CMJ. An open controlled assessment of osteopathic manipulation

in non-specific low back pain. Spine 1990;15:364-370.

45. Maitland GD. Vertebral Manipulation. 5th

ed. London: Butterworth-Heinemann, 1986.

28

46. Maniadakis N, Gray A. The economic burden of back pain in the UK. Pain 2000;84:95-

103.

47. Meade TW, Dyer S, Browne W, et al. Low back pain of mechanical origin: randomised

comparison of chiropractic and hospital outpatient treatment. BMJ 1990;300:1431-

1437.

48. Melzack R, Wall P. Pain mechanisms: a new theory. Science 1965;150:971-979.

49. Melzack R. The McGill pain questionnaire: major properties and scoring methods. Pain

1975;1:227-299.

50. Melzack R. The McGill Pain Questionnaire. In R. Melzack (ed) Pain measurements

and assessment. New York:Raven Press,1983.

51. Mielenz TJ, Carey TS, Dyrek DA, et al. Physical therapy utilization by patients with

acute low back pain. Phys Ther 1997;77:1040-1051.

52. Mohseni-Bandpei MA, Stephenson R, Richardson B. Spinal manipulation in the

treatment of low back pain: a review of the literature with particular emphasis on

randomised controlled clinical trials. Phys Ther Rev 1998;3:185-194.

53. Moore A. An audit of outcome of physiotherapy intervention for outpatients with back

pain against set clinical standards. Brighton: South Thames Clinical Audit Programme,

University of Brighton and Mid Kent Healthcare Trust, 1998.

54. Morton JE. Manipulation in the treatment of acute low back pain. J Man Manip Ther

1999;7:182-189.

55. Noble G. Current use of interferential therapy in physiotherapy outpatient departments

for treatment of low back pain. BSc(Hons) Thesis, University of Ulster, 1998.

56. Partridge CJ, Kitchen S. Adverse effects of electrotherapy used by physiotherapists.

Physiotherapy 1999;85:298-303.

57. Pocock SJ. Clinical trials: A practical approach. Chichester: John Wiley,1983.

58. Pope GD, Mockett SP, Wright JP. A survey of electrotherapeutic modalities: Ownership

and use in the National Health Service in England. Physiotherapy 1995;81:82-91.

59. Pope MH, Phillips RB, Haugh LD, et al. A prospective randomised three-week trial of

spinal manipulation, transcutaneous electrical nerve stimulation, massage and corset in

the treatment of subacute low back pain. Spine, 1994;19:2571-2577.

60. Postacchini F, Facchini M, Paliere P. Efficacy of various forms of conservative

treatment in low back pain: a comparative study. Neuro-Orthopaedics 1988;6:28-35.

61. Roland M, Morris R. A study of the natural history of back pain: Part I: Development of

a reliable and sensitive measure of disability in low back pain. Spine 1983;8:141-144.

29

62. Roland M, Fairbank J. The Roland-Morris Disability Questionnaire and the Oswestry

Disability Questionnaire. Spine 2000;25:3115-3124.

63. Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175-184.

64. Scott J, Huskisson EC.Vertical or horizontal visual analogue scales. Ann Rheum Dis

1978;38 560.

65. Seferlis T, Nemeth G, Carlsson AM, et al. Conservative treatment in patients sick-listed

for acute lowback pain: A prospective randomised study with 12 months’ follow-up.

Euro Spine J 1998;7:461-470.

66. Shekelle PG, Adams AH, Chassin MR, et al. Spinal manipulation for low back pain.

Ann Int Med 1992;117:590-598.

67. Sim J, Wright C. Research in Health Care: Concepts, designs and methods.

Cheltenham:Stanley Thornes Publishers Ltd, 2000.

68. Simmonds MJ. Pain and the placebo in physiotherapy: A benevolent lie? Physiotherapy

2000;86:631-637.

69. Skargren EI, Carlsson PG, Oberg BE. One year follow-up comparison of the cost

effectiveness of chiropractic and physiotherapy as primary management for back pain.

Spine 1998 23;1875-1884.

70. Spitzer WO, LeBlanc FE, Dupuis M. (eds). Scientific approach to the assessment and

management of activity-related spinal disorders; A monogram for clinicians. Report of

the Quebec Task Force of Spinal Disorders. Spine 1987;12:1-59-S33.

71. Stratford PW, Binkley JM, Riddle DL, et al. Sensitivity to change of the Roland-Morris

Back Pain Questionnaire Part 1. Phys Ther 1998;78:1186-1196.

72. Van Tulder MW, Koes BW, Bouter LM. Conservative treatment of acute and chronic

non-specific LBP, a systematic review of RCTs of the most common interventions.

Spine 1997;22:2128-2156.

73. Van Tulder MW. Point of view. Spine 1999;24:1584.

74. Waddell G. The Back Pain Revolution. Edinburgh: Churchill Livingstone, 1998.

75. Waddell G, McIntosh A, Hutchinson A, et al. Low Back Pain Evidence Review. Royal

College of General Practitioners, London, 1999.

76. Ware JE, Kosinski M, Keller SK. SF-36 Physical and Mental Health Summary Scales:

A User’s Manual. Boston: The Health Institute, 1994.

77. Waxman R, Tennant A, Helliwell P. A prospective follow-up study of low back pain in

the community. Spine 2000;25:2085-2090.

30

78. Werners R, Pynsent PB, Bulstrode CJK. Randomized trial comparing interferential

therapy with motorized lumbar traction and massage in the management of low back

pain in a primary care setting. Spine 1999;24:1579-1583.

79. Wright A, Sluka KA. Nonpharmacological treatments for musculoskeletal pain. Clin J

Pain 2001;17:33-46.

80. Yammarino FJ, Skinner SJ, Childers TL. Understanding mail survey response behavior.

A meta-analysis. Public Opinion Quarterly 1991;55 :613-639.