Thiese et al.

BMC Musculoskeletal Disorders 2013, 14:117

http://www.biomedcentral.com/1471-2474/14/117

STUDY PROTOCOL Open Access

Electrical stimulation for chronic non-specific low

back pain in a working-age population: a 12-week
double blinded randomized controlled trial
Matthew S Thiese*, Matthew Hughes and Jeremy Biggs

Abstract
Background: Non-invasive electrotherapy is commonly used for treatment of chronic low back pain. Evidence for
efficacy of most electrotherapy modalities is weak or lacking. This study aims to execute a high-quality, double-
blinded randomized controlled clinical trial comparing 1) H-WaveW Device stimulation plus usual care with 2)
transcutaneous electrical nerve stimulation (TENS) plus usual care, and 3) Sham electrotherapy plus usual care to
determine comparative efficacy for treatment of chronic non-specific low back pain patients.
Methods/Design: Patients- Chronic non-specific low back pain patients between ages of 18–65 years, with pain of
at least 3 months duration and minimal current 5/10 VAS pain. Patients will have no significant signs or symptoms
of lumbosacral nerve impingement, malignancy, spinal stenosis, or mood disorders.
Study design- Double blind RCT with 3 arms and 38 subjects per arm. Randomization by permuted blocks of
random length, stratified by Workers Compensation claim (yes vs. no), and use of opioids. The null hypothesis of
this study is that there are no statistically significant differences in functional improvement between treatment
types during and at the end of a 12-week week treatment period.
TM
Data collection- Subjective data will be collected using Filemaker Pro database management collection tools.
Objective data will be obtained through functional assessments. Data will be collected at enrollment and at 1, 4, 8,
and 12 weeks for each participant by a blinded assessor.
Interventions- H-WaveW device stimulation (Intervention A) plus usual care, transcutaneous electrical nerve
stimulation (TENS) (Intervention B) plus usual care, and sham electrotherapy plus usual care (control). Each
treatment arm will have identical numbers of visits (4) and researcher contact time (approximately 15 hours).
Outcomes- Primary outcome measure: Oswestry Disability Index. Secondary measures include: Rowland Morris
Instrument, VAS pain score, functional evaluation including strength when pushing and pulling, pain free range of
motion in flexion and extension. Outcome measures assessed at baseline, 1, 4, 8, and 12 weeks. Treatment failure
will be defined if patient terminates assigned treatment arm for non-efficacy or undergoes invasive procedure or
other excluded cointerventions. Data will be analyzed using intention-to-treat analysis and adjusted for covariates
related to LBP (e.g. age) as needed.
Discussion: Study strengths include complex randomization, treatment group allocation concealment, double
blinding, controlling for co-interventions, rigorous inclusion criteria, assessment of compliance, plans for limiting
dropout, identical assessment methods and timing for each treatment arm, and planned intention-to-treat analyses.
Keywords: Chronic low back pain, Transcutaneous electrical nerve stimulation, Double-blind randomized controlled
trial, H-wave, TENS, Usual care

* Correspondence: matt.thiese @hsc.utah.edu

Rocky Mountain Center for Occupational & Environment Health, Department
of Family and Preventive Medicine, University of Utah, 391 Chipeta Way,
Suite C, Salt Lake City, UT 84108, USA

© 2013 Thiese et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 2 of 9
http://www.biomedcentral.com/1471-2474/14/117

Background highly prescribed due to low cost and low occurrence of

Low Back Pain (LBP) is one of the most common rea- side effects [21]. Identification of electrotherapy treat-
sons for medical visits to physician offices and emer- ments with quality evidence of efficacy, low cost, and low
gency departments, with an estimated 61 Million visits adverse effect profile would be expected to result in im-
in 2007, up from an estimated 15 million visits in 1990 proved patient function, reduced pain, reduced morbidity,
[1-3]. Chronic low back is associated with significant co- improved productivity, and overall lower health care costs.
morbidity and direct health care costs due to increased In this study, we evaluate anecdotal properties of the H-
health care utilization [4]. The economic burden of low Wave W device, which is reported to be highly efficacious
back pain is estimated between $20 and $50 billion annual for many patients, and compares both economically and
cost to the U.S. economy [5]. Additionally, low back injur- in safety with achieved TENS treatment protocols.
ies result in an estimated 149 million lost workdays per The H-WaveW Device (Electronic Waveform Lab, Inc)
year and an additional $28 billion in productivity losses stimulation is an electrotherapy modality. The primary
[6]. Despite the relative pervasiveness of low back pain, difference between other electrotherapies and H-WaveW
and the enormous financial burden on the health care sys- is the waveform, with larger pulse durations and uni-
tem, chronic low back pain remains a difficult condition to quely low frequencies. H-WaveW utilizes a bipolar ex-
treat [7,8]. Identification of efficacious noninvasive, non- ponentially decaying pulse with frequencies between
pharmacologic therapies could yield meaningful gains and 1-70Hz and a pulse duration of 5 milliseconds along
result in substantial population improvement in morbidity with some other proprietary factors. The H-WaveW de-
and costs associated with LBP. A well-designed random- vice has received FDA clearance for a number of indica-
ized trial is needed to determine comparative treatment tions, including home-based treatment of chronic pain,
efficacy of electrical stimulation treatments that are rela- post-surgical pain, circulation, range of motion, and
tively inexpensive, noninvasive and nonpharmacologic. muscle spasms. The mechanism(s) of action for the
Myriad treatments have been described for chronic H-WaveW Device are unique and include nitric-oxide
LBP, yet the body of quality randomized trials demon- dependent enhanced microcirculation and angiogenesis,
strating efficacy for many commonly used treatments is stimulation and contraction of muscle fibers within the
limited [9-11]. A recent evidence based review demon- subdermal lymphatic system, and deactivation of the so-
strated that only 47 of 148 treatment interventions for dium pump within the nerve fiber [22,23]. While the
Chronic LBP have quality randomized trials as the basis exact mechanism of LBP treatment is unknown there
of recommendation for the intervention for low back are multiple mechanisms, including the one that is refe-
pain [8]. Thus, the majority of commonly used interven- renced above that may demonstrate efficacy for treating
tions have little or no evidence of efficacy. LBP. Other mechanisms include stimulation of nocire-
Specific to electrotherapy modalities, multiple devices ceptors and other nerve types to mask pain and allow
producing varying electronic waveform types, wave fre- patients to improve function.
quency, and wave amplitude are used for analgesic relief The intent of this trial is to execute a high-qualitya
of chronic pain conditions, including nonspecific low back double-blinded randomized controlled clinical trial that
pain. An evidence based review of electrotherapy literature compares the efficacy of the H-WaveW Device with a com-
identified a modest number of moderate and low quality monly used TENS device and with a control group using
trials that demonstrated evidence of transcutaneous elec- a sham electrotherapy device [24]. A well-designed, high-
trical nerve stimulation (TENS) efficacy for select patients quality trial will provide definable level of evidence for
with chronic low back pain, however there is still no con- treatment efficacy, and provide a basis for evidence-based
sensus regarding efficacy of TENS for treating chronic recommendation for or against utilization of the use of
LBP [8,12-20]. Additionally, the reviews found insufficient the H-WaveW device. The results for H-WaveW device, if
quality evidence for recommending for or against the use positive, could significantly impact morbidity by providing
of high-voltage galvanic therapy, interferential therapy, a non-invasive, non-pharmacologic treatment for symp-
iontophoresis, percutaneous electrical nerve stimulation tomatic relief, and reduce overall disability and health care
(PENS), microcurrent electrical stimulation, and sympa- costs associated with chronic low back pain.
thetic electrotherapy for treating chronic low back pain.
Similarly, although there is strong anecdotal support, Methods/Design
there are no reported quality randomized controlled trials Participants
demonstrating efficacy of the H-WaveW Device for symp- Subjects will be recruited from University and community
tomatic relief of chronic low back pain. Regardless of the practitioners, as well as through media advertisements.
evidence of efficacy of TENS in treating chronic LBP, it is Enrollment and treatment will be conducted in clinical/la-
a common modality for treating LBP due to high demand boratory facilities at the University of Utah’s Rocky Moun-
for noninvasive, nonpharmacologic interventions. It is tain Center for Occupational and Environmental Health.
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 3 of 9
http://www.biomedcentral.com/1471-2474/14/117

Potential study subjects identified through the institu- preclude successful patient participation will also be
tional review board (IRB) approved study marketing ma- excluded. Active psychiatric disorders will be excluded
terials will be screened based on inclusion and exclusion (e.g. use of antipsychotic medication, bipolar disorder,
criteria. If all criteria are met, they will be scheduled for schizophrenia). Patients diagnosed with history of signifi-
a consent, randomization, and initial treatment. cant mood disorder will be excluded (e.g., depression or
Inclusion criteria for study eligibility includes: ages anxiety with adequate control would be acceptable). Pa-
18–65 with chronic low back pain (≥3 months duration; tients currently are or who become pregnant will be
a current VAS pain rating ≥5/10; no radiating pain below excluded.
the knee; ≥75% back or buttock pain rather than lower Consent will be obtained prior to randomization. Po-
extremity pain; English speaking, and able to complete tential study subjects will be given a copy of the consent
and tolerate treatment for the study period. document and the document will be outlined in detail
Exclusion criteria include: Prior home use of H-WaveW by a research team member. Potential study subjects will
Device or TENS. Prior history of spinal fusion or failed be given opportunities to ask questions prior to signing
spinal surgery syndrome. Laminectomy, laminotomy or the consent form and enrolling in the study.
discectomy within 12 months of enrollment. Diagnostic or This study has been approved through the University
interventional injections or any low back surgeries not of Utah Institutional Review Board (IRB # 52918). Study
mentioned above, including radiofrequency neuroablation participants will receive compensation for participation
within 6 months of enrollment. Current implanted cardiac in this study (Figure 1).
demand pacemakers, defibrillators, cardiac pumps, spinal
stimulators or other implanted electronic devices. Patients Randomization and allocation procedures
using personal home based electrical stimulation devices Subjects will be allocated through central, computerized
are excluded to reduce risk of blinding failure. Patients randomization utilizing a stratified permuted-block
with other concomitant illnesses (e.g., malignancy, osteo- randomization employing random block sizes. After
porosis) which, in the opinion of the investigator, would confirming eligibility and completion of baseline data

Figure 1 Trial design, procedures, stages and data collection.

Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 4 of 9
http://www.biomedcentral.com/1471-2474/14/117

collection, the Study Coordinator will contact the desig- arm for non-efficacy or election to pursue surgical or inva-
nated investigator (MST) for allocation. He will consult sive procedure or other excluded cointerventions.
the prepared block randomized list and the next alloca- The primary outcome measure will be the Oswestry
tion contained in sealed opaque envelope will be given Disability Index (ODI). The ODI is widely used in low
to the treatment provider, while also recording the sub- back pain intervention trials, and provides a measure of
ject ID number to allow for ascertainment of compli- functional improvement The ODI was selected com-
ance with treatment allocation. The three treatment pared to a subjective pain rating [25-29]. Secondary out-
arms are labeled using letters “L”, “W” and “E” and come measures include Rowland Morris Instrument
correspond to a labeled device. The names of particular (0–24 scale of disability); VAS Pain score (0–100); phys-
devices will not be used. The treatment provider will ical activity levels objectively measured using accelerom-
open the envelope and provide the assigned treatment eters worn by participants during waking hours over
device to the participant. Block stratification will be during the entire 12 week study period; functional cap-
performed for opioid use (current vs. past/never) and acity evaluation including strength when pushing and
workers’ compensation claim (yes vs. no). pulling (3 repetitions lasting 5 seconds each of both
pushing and pulling measuring both peak and average
Blinding kg of force according to standard EvalTech Functional
We will attempt to blind patients and assessors to Systems protocol which utilizes a load cell to measure
achieve double blinding. Patients will not be told what forces exerted); range of lumbar spine motion (up to parti-
electrotherapy techniques are being utilized in this cipants maximum range of specific movement) in flexion,
study, only that there are two active electrotherapies out extension, lateral rotation; duration of time (minutes) sit-
of the spectrum of electrotherapy devices and a sham ting comfortably (subjective estimate made by participant
group. All study materials will be void of the words H- during questionnaire); duration of time (minutes) standing
WaveW or TENS. Patient blinding will be accomplished comfortably (subjective estimate made by participant
by providing all patients either a H-WaveW device, a during questionnaire); duration (minutes) and distance
fully operation TENS Unit modified to function within a (meters) walking comfortably (subjective estimate made
H-WaveW device housing, or a sham electrotherapy de- by participant during questionnaire); summed VAS pain ×
vice with minimally perceptible electrical current output days rating; days of impairment (days); lost workdays
modified to function within a H-WaveW device housing. (days); time to exacerbation (days until increase 3/10
Researchers providing the treatment are not blinded. As- pain); rescue acetaminophen tablets consumed (% and
sessors are blinded and will not be present for the inter- number); NSAID discontinuation (%); opioid discontinu-
vention. Participants will be instructed not to reveal or ation (%). The pain VAS score is assessed immediately be-
discuss the type of treatment characteristics they are fore and after treatment at weeks 0, 1, and daily scores pre
experiencing. and post home-treatment in a diary that the participant
Sham devices will appear to operate in the same exact keeps (Table 1).
way as a fully functional electrotherapy device, including Functional assessments will be completed using a stan-
battery drain and usage log. At maximal settings, weak dardized protocol on a BTE EvalTech system. Standard
electrical current will be flowing to the pads so that the protocols for Lumbar Flexion/Extension, Lumbar Lateral
patients may perceive the device as active treatment. Flexion, Shoulder Height Push and Pull, Cart Height Push
Additional blinding of subjects or treatment within and Pull are used. Researchers performing assessments
this trial is not possible. The research team will be re- are trained to perform all assessments exactly the same
quired to wear color-coded nametags indicating whether following the BTE EvalTtech protocol (Figure 2).
or not the researcher is blinded or non-blinded to avoid Subjective outcomes data will be collected using com-
TM
any confusion at the time of assessments. Efficacy of as- puterized data collection tools utilizing Filemaker Pro
sessor blinding will be conducted at all onsite assess- database management to assure complete and accurate
ments by recording assessor’s opinion of treatment arm data collection. Data will be collected at enrollment and
for each subject. subsequent treatments. Data after treatments end will be
collected in person using the same tools for every par-
Outcome measures ticipant study by a blinded assessor.
Outcome measures will be assessed at baseline, 1 week,
4 weeks, 8 weeks, and 12 weeks, with the primary out- Assessments
come being at 4 weeks. Assessment will occur in the same Timing of assessments is identical in each arm. Assess-
manner and at the same time intervals for all treatment ments will be conducted by blinded assessors. Enrollment,
arms utilizing standardized methods. Treatment failure randomization and initial treatment will occur at the time
will be defined as if patient terminates assigned treatment of enrollment. All participants will return 1 week after
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 5 of 9
http://www.biomedcentral.com/1471-2474/14/117

Table 1 Outcome measures measures which are assessed via electronic device
Primary outcome measure: (Actigraph GT3X + Accelerometers and EvalTech func-
Oswestry Disability Index (ODI) tional testing equipment manufactured by BTE Tech-
Secondary measures: nologies). The accelerometers provide physical activity
a. Rowland Morris Instrument (0–24 scale of disability)
levels on a per minute basis and store up to 40 days of
data. These accelerometers have demonstrated validity
b. VAS Pain score (0–10)
and reliability in a variety of settings [30-38]. Functional
c. Physical activity levels (objectively measured using accelerometers)
strength and lumbar range of motion will be captured
d. Functional evaluation including strength when pushing (peak and
average kg of force)
using EvalTech functional evaluation tools. These in-
clude fully modifiable load cells allowing different
e. Functional evaluation including strength when pulling (peak and
average kg of force) heights, widths, and orientations, as well as electronic
f. Comfortable (up to maximum baseline pain level) range of motion
range of motion devices for adequate capture. All trials
in flexion (degrees from neutral) will be performed 3 times. Pushing and pulling strength
g. Comfortable (up to maximum baseline pain level) range of motion will be three trials each with exertions for 5 seconds and
in extension (degrees from neutral) a 20 second break between trials. These will be at both
h. Duration of time sitting comfortably (up to maximum baseline pain shoulder height and cart height (approximately 49
level) (minutes) inches). There will be a minimum of 6 trials (3 push and
i. Duration of time standing comfortably (up to maximum baseline 3 pull at both shoulder height and cart height). Peak and
pain level) (minutes)
average forces will be recorded for each trial.
j. Duration and distance walking comfortably (up to maximum
baseline pain level) (minutes and meters)
Treatment protocols
k. Summed VAS pain × days rating
Subjects will be allocated to one of the three arms. Each
l. Days of impairment (days) treatment arm will have identical numbers of visits and
m. Lost workdays (days) contact time. Each arm will receive the uniform ins-
n. Time to exacerbation (days until increase 3/10 pain) tructions for using of their assigned electrotherapy device.
o. Rescue acetaminophen tablets consumed (% and number) Patients will be instructed to use their assigned device
p. NSAID discontinuation (%) daily upon waking and before bedtime, and up to two add-
q. Opioid discontinuation (%) itional treatments during the day as needed (e.g., 4 1-hour
treatment sessions). Providers will place electrodes and

initial treatment for assessment of subject ability to cor-

rectly use their assigned device; have questions answered
about the protocol, and to complete week 1 assessment.
Subjective assessments will be measured at baseline, 1 -
week, 4 weeks, 8 weeks, and 12 weeks. Objective assess-
ments will be measured pre and post-treatment at
baseline and post-treatment at week 1, week 4, and week
12. No treatments are provided by the researchers after
week 1, although patients will be performing home treat-
ments per protocol for the entire 12 week period. Compli-
ance will be assessed using digital time data logs recorded
automatically by each device and diaries kept by partici-
pants. Participants in each arm will have identical and
comparable contact time with providers and assessors to
minimize treatment bias. All participants will be contacted
weekly by a blinded assessor via telephone to encourage
compliance and use of daily treatment diaries.
At each assessment, all participants receive the same
assessment, including questionnaire, physical functio-
ning assessment form, and recording of accelerometer
use. All measures will be assessed using computerized
data collection methods at baseline and each follow-up,
Figure 2 Shoulder height push assessment using BTE
except at the 8-week follow-up where outcomes will be
EvalTech system.
assessed over the phone. There are four objective
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 6 of 9
http://www.biomedcentral.com/1471-2474/14/117

activate the device as per the protocol for each arm. effectiveness for LBP, however this will likely blind the
Patients will be told that it is possible to not immediately patients because none of them have ever experienced
feel effects of electrical treatment due to current levels the H-Wave Device. Blinding of patients is achieved as
that do not meet sensory thresholds. Patients will return none will have previously experienced H-Wave, home
after 1 week for reevaluation on electrode placement and based TENS therapy, or subtherapeutic electrotherapy.
ability to achieve the required self-administered stimula- Further, patients are kept blind as to which electrother-
tion levels. Providers are trained in standardized fashion apy devices out of the myriad devices on the market, are
to educate each participant on their assigned therapy. included in the study and therefore have few precon-
The H-WaveW device delivers electrical stimulation ceived notions to their assigned device.
through the placement of 4 electrode pads in a rectangu-
lar geometric pattern at the circumference of the low Compliance
back pain nidus. Patients will be instructed on optimal Compliance will be assessed by device logs and diaries.
electrode placement for home use. The H-WaveW Device Each device will be capable of recording treatment appli-
is controlled by the patient after instruction from a ther- cations and duration, and will be downloaded electronic-
apist or practitioner. The stimulation parameters in ally at each in-person assessment. Total treatment time
terms of pulse duration and frequency are proprietary of 3.5 hours per week will be set as the threshold for
for the manufacturer and thus are not included in this compliance. If VAS Pain drops to a daily average of less
protocol. However, the maximal intensity offered by the than or equal to 2 out of 10, patients will be considered
device is the goal of the treatment intensity. Patients are in compliance regardless of the number of treatments
instructed to achieve the highest intensity possible in completed on that day. Patients will be called once per
graduated fashion. Lower intensities are accepted if the week to promote compliance and answer any questions.
patient cannot comfortably reach the maximal device
intensity and will be noted on diary logs. There are con- Discontinuation
trols for intensity and frequency, which the patient Subjects will be encouraged to continue to follow-up
manipulates according to specific instructions and toler- throughout the study. There are no penalties for early
ances. Patients in this arm will be required to reach near dropouts other than the amount of incentive earned is
maximal intensity on low frequency. The maximal inten- reduced if the entire follow-up period is not completed.
sity may, in rare cases, induce transient muscular con- Each device (H-WaveW, TENS) has very low adverse ef-
tractions. Patients are instructed not to increase the fect profile. Although unexpected, those developing an
intensity to a painful level. The treatment is then self- adverse event will be reported to the IRB.
administered for the duration of the study.
There are no accepted standardized protocols found in Usual care and co-interventions
the literature for TENs treatment of chronic low back Study participants will be asked to avoid significant
pain. There are no trials that demonstrate maximum ef- changes in medical/healthcare management of their LBP
ficacy of pulse duration, wave type, or frequency. There- during the study period. All participants will be allowed
fore, the TENS units in this study allow adjustment of to pursue usual care with their own personal treating
amplitude (intensity) only at a fixed frequency of 100 Hz physician. No other treatment other than the assigned
with steady state pulsing. Modulation and pulse modes electrotherapy intervention will be provided by the trial
will be disabled in order to maintain uniformity of elec- researchers. Usual care will be limited to non-invasive
trotherapy protocols. Pad placement will be directed by therapies, including medications such as NSAIDS (pre-
the provider to provide maximal clinical efficacy. The scription or over the counter) and opioids, physical
device is two channel TENS and will be administered by therapy, manipulation, stretching and aerobic exercise.
the patient after training by the provider. Self-applications of heat or ice are also acceptable. All
The sham or control device will appear to operate in usual care medications and interventions will be recorded.
the same exact way as a fully functional TENS device,
including battery drain and usage log. Subtherapeutic Interventions not allowed
electrical current will be flowing to the pads so that the Any lumbosacral invasive procedure such as discectomy,
patient may perceive it as an active therapy. They will be microdiscetomy, laminectomy, lumbar fusion, disc desic-
instructed to use the device for the same duration as ac- cation by any chemical or thermal means. No injections
tive treatments, daily upon waking and before bedtime, into the lumbosacral area other than localized trigger
and up to two additional treatments during the day as point without the use of glucocorticoids. No other elec-
needed (e.g., 4 1-hour treatment sessions). There is no trical therapies (PENS, Interferential therapy). No use of
quality evidence that subtherapeutic electrical current oral or systemic glucocorticosteroids. TENS and H-WaveW
provides any clinical benefit over no electrotherapy are not allowed except in the arms assigned to those
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 7 of 9
http://www.biomedcentral.com/1471-2474/14/117

treatments. Patients are asked about any additional treat- later on in the follow-up, while the intervention least
ments received at each assessment to determine con- effective to have dropouts early on. The analysis will
tinued eligibility. emphasize the 4-week time point as the primary end
point, which ends the intensive intervention period. At
Statistical approach 4 weeks compliance to the protocol and complete
Statistical power/Sample size follow-up should be at a maximum. The remaining time
Sample size determination is based on the comparisons to 12 weeks will measure long term effectiveness and ac-
of the study arms for the primary outcome variable, ceptance of the study interventions.
ODI. In previous studies of low back pain patients , the
baseline ODI values for one study group were mean Secondary analyses
(SD) of 60.2 (15.0) and for a second study group were The changes from baseline to each follow-up visit in the
59.9 (14.4) [39,40]. The total score for the ODI is a per- pain VAS (the main secondary endpoint) will be ana-
centage, ranging from 0 to 100. It is suggested that the lyzed using the same longitudinal model described above
reliability of the ODI requires a minimum detectable for the Oswestry score. Key secondary comparisons will
change of at least 10 points, as any smaller difference include the three pair wise comparisons between the
may be attributable to error in the measurement [25]. treatment groups in the mean change in the Roland
Assuming equality of the study arms at baseline, which Morris Instrument, VAS, strength, and flexibility mea-
can be assured by adjusting for baseline ODI in the re- sures at the different time points.,
gression model, and a minimal detectable effect size of a The changes to 12 weeks will evaluate the persistence
10 point difference on the post-treatment value, sample of treatment effects of the interventions to that time;
size is based on a comparison of mean (SD) of 60 (15) changes to 4 weeks evaluate treatment effects at the
vs. 50 (15), a 17% relative difference post treatment, completion of the intensive intervention period, and the
using a two-sided comparison with alpha of 0.025 to slopes from weeks 4 to 12 evaluate changes in the treat-
allow for multiplicity adjustment. Under these assump- ment effects after 4 weeks. The null hypothesis is that
tions, a sample size of n = 43 subjects per study arm is there is no difference between the three treatment arms
required to achieve 80% power. However, it is expected at 12 weeks.
that the stimulation arms will produce at least a 25%
relative improvement over the control arm. Assuming Intent-to-treat analysis
60 (15) vs. 45 (15), a 25% relative difference, using a These data will be analyzed using an intent-to-treat ana-
two-sided alpha 0.025 comparison, a sample size of n = lysis. As H-WaveW Device is not widely utilized in study
20 subjects per study arm is required for 80% power, or catchment area it is doubtful that individuals will inad-
n = 25 subjects for 90% power. Allowing for a 20% drop- vertently crossover to that arm from the controls. How-
out rate, a sample size of n = 38 subjects will be enrolled ever, that possibility will be both discouraged and
in each study arm to provide n = 25 evaluable subjects tracked.
for data analysis.
Multiplicity
Statistical analysis A multiplicity adjustment will be made for the compari-
Primary analysis son of the primary outcome variable, Oswestry Disability
The primary analysis is the comparison of the three Index (ODI). The two-sided p values from the hypoth-
study arms on the primary outcome, Oswestry Disability esis tests will be adjusted for two comparisons using the
Index (ODI). Given that repeated measurements are col- Hochberg procedure [41]. With that procedure, the
lected, a mixed effects linear regression model will be fit- smaller of the two p values is compared against alpha =
ted with an unstructured covariance matrix to account 0.025 and the larger p value is compared against alpha =
for lack of independence introduced by the repeated 0.05. This maintains the overall alpha at 0.05 for the ef-
measurements. In this model, the repeated measure- fectiveness win strategy evaluation.
ments are nested within study subject (weeks 0, or base-
line, and weeks 1, 4, 8, and 12). The primary predictor is Co-variates
study group. A time variable will be included as a covari- Age, gender, tobacco, undiagnosed anxiety, undiagnosed
ate, since visits will not always be at the planned inter- depression, past back pain history, prior and current
vals. Other covariates will be assessed for possible NSAID use, rescue acetaminophen use.
confounding in the relationship between treatment arm
and outcome. It is likely that the pattern of dropouts will Post-hoc analyses
vary between the study groups, where the study inter- Post-hoc analyses will be conducted to attempt to ascer-
ventions that are most effective having the dropouts tain what factors predicted better responders among the
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 8 of 9
http://www.biomedcentral.com/1471-2474/14/117

two treatment arms. Potential prognostic factors include maintain blinding of the assessors. Additionally, this
gender, tobacco, anxiety, depression, (age and past back protocol outlines concealing the treatment allocation by
pain history will be assessed for residual confounding, using opaque envelopes that appear identical. The utili-
but controlled in study design), belief in effectiveness of zation of a complex randomization schedule is another
the treatments, NSAID use, rescue acetaminophen use. strength that may ensure equality of treatment groups at
baseline. The intent-to-treat analysis may account for
Other considerations unequal dropout between groups and any contamination
The primary and main secondary analyses will be per- that may occur.
formed using an intent-to-treat strategy in which patients Potential weaknesses include the possibility of high
are analyzed based on their randomized assignment irre- dropout in one or more treatment arm. To combat this,
spective of their compliance to the treatment protocol. we are offering incentives and encouraging participants
Formal interim analyses of efficacy will not be conducted. to continue for the entire duration of the study. There is
However, the operational aspects of the study (recruit- the possibility of a randomization failure, which cannot
ment, retention, adherence rates, and rates of missed be overcome. Additionally, due to the nature of the
visits) will be monitored as the study progresses so that treatment, specifically the differences between the ex-
corrective action may be taken to address any unexpec- perimental treatment and the TENS treatment, it was
ted difficulties. Monitoring for safety will be conducted not feasible to blind the interventionalist. To address
throughout the study, investigating all possible unantici- this all interventionalists will undergo strict training and
pated safety events, including increased pain and discom- re-standardization based on the treatment protocol. This
fort which are the two most likely types of events to training will stress treating all patients similarly with
occur. These treatments are widely used and are consid- regards to demeanor and contact time, regardless of
ered safe with low rates of adverse effects. Efforts will be their randomized treatment arm.
made to minimize the rates of missing data, including
telephone follow-up of missed visits. The restricted max- Endnotes
imum likelihood procedure of the primary analysis pro- a
Based on the ACOEM criteria, this study protocol is
duces approximately unbiased estimates of the treatment intended to provide a high-quality score (Randomization –
effects so long as data are missing at random [42] and is 1, Blinded Allocation – 1, Baseline Comparability – 1, Pa-
less sensitive to missing data than analyses restricting to tient blinding – 1, Interventionist blinding – 0, Assessor
complete data [43]. In sensitivity analyses, multiple imput- Blinding – 1, Control of co-interventions – 0.5, Compli-
ation incorporating baseline and follow-up predictors of ance – 1.0, Drop Out < 20% – 1, Timing of assessments –
outcome [44] will be used to assess the effect of potential 1, ITT analysis – 1 (Total score 9.5/11) High-quality ≥ 8.0.
violations of the missing at random assumption.
Competing interests
The authors declare they do not have any competing interests.
Discussion
Chronic low back pain is a significant issue, resulting in Authors’ contributions
significant costs, lost productivity, and morbidity [1-6]. All listed authors contributed to the conception and design of the study.
All authors were involved in drafting the manuscript and read and approved
Therefore, even small improvements in treatment effi- the final manuscript.
cacy, particularly for treatments with few side effects,
can have a meaningful impact on improving LBP. While Sponsor
Electronic Waveform Lab, Inc.
there have been some studies assessing the therapeutic
effectiveness of TENS, the effect generally is relatively Received: 7 August 2012 Accepted: 20 March 2013
small improvements in chronic LBP as compared to Published: 28 March 2013
sham treatment. The use of two comparison arms in this
References
study, both TENS and sham treatment, allows for com- 1. Deyo RA: Promises and limitations of the Patient Outcome Research
parisons of experimental device with both TENS as well Teams: the low-back pain example. Proc Assoc Am Physicians 1995,
as sham treatment. 107(3):324–328.
2. Friedman BW, et al: Diagnostic testing and treatment of low back pain in
Blinding within this study is a significant strength. United States emergency departments: a national perspective. Spine
All devices are identical in appearance, weight, display, (Phila Pa 1976) 2010, 35(24):E1406–E1411.
charge time, and sound. It is anticipated that complete 3. Licciardone JC: The epidemiology and medical management of low back
pain during ambulatory medical care visits in the United States.
blinding of the assessors and patients will be achieved. Osteopath Med Prim Care 2008, 2:11.
Care will be taken to limit contact between participants, 4. Gore M, et al: The burden of chronic low back pain: clinical
e.g. schedule participants one at a time so they are not comorbidities, treatment patterns, and health care costs in usual care
settings. Spine (Phila Pa 1976) 2012, 37(11):E668–E677.
waiting in the same room together. All treatment ques- 5. Pai S, Sundaram LJ: Low back pain: an economic assessment in the
tions will be directed to the interventionalists, as to United States. Orthop Clin North Am 2004, 35(1):1–5.
Thiese et al. BMC Musculoskeletal Disorders 2013, 14:117 Page 9 of 9
http://www.biomedcentral.com/1471-2474/14/117

6. Maetzel A, Li L: The economic burden of low back pain: a review of 31. Brage S, et al: Reexamination of validity and reliability of the CSA monitor
studies published between 1996 and 2001. Best Pract Res Clin Rheumatol in walking and running. Med Sci Sports Exerc 2003, 35(8):1447–1454.
2002, 16(1):23–30. 32. Crouter SE, Clowers KG, Bassett DR Jr: A novel method for using
7. Fourney DR, et al: Chronic low back pain: a heterogeneous condition accelerometer data to predict energy expenditure. J Appl Physiol 2006,
with challenges for an evidence-based approach. Spine (Phila Pa 1976) 100(4):1324–1331.
2011, 36(21 Suppl):S1–S9. 33. Freedson PS, Melanson E, Sirard J: Calibration of the Computer Science and
8. Hegmann KT, Thiese MS, Hughes MA, Edwards H, Ott U, Robertson D, Russo Applications, Inc. accelerometer. Med Sci Sports Exerc 1998, 30(5):777–781.
R, Srisukhumbowornchai S, Frischknecht M, Ording JA, Turner GE, Machata 34. Fudge BW, et al: Estimation of Oxygen Uptake during Fast Running Using
N: Update. Chapter 6: Chronic Pain. In American College of Occupational Accelerometry and Heart Rate. Med Sci Sports Exerc 2007, 39(1):192–198.
and Environmental Medicine’s Occupational Medicine Practice Guidelines. 2nd 35. Hendelman D, et al: Validity of accelerometry for the assessment of
edition. Edited by Hegmann K. Elk Grove Village, IL:; 2008. moderate intensity physical activity in the field. Med Sci Sports Exerc 2000,
9. Arnau JM, et al: A critical review of guidelines for low back pain 32(9 Suppl):S442–S449.
treatment. Eur Spine J 2006, 15(5):543–553. 36. Matthews CE, et al: Sources of variance in daily physical activity levels as
10. Schnitzer TJ: Update on guidelines for the treatment of chronic measured by an accelerometer. Med Sci Sports Exerc 2002, 34(8):1376–1381.
musculoskeletal pain. Clin Rheumatol 2006, 25:22–29. 37. Swartz AM, et al: Estimation of energy expenditure using CSA
11. Kamper SJ, et al: Treatment-based subgroups of low back pain: a guide accelerometers at hip and wrist sites. Med Sci Sports Exerc 2000,
to appraisal of research studies and a summary of current evidence. Best 32(9 Suppl):S450–S456.
Pract Res Clin Rheumatol 2010, 24(2):181–191. 38. Yngve A, et al: Effect of monitor placement and of activity setting on the
12. Brosseau L, et al: Efficacy of the transcutaneous electrical nerve MTI accelerometer output. Med Sci Sports Exerc 2003, 35(2):320–326.
stimulation for the treatment of chronic low back pain: a meta-analysis. 39. Jarzem PF, et al: Transcutaneous electrical nerve stimulation [TENS] for
Spine 2002, 27(6):596–603. short-term treatment of low back pain- randomized double blind
13. van Middelkoop M, et al: A systematic review on the effectiveness of crossover study of sham versus conventional TENS. Journal of
physical and rehabilitation interventions for chronic non-specific low Musculoskeletal Pain 2005, 13(2):11–17.
back pain. Eur Spine J 2011, 20(1):19–39. 40. Khadilkar A, et al: Transcutaneous electrical nerve stimulation (TENS) for
14. Buchmuller A, et al: Value of TENS for relief of chronic low back pain with chronic low-back pain. Cochrane Database Syst Rev 2005, 3, CD003008.
or without radicular pain. Eur J Pain 2012, 16(5):656–65. 41. Wright S: Adjusted p-values for simultaneous inference. Biometrics 1992,
15. Furlan AD, et al: A systematic review and meta-analysis of efficacy, 48(4):1005–1013.
cost-effectiveness, and safety of selected complementary and alternative 42. Little RJA: Statistical Analysis with Missing Data Second Ed. New York: Wiley &
medicine for neck and low-back pain. Evidence-Based Complementary and Sons; 2002.
Alternative Medicine 2011, 2012. 43. Tukey JW: Exploratory Data Analysis. Reading, MA: Addison-Wesley; 1977.
16. Khadilkar A, et al: Transcutaneous electrical nerve stimulation for the 44. Brinkhaus B, et al: Acupuncture in patients with chronic low back pain: a
treatment of chronic low back pain: a systematic review. Spine 2005, randomized controlled trial. Arch Intern Med 2006, 166(4):450–457.
30(23):2657–2666.
17. Milne S, et al: Transcutaneous electrical nerve stimulation (TENS) for doi:10.1186/1471-2474-14-117
chronic low back pain (Cochrane Review). The Cochrane Library 2002, 2. Cite this article as: Thiese et al.: Electrical stimulation for chronic non-
18. Khadilkar A, et al: Transcutaneous electrical nerve stimulation (TENS) for specific low back pain in a working-age population: a 12-week double
chronic low-‐back pain. The Cochrane Library 2008(4):CD003008. blinded randomized controlled trial. BMC Musculoskeletal Disorders 2013
14:117.
19. Milne S, et al: Transcutaneous electrical nerve stimulation (TENS) for
chronic low-‐back pain. The Cochrane Library 2001(2):CD003008.
20. Poitras S, Brosseau L: Evidence-informed management of chronic low
back pain with transcutaneous electrical nerve stimulation, interferential
current, electrical muscle stimulation, ultrasound, and thermotherapy.
Spine J 2008, 8(1):226–233.
21. Williams CM, et al: Low back pain and best practice care: A survey of
general practice physicians. Arch Intern Med 2010, 170(3):271.
22. Smith TL, et al: H-Wave induces arteriolar vasodilation in rat striated
muscle via nitric oxide-mediated mechanisms. J Orthop Res 2009,
27(9):1248–1251.
23. Blum K, et al: The H-Wave((R)) Device Induces NODependent Augmented
Microcirculation and Angiogenesis, Providing Both Analgesia and Tissue
Healing in Sports Injuries. Phys Sportsmed 2008, 36(1):103–114.
24. Harris JS, et al: Methodology to update the practice recommendations in
the American College of Occupational and Environmental Medicine's
Occupational Medicine Practice Guidelines. J Occup Environ Med 2008,
50(3):282–295.
25. Fairbank JC, Pynsent PB: The Oswestry disability index. Spine 2000,
25(22):2940–2953.
26. Roland M, Fairbank J: The Roland–Morris disability questionnaire and the Submit your next manuscript to BioMed Central
Oswestry disability questionnaire. Spine 2000, 25(24):3115–3124. and take full advantage of:
27. Gronblad M, et al: Intercorrelation and Test—Retest Reliability of the
Pain Disability Index (PDI) and the Oswestry Disability. Clin J Pain 1993,
• Convenient online submission
9(3):189–195.
28. Smeets R, et al: Measures of function in low back pain/disorders: Low • Thorough peer review
Back Pain Rating Scale (LBPRS), Oswestry Disability Index (ODI), • No space constraints or color figure charges
Progressive Isoinertial Lifting Evaluation (PILE), Quebec Back Pain
Disability Scale (QBPDS), and Roland-‐Morris Disability Questionnaire • Immediate publication on acceptance
(RDQ). Arthritis Care Res 2011, 63(S11):S158–S173. • Inclusion in PubMed, CAS, Scopus and Google Scholar
29. McCormick JD, Werner BC, Shimer AL: Patient-reported Outcome • Research which is freely available for redistribution
Measures in Spine Surgery. J Am Acad Orthop Surg 2013, 21(2):99–107.
30. Bassett DR Jr: Validity and reliability issues in objective monitoring of
physical activity. Res Q Exerc Sport 2000, 71(2 Suppl):S30–S36. Submit your manuscript at
www.biomedcentral.com/submit