Research Briefs

Edited by Meredith Wallace

Meredith Wallace, PhD, RN, CS-ANP, is an Assistant Professor
of Nursing at Southern Connecticut State University, New Ha-
ven, CT.

The Effect of Neuromuscular Electrical Stimulation on Arthritis Knee

Pain in Older Adults With Osteoarthritis of the Knee
Jean M. Gaines, E. Jeffrey Metter, and Laura A. Talbot

The objective of this study was to examine the short- and long-term effects of a home-based, 12-week
neuromuscular electrical stimulation (NMES) of the quadriceps femoris to decrease arthritis knee pain in
older adults with osteoarthritis of the knee. The study sample (N ⫽ 38) was randomly assigned to the
NMES treatment plus education group or the arthritis education-only group. Pain was measured in both
groups with the McGill Pain Questionnaire (MPQ) at baseline, during the intervention at weeks 4, 8, 12,
and at follow-up and with the Arthritis Impact Measurement Scale 2–Pain Subscale (AIMS2-PS) at
baseline and week 12. The NMES Pain Diary (PD) was completed 15 minutes before and after each
stimulation session. There was a significant 22% decline in pain 15 minutes after as compared with
immediately before each NMES treatment (p ⬍ .001), as measured by the PD. No significant group
differences were found between the 2 groups over the course of the intervention and follow-up. These
findings indicate that a home-based NMES intervention reduced arthritis knee pain 15 minutes after a
NMES treatment.
© 2004 Elsevier Inc. All rights reserved.

O STEOARTHRITIS OF THE knee is a debil-

itating disease that affects 52% of adults
over the age of 75 years (United States Department
response to the stimulation and its overall impact
on pain after completion of the intervention.

of Health and Human Services, 1999). The 2 most

common, nonpharmacological treatment strategies
for osteoarthritis are patient education and exercise
(Flores & Hochberg, 1995). Both education (Lorig,
Gonzalez, Laurent, Morgan, & Laris, 1998) and
exercise (Ettinger et al., 1997) have limited effec-
tiveness in relieving arthritis knee pain. The pri-
mary aim of this study was to compare the effec-
tiveness of a home-based 12-week NMES protocol
plus arthritis education with an arthritis education-
only protocol in improving knee strength and func-
tional performance. These results are reported else-
where (Talbot, Gaines, Ling, & Metter, 2003). A
secondary aim was to examine the immediate pain
Jean M. Gaines, PhD, RN, Erickson Foundation, Baltimore, MD;
E. Jeffrey Metter, MD, National Institutes of Health, National Insti-
tute on Aging, Gerontological Research Center, Baltimore, MD;
Laura A. Talbot, EdD, PhD, RN, Uniformed Services University of
the Health Sciences, Graduate School of Nursing, Bethesda, MD.
Supported in part by the Fund for Geriatric Medicine and
Nursing, Johns Hopkins University.
The views expressed are those of the author and do not reflect
the official policy or position of USUHS, the Department of
Defense, or the United States Government.
Address reprint requests to Jean M. Gaines, PhD, RN, Re-
search Associate in Geriatric Medicine & Gerontology, The
Erickson Foundation, 701 Maiden Choice Lane, Baltimore, MD
21228. E-mail: jgaines@ericksonmail.com
© 2004 Elsevier Inc. All rights reserved.
0897-1897/04/1703-0009$30.00/0
doi:10.1016/j.apnr.2004.06.004

Applied Nursing Research, Vol. 17, No. 3 ( August), 2004: pp 201-206 201
202
NMES is a noninvasive treatment modality that
uses low-frequency, low-amplitude electrical cur-
rent to activate motoneurons resulting in an invol-
untary muscle contraction. NMES may reduce pain
in a manner similar to transcutaneous electrical
nerve stimulation (TENS). Two randomized con-
trolled trials of the effect of high-frequency TENS
capable of inducing a muscle contraction (Jensen,
Zesler, & Christensen, 1991; Zizic et al., 1995)
found a significant reduction in osteoarthritic pain.
Oldham and colleagues (1995) examined the use of
NMES in a randomized pattern in older adults with
osteoarthritis of the knee and found no significant
reduction in pain. Conversely, Moore and Shur-
man, in 1997, studying the use of NMES for
chronic back pain, found a significant decline in
pain immediately following a NMES treatment.
Given the mixed results shown in these studies, the
effect of NMES on arthritis knee pain was exam-
ined.
METHOD
Design
This study was a preliminary randomized con-
trol study using a 2 ⫻ 5 (group ⫻ time) repeated
measures design with measurements taken at base-
line, weeks 4 and 8 (during the intervention), week
12 (postintervention), and week 16 (follow-up)
(Talbot, Gaines, Ling, & Metter, 2003). In addi-
tion, subjects who received NMES were assessed
for their level of pain 15 minutes before and after
each stimulation session. The study was approved
by the Johns Hopkins University Joint Committee
for Clinical Investigations. Written informed con-
sent was obtained from all participants.
Sample
Healthy volunteers (N ⫽ 43) were randomly
assigned to the NMES plus education or education-
only group. Inclusion criteria were 60 years of age
or older with radiographic and clinical evidence of
knee osteoarthritis. Exclusion criteria were the
presence of a cardiac pacemaker; cognitive impair-
ment (a score less than 24 on the Mini-Mental
State Examination); and uncontrolled conditions of
diabetes, hyper- or hypotension, and cardiac dis-
ease. A total of five participants withdrew from the
study leaving a sample size of 38 participants
(NMES group ⫽ 20; education-only group ⫽ 18).
GAINES, METTER, AND TALBOT
Arthritis Self-Management Program
The Arthritis Self-Help Course was used as the
standard of care with all participants attending the
course (Arthritis Foundation, 1996) and was taught
by trained leaders. The Arthritis Self-Help Course
is a standardized 12-hour community-based educa-
tion course designed to provide accurate informa-
tion about arthritis, instill positive attitudes toward
self-management (including pain management),
and assist in developing personalized action plans
(including exercise) for the management of arthri-
tis (Arthritis Foundation, 1996).
NMES Protocol
The NMES intervention was a nurse-managed,
home-based program in which electrical stimula-
tion was used to induce an involuntary contraction
of the quadriceps muscles of the index leg. The
index knee (the leg receiving the NMES therapy)
was the more severe knee as determined before the
start of the protocol and was based on weight-
bearing x-rays and a physical examination includ-
ing the participants’ report of pain, morning stiff-
ness ⱕ30 minutes duration, and crepitus on motion
(Talbot et al., 2003). A portable electrical home
stimulator, the Respond Select (EMPI, St. Paul,
MN), powered by a standard 9-volt battery, was
used to induce the contraction. The following stim-
ulation parameters were programmed into the Re-
spond Select to induce a contraction while reduc-
ing muscle fatigue: rectangular waveform; pulsed,
symmetric, biphasic current; 50 bursts per second;
and a ramp up time of 3 seconds each with an “on”
time of 10 seconds followed by a 50-second “off”
time. High impedance, reusable, self-adhesive
electrodes (Stympac; EMHI, Miami, FL, 4” ⫻ 5”)
were positioned over the vastus medialis oblique
and proximal vastus lateralis of the index leg.
Participants were asked to use the NMES device
for 15 minutes per day 3 days a week on the index
leg for a total of 36 sessions.
During the first 4 weeks of the protocol, the
intensity of the electrical stimulation was set to
induce a muscle contraction that was 10% to 20%
of the isometric maximum voluntary contraction
(MVC). The MVC was obtained at the start of the
protocol using a Kin-Com 125E isokinetic dyna-
mometer (KINCOM; Chattanooga Corporation,
Chattanooga, Tennessee) set at zero velocity and
maintaining a knee joint angle of 120°. All testing
NMES AND ARTHRITIS KNEE PAIN
was performed by trained assessors who were not
blinded to group assignment. Over the 12 weeks of
the protocol, electrical current intensity levels were
incrementally increased to achieve higher percent-
ages of MVC: 20% to 30% of MVC during weeks
5 to 8 and 30% to 40% during weeks 9 to 12.
Before each increase in electrical current intensity,
the participant’s MVC was assessed and the elec-
trical current intensity needed to achieve the de-
sired percentage of MVC was determined. Partic-
ipants received both individual and written
instructions for home use of the Respond Select.
To assure standardization and proper use of the
equipment, participants showed the use of the Re-
spond Select at the monthly testing sessions with
weekly contact at the arthritis self-management
classes.
Outcome Measures
The NMES Pain Diary was used to examine the
immediate effect of the NMES treatment on arthri-
tis knee pain by assessing knee pain 15 minutes
before the treatment and again 15 minutes after
completion of the treatment. The NMES Pain Di-
ary used a numerical scale with anchored end
points of 1 (no pain) to 10 (worst pain). A mean of
29.45 stimulation sessions (SD ⫽ 6.83) were re-
ported by the NMES group participants.
Knee pain was assessed at weeks 0, 4, 8, 12, and
16 using the Present Pain Intensity scale (PPI) and
the Pain Rating Index–Total (PRIT) of the McGill
Pain Questionnaire (Melzack, 1975). The Arthritis
Impact Measurement Scale 2–Pain Subscale
(AIMS2-PS) was assessed at weeks 0 and 12
(Meenan, Mason, Anderson, Guccione, & Kazis,
1992).
Data Analysis
Descriptive statistics were used to describe the
sample. Chi-square, Fisher Exact Test, and inde-
pendent t test were used to test for group differ-
ences. Repeated-measures analysis of variance was
used to evaluate the main effects of time and group
and the time ⫻ group interaction for the McGill
Pain Questionnaire. An analysis of covariance was
used to assess group differences in the AIMS2-PS
with the baseline scores as the covariate. The
NMES Pain Diary pre- and posttreatment scores,
from the same session, were compared by using a
paired t test. All analyses were conducted using
SPSS v 10.1 with level of significance set at 0.05.
203
RESULTS
The sample is described in Table 1. No signifi-
cant differences were found between the NMES
and education-only groups for demographic vari-
ables or any of the pain variables. There was a
significant difference between the groups in
strength in the index leg with the education-only
group having a higher isometric peak torque (p ⫽
.048).
No significant differences were found
between the NMES and education-
only groups for demographic vari-
ables or any of the pain variables.
Within the NMES group, the mean pain score 15
minutes before the NMES session was 3.5,
whereas after the NMES session, the mean pain
score demonstrated a significant 21.77% decline to
2.72 (p ⬍ .001). To determine if there was a
decline in pain after every NMES session, a change
score was calculated (pain after ⫺ pain before) for
each NMES session and is represented as a distri-
bution graph in Figure 1. A decrease in pain was
reported after 74% of the sessions, 17% of sessions
reported no change in pain, and 9% reported an
increase in pain.
Both the NMES group and the education-only
group showed a decline in pain from baseline to
postintervention on the PPI score and the PRIT
(Table 2). However, the interaction effects were
not statistically significant (Table 3). A paired t test
was used to analyze baseline to postintervention
scores for the AIMS-PS. Results showed a nonsig-
nificant 7% increase in pain for the NMES group
(p ⫽ .69) and a significant 66% increase in pain for
the education-only group (p ⫽ .03), although the
analysis of covariance showed no significant group
differences (F [1, 35] ⫽ .028, p ⫽ .869).
DISCUSSION
NMES leads to an immediate reduction in pain
15 minutes after the NMES treatment sessions with
parameters set for muscle strengthening. This im-
mediate reduction in pain is similar to the imme-
diate reduction in chronic back pain following the
204 GAINES, METTER, AND TALBOT

Table 1. Comparison of Demographics and Physical Characteristics Between the NMES and
Education-Only Groups
Variable NMES (n ⫽ 20) Education (n ⫽ 18) p

Gender Women 17 (85%) 13 (72%) .286†

Men 3 (15%) 5 (28%)
Age (years) 70.75 70.94 .913‡
Race White 16 (80.0%) 17 (94.4%) .205†
Nonwhite 4 (20.0%) 1 (5.6%)
Marital Married 8 (40.0%) 8 (44.4%) .782*
Not married 12 (60.0%) 10 (55.6%)
Education ⬍12 yrs 4 (20.0%) 1 (5.9%) .225†
ⱖ12 yrs 16 (80.0%) 16 (94.1%)
Income ⱕ$29,999 10 (55.6%) 11 (73.3%) .245†
⬎$30,000 8 (44.4%) 4 (26.7%)
Grade of OA—index knee Grades 1/2 12 (60.0%) 11 (61.1%) .944*
Grades 3/4 8 (40.0%) 7 (38.9%)
Grade of OA—contra knee Grades 1/2 11 (57.9%) 9 (56.3%) .922*
Grades 3/4 8 (42.1%) 7 (43.8%)
BMI (kg/m2) (SD) 31.52 (5.43) 31.55 (7.25) .986‡
Peak Torque
Index knee (N) (SD) 300.35 (74.38) 360.63 (102.15) .048‡

Abbreviation: NMES, neuromuscular electrical stimulation.

*Chi-square.
†Fisher exact test.
‡Independent t test.

use of NMES reported by Moore and Shurman
(1997) and analogous to the pain relief reported
with the use of TENS (Jensen et al., 1991; Zizic et
al., 1995). However, similar to the results reported
by Oldham et al. (1995), no long-term improve-
ment in pain was found for either the NMES or
education interventions.
The immediate decline in pain may be caused by
the transcutaneous transmission of electrical cur-
rent stimulating the large-diameter afferent nerve
fibers that inhibit second-order neurons in the dor-
sal horn. This prevents nociceptive impulses from
reaching the higher brain centers, specifically the
periaquaductal gray matter and the thalamus
(Sluka, Deacon, Stibal, Strissel, & Terpstra, 1999).

Table 2. Level of Pain by Group and Time

NMES Group Education Group
Variable Time Mean (SD) Mean (SD)

PPI Baseline 1.50 (1.07) 1.41 (1.42)

Post 1.17 (1.04) 1.25 (0.94)
Follow-up 1.08 (0.82) 0.89 (0.54)
PRIT Baseline 19.68 (11.03) 14.00 (10.32)
Post 14.95 (13.07) 10.63 (4.84)
Follow-up 19.38 (13.66) 10.44 (5.25)
AIMS2-PS Baseline 4.85 (2.20) 3.61 (2.26)
Post 5.18 (2.11) 5.99 (2.40)
Figure 1. The NMES Diary Pain Score change for
each NMES session. Participants reported a decline in Abbreviations: NMES, neuromuscular electrical stimulation
pain after 74% of the sessions, no change in pain after group; PPI, present pain intensity; PRIT, Pain Rating Index–
17% of the sessions and an increase in pain after 9% of Total; AIMS2-PS, Arthritis Impact Measurement Scale 2–Pain
the sessions. Subscale.
NMES AND ARTHRITIS KNEE PAIN 205

Table 3. Repeated Measures Analysis of Variance Summary for Outcome Variables

Dependent Source of
Variable Variation SS df MS F p

PPI Time 2.259 2.173 1.040 .996 .380

Time x group 2.182 2.173 1.004 .962 .393
Group .516 1 .516 2.211 .146
PRIT Time 1575.893 3 525.298 7.200 ⬍.001
Time x group 34.868 3 11.623 .159 .923
Group 220.435 1 220.435 6.290 .017

Abbreviations: NMES, neuromuscular electrical stimulation experimental group; PPI, present pain intensity; PRIT, Pain Rating
Index–Total; SS, sum of squares; MS, mean square.

A second mechanism of pain relief may occur
through activation of the endogenous opioid sys-
tem with the release of endorphins (Hoffmann,
Carlsson, & Thoren, 1990; Sluka et al., 1999).
However, a placebo effect cannot be discounted.
Participants in the NMES protocol self-reported
completing 81.82% of stimulation sessions. A hid-
den compliance monitor with the stimulation de-
vice recording a mean of 7.88 hours (SD ⫽ 3.07)
of the 9 hours of possible stimulation for an ad-
herence rate of 87.5%. However, because this was
a home-based protocol, it is possible that partici-
pants adjusted the electrical stimulation to a lower
level thereby reducing the intensity of the contrac-
tion and the pain-relieving effects of the stimula-
tion. Within the NMES protocol, only the partici-
pant’s index leg was stimulated producing pain
relief. Participants frequently stated they had pain
in both knees. The treatment of one leg may not be
enough to overcome the pain from the second leg
and the participant would continue to report little
or no change in pain.
Generalizability in this study is limited to com-
munity-dwelling older adults with osteoarthritis of
the knee functionally able and willing to attend a
local senior center. In this sample, men were un-
derrepresented. Although this study had 13% mi-
nority representation, all were women.
IMPLICATIONS FOR NURSING
These preliminary results suggest that a home-
based protocol of NMES improves arthritis pain 15
minutes after stimulation. Using battery-powered
units, this home-based program is safe, feasible,
and well accepted by older individuals with knee
OA. Through further testing, this method may
prove to be one component in an arsenal of meth-
ods nurses use to relieve arthritis knee pain.
These preliminary results suggest
that a home-based protocol of NMES
improves arthritis pain 15 minutes af-
ter stimulation.
CONCLUSION
Within this study, there was an immediate de-
cline in arthritis knee pain that occurred when
NMES was used only 15 minutes/d, 3 days/wk
with parameters set for muscle strengthening.
NMES, as used in this study, was a low-risk, rel-
atively low-cost strategy that can be taught and
monitored by nurses. This study has provided the
preliminary support for continuing research into
the use of the electrical stimulator among people
with osteoarthritis of the knee.
Acknowledgments
We would like to thank Drs Shari Ling and Rob McKinney
for screening the participants and our research assistant Trina
Duke. We would like to thank the participants in this study.

REFERENCES
Arthritis Foundation. (1996). Arthritis self-help course: lead- M., Monu, J., & Craven, T. (1997). A randomized trial compar-
er’s manual and reference materials. Atlanta, GA: The Arthritis ing aerobic exercise and resistance exercise with a health edu-
Foundation. cation program in older adults with knee osteoarthritis. The
Ettinger, W.H. Jr., Burns, R., Messier, S.P., Applegate, W., Fitness Arthritis and Seniors Trial (FAST). Journal of American
Rejeski, W.J., Morgan, T., Shumaker, S., Berry, M.J., O’Toole, Medical Association, 277(1), 25-31.
206
Flores, R.H., & Hochberg, M.C. (1995). Medical management of
osteoarthritis. Contemporary Internal Medicine, 7(6), 32-3641-32.
Hoffmann, P., Carlsson, S., & Thoren, P. (1990). The effects
of mu, delta- and kappa-opioid receptor antagonists on the pain
threshold increase following muscle stimulation in the rat. Acta
Physiologica Scandinavica, 140, 535-538.
Jensen, H., Zesler, R., & Christensen, T. (1991). Transcuta-
neous electrical nerve stimulation (TNS) for painful osteoar-
throsis of the knee. International Journal of Rehabilitation
Research, 14(4), 356-358.
Lorig, K., Gonzalez, V.M., Laurent, D.D., Morgan, L., &
Laris, B.A. (1998). Arthritis self-management program varia-
tions: Three studies. Arthritis Care and Research, 11(6), 448-
454.
Meenan, R.F., Mason, J.H., Anderson, J.J., Guccione, A.A.,
& Kazis, L.E. (1992). AIMS2. The content and properties of a
revised and expanded Arthritis Impact Measurement Scales
Health Status Questionnaire. Arthritis and Rheumatism, 35(1),
1-10.
Melzack, R. (1975). The McGill Pain Questionnaire: Major
properties and scoring methods. Pain, 1(3), 277-299.
Moore, S.R., & Shurman, J. (1997). Combined neuromuscu-
lar electrical stimulation and transcutaneous electrical nerve
GAINES, METTER, AND TALBOT
stimulation for treatment of chronic back pain: A double-blind,
repeated measures comparison. Archives of Physical Medicine
and Rehabilitation, 78(1), 55-60
Oldham, J.A., Howe, T.E., Petterson, T., Smith, G.P., &
Tallis, R.C. (1995). Electrotherapeutic rehabilitation of the
quadriceps in elderly osteoarthritic patients: A double blind
assessment of patterned neuromuscular stimulation. Clinical
Rehabilitation, 9(1), 10-20.
Sluka, K.A., Deacon, M., Stibal, A., Strissel, S., & Terpstra,
A. (1999). Spinal blockade of opioid receptors prevents the
analgesia produced by TENS in arthritic rats. Journal of Phar-
macology and Experimental Therapeutics, 289(2), 840-846.
Talbot, L.A., Gaines, J.M., Ling, S., & Metter, E.J. (2003). A
home-based protocol of electrical muscle stimulation improves
quadriceps muscle strength in older adults with osteoarthritis of
the knee. Journal of Rheumatology, 30(7), 1571-1578.
United States Department of Health and Human Services.
(1999). Current estimates from the National health Interview
Survey, 1996. Vital and Health Statistics, 10(200), 4, 81.
Zizic, T.M., Hoffman, K.C., Holt, P.A., Hungerford, D.S., &
O’Dell, J.R. et al. (1995). The treatment of osteoarthritis of the
knee with pulsed electrical stimulation. Journal of Rheumatol-
ogy, 22(9), 1757-1761.