Biofeedback Relaxation for Pain Associated

With Continuous Passive Motion in

Taiwanese Patients After Total Knee
Arthroplasty
Tsae-Jyy Wang, Ching-Fen Chang, Meei-Fang Lou, Man-Kuan Ao, Chiung-Chen Liu, Shu-Yuan
Liang, Shu-Fang Vivienne Wu, Heng-Hsing Tung

Correspondence to Chiung-Chen Liu Abstract: Effective pain management is crucial for patient recovery after total
Email: jane20080606@yahoo.com.tw knee arthroplasty (TKA). Biofeedback therapy, which encourages relaxation and
helps alleviate various conditions associated with stress, may help to decrease
Tsae-Jyy Wang
postoperative pain in patients undergoing TKA. A quasi- experimental design was
Professor
used to investigate the efficacy of a biofeedback relaxation intervention in reducing
Department of Nursing
pain associated with postoperative continuous passive motion (CPM) therapy.
National Taipei University of Nursing and
Health Science Sixty-six patients admitted to a general hospital in Taiwan for TKA were recruited
Taipei, Taiwan, ROC and randomly assigned to the intervention or control group. The intervention group
received biofeedback training twice daily for 5 days, concurrent with CPM therapy,
Ching-Fen Chang whereas the control group did not receive the biofeedback intervention. Pain was
Lecturer measured using a numeric rating scale before and after each CPM therapy session
Department of Nursing on postoperative days 1 through 5. The CPM-elicited pain score was calculated by
National Taipei University of Nursing and subtracting the pre-CPM pain score from the post-CPM pain score. Results of
Health Science repeated-measures analysis of variance showed intervention group reported signif-
Taipei, Taiwan, ROC icantly less pain caused by CPM than did the control group (f ¼ 29.70, p < 0.001).
Meei-Fang Lou The study results provide preliminary support for biofeedback relaxation, a non-
Associate Professor invasive and non-pharmacological intervention, as a complementary treatment
Department of Nursing option for pain management in this population. ß 2014 Wiley Periodicals, Inc.
National Taiwan University
Keywords: osteoarthritis; total knee replacement; postoperative pain; biofeed-
Taipei, Taiwan, ROC
back; progressive muscle relaxation training; continuous passive motion
Man-Kuan Ao Research in Nursing & Health, 2015, 38, 39–50
Director of Orthopedics Department Accepted 4 November 2014
Cheng Hsin General Hospital DOI: 10.1002/nur.21633
Taipei, Taiwan, ROC Published online 30 December 2014 in Wiley Online Library (wileyonlinelibrary.com).
Chiung-Chen Liu
Nurse Practitioner
Department of Nursing and Department of
Pediatrics
Tri-Service General Hospital
No.325 Chenggong Rd. Sec. 2, Neihu
District, Taipei 114, Taiwan, ROC

Shu-Yuan Liang
Associate Professor
Department of Nursing
National Taipei University of Nursing and
Health Science
Taipei, Taiwan, ROC
,
Note: Additional authors are listed on the last page.


C 2014 Wiley Periodicals, Inc.
40 RESEARCH IN NURSING & HEALTH
Total knee arthroplasty (TKA) is most commonly performed
for knee osteoarthritis when nonsurgical treatments are no
longer helpful (Ibrahim, Bloch, Esler, Abrams, & Harper,
2010; Otten, van Roermund, & Picavet, 2010; Tien et al.,
2009). The primary goal of TKA is to allow continued
motion of the knee (Zhang et al., 2008); the damaged knee
joint surface is replaced with metal and plastic material to
relieve pain and improve knee function (American Acad-
emy of Orthopaedic Surgeons, 2010; Hohler, 2008). With
proper rehabilitation, a successful outcome is achieved in
the majority of patients (Hohler, 2008; Nilsdotter, Toksvig-
Larsen, & Roos, 2009).
With an overweight and aging population and advan-
ces in arthroplasty surgery, the number of TKAs performed
has been escalating over the years and is projected to con-
tinue increasing in the near future (Otten et al., 2010). The
incidence of TKA nearly doubled from 2000 to 2010 in 15
member countries of the Organization for Economic Co-
operation and Development (OECD; 2013). In 2010, there
were 719,000 TKAs in the United States alone (CDC/
National Center for Health Statistics, 2014). In Taiwan, the
rate of TKA increased from 39.1/100,000 in 2000 to 86.6/
100,000 in 2010, an increase of 122% (National Health
Insurance Administration, 2012; Tien et al., 2009). In 2010,
about 20,000 TKA procedures were done in Taiwan
(National Health Insurance Administration, 2012).
In Taiwan, patients who undergo TKA typically
receive continuous passive motion (CPM) treatment.
Postoperative pain is often intensified during CPM. The
increased pain can hinder patients’ recovery and prevent
them from following the prescribed CPM therapy regimens.
Biofeedback therapy, a biopsychosocial approach to symp-
tom management, may aid conventional pain control and
decrease the need for analgesics in TKA patients receiving
CPM. Biofeedback therapy enhances the patient’s self-
understanding of body signals, applying principles of self-
regulation to alleviate discomfort (Durand, & Barlow, 2009;
Shaffer, & Moss, 2006). Using biofeedback to facilitate
muscle relaxation has been found effective for decreasing
cancer pain (Tsai, Chen, Lai, Lee, & Lin, 2007) and acute
postoperative pelvic pain (Cheema, Lebovits, & Dubois,
2008). However, no evidence was found on the effect of
biofeedback on postoperative pain relief after TKA. There-
fore, the purpose of this study was to test the efficacy of a
biofeedback relaxation intervention in reducing pain
associated with CPM therapy in TKA patients. The hypoth-
esis was that patients who received biofeedback relaxation
intervention would report less pain associated with CPM
therapy than patients who did not receive the intervention.
Pain Management During CPM Therapy
CPM has been used since the early 1980s to facilitate
rehabilitation after TKA (Harvey, Brosseau, & Herbert,
2010). With the patient in a supine position, the foot on the
affected leg is fixed in a track on which the foot slides
Research in Nursing & Health
toward and away from the body, maintaining a constant
passive range of motion (ROM) on the postoperative joint.
CPM may reduce blood and edema (fluid buildup) in
and around the joint, which can lead to joint stiffness
and the development of joint fibrosis and contractures
(Viswanathan & Kidd, 2010). Some researchers found
short-term reduction of postoperative CPM on soft-tissue
swelling and increase in knee ROM in patients with a TKA
(Bennett, Brearley, Hart, Bailey, 2005; Lenssen et al.,
2008), but others reported no beneficial effects of CPM
(Bruun-Olsen, Heiberg, & Mengshoel, 2009; Denis et al.,
2006; Harvey et al., 2010). Due to the inconsistent findings,
CPM is not part of standard postoperative management of
TKA patients in some countries (Congdon, 2012; Rudisile,
2011), but prescribing postoperative CPM therapy is still a
common practice and is covered by universal health insur-
ance in Taiwan.
Significant pain can be associated with postopera-
tive CPM therapy. Effective postoperative pain relief is
crucial for patients’ recovery and has been associated
with a smoother postoperative course and an earlier hos-
pital discharge (Bader et al., 2010; Polomano, Punwoody,
Krenzischek, & Rathmell, 2008). In Taiwan, postoperative
pain after TKA is routinely managed with a combination
of analgesics, including acetaminophen, COX-2 selective
or non-selective nonsteroidal anti-inflammatory drugs,
and opioids. These analgesics are typically administered
on an around-the-clock schedule or with patient-controlled
analgesia pumps. The pain management for patients
undergoing TKA in Taiwan is consistent with the guide-
lines for perioperative acute pain management of the
American Society of Anesthesiologists (Apfelbaum, Ash-
burn, Connis, Gan, & Nickinovich, 2012), with the excep-
tion that local anesthetics are not commonly used during
or after TKA surgery in Taiwan.
Taiwanese Pain Beliefs
Pain is a multifaceted experience shaped by individual cul-
tural context (Narayan, 2010). In Taiwan, there is a strong
cultural value on self-conduct. To be a good patient, an
individual may suppress his/her emotions and expressions
when responding to pain and avoid making demands on
health care providers (Tung & Li, 2014). As a result,
patients may underreport their pain and not want to trouble
the nurses for pain medications (Chen, Miaskowski, Dodd,
& Pantilat, 2008).
Patients influenced by Taiwanese folk beliefs may
consider pain to be a result of their bad conduct in a previ-
ous life. Therefore, they may refuse to take prescribed
pain medications, in order to accept their fates (Chen et al.,
2008). Concern about side effects or addiction of analge-
sics is also prevalent among Taiwanese patients (Lai et al.,
2002; Yin, Tse, & Wong, 2012). They may seek culture-
based therapies or traditional Chinese medicine (e.g.,
herbal medications, acupressure, or acupuncture) instead
 BIOFEEDBACK RELAXATION FOR ACUTE POSTOPERATIVE PAIN/ WANG ET AL.
of taking prescribed medications for pain (Chen, Kung,
Chen, & Hwang, 2006; Chen et al., 2007). From the tradi-
tional Chinese medicine perspective, pain is considered as
a result of disharmony in interaction between functional
entities and the outside world, and the purpose of treat-
ments is to regain the balance between them (Chen et al.,
2008).
Biofeedback as Facilitator of Progressive
Muscle Relaxation for Pain Relief
Biofeedback may be seen as assisting in regaining this bal-
ance. Biofeedback is a monitoring tool with which individu-
als can be coached to regulate bodily processes that
usually occur involuntarily, including heart rate, blood pres-
sure, muscle tension, and skin temperature. By increasing
awareness of one’s own bodily functions and understand-
ing the power of the mind to influence them, an individual
can have more control over his health (Durand & Barlow,
2009; Shaffer & Moss, 2006).
Biofeedback monitoring is performed using several
devices and physiological endpoints, including electromy-
ography, thermal biofeedback, neurofeedback or electroen-
cephalography, electrodermal activity, and heart rate
variability (Yucha & Montgomery, 2008). In a biofeedback
session, electrodes are attached to the patient’s skin.
These electrodes transmit signals to a monitor that displays
sounds, images, or light-flashes representing skin tempera-
ture, breathing rate, heart rate, blood pressure, muscle ten-
sion, or sweating. By observing the display, individuals
train themselves to recognize the sensations associated
with body functions and are guided by a biofeedback thera-
pist to control these functions, using relaxation methods
and mental exercises, including progressive muscle relaxa-
tion, deep breathing, guided imagery, and mindfulness
meditation (Yucha & Montgomery).
Among these, progressive muscle relaxation has
been predominantly used for reducing overall body tension
and anxiety (Craske & Barlow, 2006; Yucha & Montgom-
ery, 2008). Progressive muscle relaxation has both physi-
cal and mental aspects. Physically, it involves tensing and
relaxing muscle groups of the legs, abdomen, chest, arms,
and face. Mentally, it addresses the contrasting emotions
associated with muscle tension and relaxation. During pro-
gressive muscle relaxation, an individual repetitively practi-
ces tensing and relaxing major muscle groups and focuses
on the associated contrasting sensations (National Center
for Complementary and Alternative Medicine, 2013). With
the assistance of a biofeedback machine, one can reduce
the tension carried in the body and feel less stress and
anxiety (Yucha & Montgomery).
Biofeedback encourages relaxation and may alleviate
some conditions associated with stress, such as migraine
headaches, high blood pressure, chronic pain, and urinary
incontinence (Nestoriuc, Martin, Rief, & Andrasik, 2008;
Yucha & Montgomery, 2008). Biofeedback is reported to be
Research in Nursing & Health
41
safe and has no known negative side effects (Yucha &
Montgomery). Using biofeedback to facilitate muscle relax-
ation may help to decrease the pain and anxiety associated
with postoperative CPM in TKA patients, therefore facilitat-
ing the implementation of CPM rehabilitation.
Method
Design
This study used a quasi-experimental design with repeated
measures. A convenience sample of 66 patients undergo-
ing primary total knee replacement were recruited and ran-
domly assigned to the intervention or control groups.
Institutional Review Board approval was obtained from the
Cheng Hsin General Hospital (IRB No. 98A-21–1).
Sample
The desired sample size was estimated using G-Power
(3rd ed) (Faul, Erdfelder, Buchner, & Lang, 2009) for six
repeated measures, with a significance level of 0.05, a
medium effect size (f ¼ 0.25), correlations of 0.5, and a
power of 80%. A sample size of 30 per group was required
to test the efficacy of the biofeedback relaxation interven-
tion for reducing pain associated with CPM therapy in TKA
patients.
Patients with osteoarthritis admitted for TKA were
recruited from orthopedic wards of a 1000-bed general hos-
pital in Taipei, Taiwan. The study was approved by the
research ethics committee of the medical center. Potential
participants were referred by their surgeons and then con-
tacted and screened by the researcher to determine their
eligibility. Patients who met the following inclusion criteria
were solicited: 25 years of age or older, diagnosed with
knee osteoarthritis, admitted to the hospital for a primary
TKA, and able to communicate in Mandarin or Taiwanese
dialect. Patients with cognitive problems and those
experiencing complications from TKA were excluded from
the study.
Seventy-three patients were screened for eligibility;
four did not meet the eligibility criteria and were excluded.
Two of these volunteers had cognitive problems, one spoke
a language other than a Mandarin or Taiwanese dialect,
and one was deaf. The 69 potentially eligible patients were
approached; three refused to participate, and 66 were
recruited and randomized to groups. All 66 participants,
with 33 in each group, completed the study (Fig. 1).
Informed consent was obtained from all participants.
Usual Care
The typical length of stay for TKA in Taiwan is 5–7 days. All
participants were prescribed the standard of care for
the study hospital of two 30-minute daily sessions of CPM
therapy, beginning the first postoperative day until the dis-
charge day. During the therapy, the patient’s postoperative
42 RESEARCH IN NURSING & HEALTH
FIGURE 1. Participant inclusion flow diagram.
limb was placed on the CPM machine with a setting of “mini-
mal” speed, and with the knee ROM set to move between 0°
of extension and 35° of flexion. The ROM was increased 5–
10° each day according to the patient’s tolerance.
Both groups received standard postoperative care
from the hospital. The routine oral and injectable pain med-
ications were acetaminophen, celecoxib, pethidine, or
tramadol.
Biofeedback Intervention
The study intervention consisted of a 30-minute biofeed-
back-assisted progressive muscle relaxation training
session during the CPM sessions twice daily for 5 days. The
intervention group received 30 minutes of individual training
on biofeedback-assisted progressive muscle relaxation skills
on the day before the scheduled TKA surgery. This training
included orientation to biofeedback and instructions for pro-
gressive muscle relaxation skills. Then in each CPM treat-
ment session, the patients practiced progressive muscle
relaxation while observing how the computerized images
changed to indicate successful muscle tension and muscle
relaxation. An interventionist guided the patient through the
biofeedback intervention in each session.
A Wireless Monitoring and Biofeedback Nexus-10
biofeedback machine (Gunjan Human Karigar, India) was
utilized in the progressive muscle relaxation training. The
Nexus-10 is a 10-channel physiological monitoring and
feedback platform with a Bluetooth wireless transmission
Research in Nursing & Health
system. Two surface electrodes were placed over the
quadriceps to record the surface electromyogram and to
monitor muscle tension. Two auxiliary sensors were placed
on the patient’s chest and fingertip to measure respiratory
rate, heart rate, and skin temperature, which have been
associated with relaxation responses (Benson & Klipper,
2000). The readings of these physical parameters were
transmitted to a notebook computer for data analysis and
represented in an audiovisual image displayed on the com-
puter screen to provide feedback to the patient.
The patients began practicing progressive muscle
relaxation by tensing all the muscles in their face, inhaling,
and counting to five. Patients were then instructed to
exhale, relax completely, and feel the tension drip out of
their facial muscles. They repeated the procedure continu-
ing down the body to the following muscle groups: neck
and shoulders, chest, abdomen, arms, hands, buttocks,
legs, and feet. During each intervention session, the inter-
ventionist gave a brief review of these skills and made sure
that every patient accurately executed the biofeedback-
assisted progressive muscle relaxation.
Data Collection and Instruments
The data were collected during 2010. At baseline, each
participant completed a demographics questionnaire. A
research nurse also collected data on disease variables,
including diagnosis, surgical procedures, CPM, and analge-
sic prescriptions, from the patients’ charts.
 BIOFEEDBACK RELAXATION FOR ACUTE POSTOPERATIVE PAIN/ WANG ET AL. 43

Data on pain intensity were collected before and after
each CPM therapy from postoperative days one through
five in both groups. The participants were asked to indicate
the current level of knee pain on an 11-point numerical rat-
ing scale (NRS) on which 0 ¼ no pain, and 10 ¼ worst pos-
sible pain (McCaffery & Beebe, 1993; National Institutes of
Health, 2003). The NRS is commonly used in healthcare
and has been found to be a reliable and valid pain intensity
measure (Hjermstad et al., 2011).
A horizontal format of NRS was used (McCaffery &
Beebe, 1993; National Institutes of Health, 2003) using
Chinese word descriptors and end points. The NRS was trans-
lated and back-translated until the meaning between the
English and Chinese was in concordance. The word transla-
tions in Chinese were validated by five Taiwanese health pro-
fessionals, and the content validity index was 0.933. Li, Liu,
and Herr (2007) used this NRS to assess pain intensity in 173
Chinese postoperative adults and found high intraclass corre-
lation coefficients (0.673–0.822) across current, worst, least,
and average pain on 7 postoperative days (p < 0.001). The
NRS was found to have concurrent validity with a visual analog
scale, verbal descriptor scale, and faces pain scale-revised as
evidenced by significant Spearman correlations among the
four scales, ranged from 0.74 to 0.95 for ratings of current
pain, 0.80 to 0.99 for worst pain, 0.71 to 0.97 for least pain, and
0.72 to 0.95 for average pain on 7 postoperative days. Zhou,
Petpichetchian, and Kitrungrote (2011) reported a test–retest
reliability (N ¼ 200) of .80–.94 in different age groups assessed
over a 3-day interval. In the current study, the test–retest reli-
ability of the NRS was 0.69 (p < 0.01), assessed at an 8-hour
interval before sessions of CPM therapy on the morning and
afternoon of the first postoperative day.
Data Analysis
The data were analyzed using the Statistical Package for
Social Sciences 18.0 (SPSS, Inc., Chicago, IL). Descriptive
statistics were used to describe participants’ demographics,

Table 1. Participant Demographics and Disease Characteristics by Group (N ¼ 66)

Intervention (n ¼ 33) Control (n ¼ 33)

Characteristic Mean SD Mean SD ta p

Age 73.5 9.5 71.7 6.5 0.9 0.378

Years since diagnosis with knee OA 3.3 3.6 4.3 4.8 488.5 0.465
BMI (Kg/m2) 27.4 4.5 27.8 4.6 519.0 0.744

n % n % x2b

Gender
Male 12 36.4 11 33.3 0.07 0.796
Female 21 63.6 22 66.7
Education
Unable to read or write 6 18.2 11 33.3 5.47 0.242
Elementary school 12 36.4 15 45.5
Middle school 5 15.2 3 9.1
High school 5 15.2 3 9.1
College or University 5 15.2 1 3.0
Marital status
Married 33 100.0 32 97.0 1.02 1.000
Single 0 0.0 1 3.0
Employment status
Unemployed 32 97.0 32 97.0 0.00 1.000
Employed 1 3.0 1 3.0
Duration of CPM therapy
3 days 2 6.1 0 0.0 2.48 0.289
4 days 2 6.1 1 3.0
5 days 29 87.9 32 97.0
Length of CPM sessions
30 minutes 32 97.0 30 90.9 1.07 0.307
Ever less than 30 minutes 1 3.0 3 9.1

Note: OA, osteoarthritis; BMI, body mass index; CPM, continuous passive motion; t (or x2), value of independent t-tests (or Chi-square
tests); SD, standard deviation.
a
Mann–Whiney U-test.
b
Fisher’s exact test.

Research in Nursing & Health

44 RESEARCH IN NURSING & HEALTH

Table 2. Intravenous Patient-Controlled Analgesia and Standing Analgesics Used by Group (N ¼ 66)

Intervention (n ¼ 33) Control (n ¼ 33)

Day Analgesic n % n % x2 p

Surgery PCA with pethidine Yes 15 45.5 16 48.5 0.06 0.81

No 18 54.5 17 51.5
Standing analgesics Acetaminophen or COX-2 26 78.8 19 57.6 3.42 0.11
þ pethidine or tramadol 7 21.2 14 42.4
Postop 1 PCA with pethidine Yes 15 45.5 15 45.5 0.00 1.00
No 18 54.5 18 54.5
Standing analgesics Acetaminophen or COX-2 26 78.8 19 57.6 3.42 0.11
þ pethidine or tramadol 7 21.2 14 42.4
Postop 2 PCA with pethidine Yes 14 42.4 15 45.5 0.06 0.80
No 19 57.6 18 54.5
Standing analgesics Acetaminophen or COX-2 26 78.8 21 63.6 1.84 0.28
þ pethidine or tramadol 7 21.2 12 36.4
Postop 3 PCA with pethidine Yes 13 39.4 13 39.4 0.00 1.00
No 20 60.6 20 60.6
Standing analgesics Acetaminophen or COX-2 26 78.8 21 63.6 1.85 0.28
þ pethidine or tramadol 7 21.2 12 36.4
Postop 4 PCA with pethidine Yes 1 3.1 0 0.0 1.05 0.49a
No 31 96.9 33 100.0
Standing analgesics Acetaminophen or COX-2 26 81.3 21 63.6 2.52 0.17
þ pethidine or tramadol 6 18.8 12 36.4
Postop 5 PCA with pethidine Yes 1 3.2 0 0.0 1.05 0.49a
No 30 96.8 32 100.0
Standing analgesics Acetaminophen or COX-2 24 77.4 21 63.6 1.67 0.27
þ pethidine or tramadol 7 22.6 12 36.4

Note: x2, Chi-square; PCA, intravenous patient-controlled analgesia; COX-2, COX-2 inhibitor; þ, acetaminophen or COX-2 in addition
to pethidine intravenous injection or tramadol oral.
a
Fisher’s exact test.

Table 3. Frequency of As-Needed Analgesics Used by Group after Total Knee Arthroplasty

Intervention (n ¼ 33) Control (n ¼ 33)

Day Frequency n % n % x2a p

Day of surgery 0 23 69.7 26 78.8 0.88 0.65

1 8 24.2 5 15.2
2 2 6.1 2 6.1
Postop day 1 0 30 90.9 30 90.9 <0.01 1.00
1 2 6.1 2 6.1
2 1 3.0 1 3.0
Postop day 2 0 31 93.9 33 100 2.06 0.36
1 1 3.0 0 0.0
2 1 3.0 0 0.0
Postop day 3 0 33 100.0 30 90.9 3.14 0.24
1 0 0.0 3 9.1
2 0 0.0 0 0.0
Postop day 4 0 30 96.8 31 96.9 2.00 0.37
1 1 3.2 0 0.0
2 0 0.0 1 3.1
Postop day 5 0 26 92.9 28 90.3 0.12 1.00
1 2 7.1 3 9.7
2 0 0.0 0 0.0

Note: x2, Chi-square. The “as-needed” analgesic prescribed was either pethidine injection or tramadol oral.
a
Fisher’s exact test.

Research in Nursing & Health

 BIOFEEDBACK RELAXATION FOR ACUTE POSTOPERATIVE PAIN/ WANG ET AL. 45

disease characteristics, and pain intensity. Chi-square tests
or Fisher’s exact tests and independent samples t-tests
were used to examine group differences in baseline values.
Repeated-measures analysis of variance (RM-ANOVA)
was used to compare between-group differences in changes
in CPM-associated pain intensity over time. The CPM-elicited
pain score was calculated by subtracting the pre-CPM pain
score from the post-CPM pain score. Age, gender, education
level, and body mass index (BMI) were added to the model as
covariates.
Results
Participants’ Characteristics and Baseline
Equivalence
The characteristics of the intervention and control groups
are presented in Table 1. There were no differences
between groups on baseline variables. In the sample as a
whole, there were 43 females and 23 males. The mean
age was 72.6 years (SD ¼ 8.2; range: 41–85). The majority
were married (n ¼ 65), unemployed (n ¼ 64), and had com-
pleted elementary school education (n ¼ 27). The duration
since knee OA diagnosis ranged from 3 months to 20 years,
with an average of 3.8 years (SD ¼ 4.2). The average BMI
was 27.6 kg/m2 (SD ¼ 4.5; range 19.1–39.4).
All 66 participants were prescribed 30 minutes of
CPM twice a day for 5 days. Among them, 60 completed
all 5 days of CPM, three (two in the intervention group and
one in the control group) completed 4 days, and three (two
in the intervention group and one in the control group) com-
pleted 3 days, all due to discharge from the hospital prior
to the fifth postoperative day.
Four of the 66 participants discontinued the CPM
before the prescribed full 30-minute session. One in the inter-
vention group and three in the control group had to pause or
terminate the 30-minute session due to pain. All of these
patients were able to continue on CPM again as soon as their

Table 4. Pain Intensity before and After Continuous Passive Motion (CPM) Therapy and CPM-Elicited Pain Intensity in
Patients After Total Knee Arthroplasty (N ¼ 66)

Intervention Control

Day CPM Time Pain Intensity Mean SD Mean SD

One Morning Pre-CPM 5.39 1.12 5.30 1.42

Post-CPM 5.90 1.88 7.33 1.90
CPM-elicited 0.52 1.58 2.03 1.55
Afternoon Pre-CPM 4.93 1.82 5.48 1.87
Post-CPM 5.54 1.97 7.15 1.60
CPM-elicited 0.61 1.12 1.67 1.29
Two Morning Pre-CPM 5.21 1.52 4.72 1.48
Post-CPM 5.06 1.77 5.27 2.03
CPM-elicited 0.00 1.39 0.55 1.00
Afternoon Pre-CPM 5.06 1.43 4.67 1.47
Post-CPM 5.06 1.73 5.39 2.09
CPM-elicited 0.00 1.30 0.73 1.23
Three Morning Pre-CPM 4.58 1.56 4.21 1.74
Post-CPM 4.21 2.04 4.70 1.76
CPM-elicited 0.36 1.39 0.48 0.91
Afternoon Pre-CPM 4.36 1.98 4.12 1.92
Post-CPM 4.03 2.13 4.73 1.77
CPM-elicited 0.33 1.02 0.61 0.90
Four Morning Pre-CPM 4.15 1.75 4.09 1.84
Post-CPM 3.60 1.92 4.70 2.02
CPM-elicited 0.55 1.03 0.61 1.00
Afternoon Pre-CPM 4.32 1.64 3.88 1.82
Post-CPM 3.71 1.79 4.73 1.93
CPM-elicited 0.61 1.02 0.95 1.25
Five Morning Pre-CPM 3.87 1.59 3.47 1.90
Post-CPM 3.58 1.75 4.16 1.63
CPM-elicited 0.29 1.01 0.69 1.03
Afternoon Pre-CPM 3.86 1.35 3.55 1.91
Post-CPM 3.36 1.47 4.23 1.67
CPM-elicited 0.50 1.07 0.68 1.01

Note: F ¼ 29.70, p < .001, for group effect on CPM-elicited pain intensity over time in repeated measures analysis of variance with one
between-group factor. CPM-elicited pain intensity ¼ post-CPM pain score minus pre-CPM pain score. A negative value indicates score
was lower after CPM. SD, standard deviation.

Research in Nursing & Health

46 RESEARCH IN NURSING & HEALTH
pain subsided. The time range during which patients stopped
or paused a session was 10–15 minutes into the session.
There was no between-group difference in the duration and
length of CPM carried out by the participants (Table 1).
Most participants were prescribed acetaminophen
1 1 1
(Paramol ), celecoxib (Celebrex ), pethidine (Demerol ), or
1
tramadol (Tramal ) for postoperative pain. The type and
frequency of use of analgesics over the 6 days (day of sur-
gery through postoperative day 5) were no different in the
two groups (Tables 2 and 3).
Intervention Adherence
Among the 33 participants randomized to the intervention
group, 29 completed the 5 days of biofeedback-assisted pro-
gressive muscle relaxation training. Of the four that did not
complete the training, two participated in the intervention for
4 days, and the other two participated for 3 days; participa-
tion was discontinued because the patients were discharged
from the hospital prior to the fifth postoperative day.
Effect of Biofeedback on Postoperative Pain
During CPM Therapy
Pain intensity measured before CPM therapy was no dif-
ferent in the two groups on any of the 5 study days
FIGURE 2. Changes in pain intensity measured before and after continuous passive motion therapy
over time. The data are shown as the mean and 95% confidence interval (error bars). Note. CPM,
continuous passive motion.
Research in Nursing & Health
(Table 4). The levels of postoperative pain associated
with CPM therapy gradually decreased over time in both
groups (Fig. 2). However, pain intensity after CPM ther-
apy was higher in the control group than in the interven-
tion group on postoperative days one, four, and five
(Table 4), and the pain intensity was intensified by CPM
therapy (CPM-elicited pain) in the control group but not in
the intervention group on the final four study days
(Fig. 2).
Results of RM-ANOVA showed a significant
group effect on CPM-elicited pain (f ¼ 24.17, p < 0.001).
No covariates were statistically significant, and the
reduced model without the covariates also showed a
significant group effect (f ¼ 29.70, p < 0.001), supporting
the study hypothesis that patients who received the bio-
feedback relaxation intervention would report less pain
than would patients in the control group following CPM
therapy.
There was a significant effect of time on CPM-elicited
pain (f ¼ 22.36, p < 0.001, based on Greenhouse–Geisser
correction); tests of within-subjects contrasts showed a sig-
nificant linear term (f ¼ 48.94, p < 0.001). Both groups
exhibited a gradual decrease in the difference between
pre-treatment and post-treatment pain scores over time
(Fig. 3), but the intervention group reported substantially
 BIOFEEDBACK RELAXATION FOR ACUTE POSTOPERATIVE PAIN/ WANG ET AL. 47

FIGURE 3. Changes in CPM-elicited pain intensity scores over time. The data are shown as the
mean and 95% confidence intervals (error bars). Note. CPM, continuous passive motion.

less CPM-elicited pain than did the control group did on all
5 study days.
Discussion
Patients who underwent TKA experienced significant post-
operative wound pain, which gradually decreased over time.
The pain intensity was significantly intensified by the CPM
therapy in the control group, but not in the intervention
group for the final 4 study days. The beneficial effect of the
intervention was observed on the first postoperative day and
continued thereafter. The between-group differences were
consistent across study days. These findings support the
efficacy of biofeedback relaxation intervention for alleviating
postoperative pain associated with CPM therapy in TKA
patients.
Some patients from a traditional Chinese medicine
perspective may believe that western pain medication is
too strong and maybe reluctant to take the full dose of the
prescribed pain medications or request an as needed
based pain medication. Biofeedback can be an alternative
choice for pain management in this group of patients,
because it is a natural fit within the Taiwanese culture with
an acceptance of the unity of mind-body-spirit (Lin, Peper,
& Weng, 2007).
Patients were able to learn biofeedback-machine
assisted progressive muscle relaxation skills with a 30-min-
ute preoperational individual training session and a brief
review prior to each biofeedback relaxation session. The
intervention was received favorably among the participants
in the intervention group, and they were willing and able to
practice the progressive muscle relaxation skills during
each CPM session. Participants in both groups adhered to
the prescribed CPM therapy. All participants completed the
twice-daily CPM therapy for 5 days, except those who were
discharged from the hospital before the final study day.
Although the patients in the control group experienced
worse pain during the CPM therapy than those in the inter-
vention group, there was no between-group difference in
CPM adherence. This finding suggests that wound pain is
not the only factor affecting participants’ adherence to CPM
therapy. Other factors such as motivation and perceived
CPM effectiveness may contribute to CPM adherence.
The study had several limitations. First, the partici-
pants were recruited from orthopedic wards of a medical

Research in Nursing & Health

48 RESEARCH IN NURSING & HEALTH
center and may vary from those admitted to other clinical
settings. Therefore, these results may not be generalizable
beyond this sample population. Second, only the short-
term effects of a 5-day biofeedback relaxation on pain
associated with CPM therapy were tested in this study; the
long-term outcomes and potential side effects remain to be
determined. Third, due to ethical considerations, a placebo
control group was not included; therefore, the potential of a
placebo effect cannot be excluded, and the intervention
effect may be overestimated. Last, issues of individual
difference in participants were not taken into account in
this study design. Previous studies showed that individuals
may respond to biofeedback-assisted relaxation training
differently due to differences in individuals’ cognitive capac-
ity for absorption and imaginative involvement on the
level of the relaxation achieved during biofeedback
(Menzies, Taylor, & Bourguignon, 2008). Additional
research is needed to explore individual differences in
responding to the study intervention.
This is the first study to investigate the effects of bio-
feedback relaxation intervention for reducing pain associ-
ated with CPM therapy in TKA patients. Other studies have
examined the effects of CPM, but none included biofeed-
back as a modality for pain management. These results
provide preliminary support for the effects of biofeedback
relaxation intervention for postoperative pain relief. How-
ever, additional randomized, controlled trials with longitudi-
nal designs, larger sample sizes, and blinded outcome
assessments are required to better define the role of bio-
feedback relaxation intervention in the treatment protocol
for postoperative pain control in TKA patients. Future work
with an appropriate design to determine who is most likely
to benefit from this intervention is also an important next
step.
Conclusion
The results of the current study showed that biofeedback
relaxation intervention was safe and effective for controlling
postoperative pain during the CPM therapy. Patients who
practiced biofeedback-assisted progressive muscle relaxa-
tion showed less pain elicited by CPM therapy compared to
those in the control group. No adverse reactions to the
intervention were observed. Although more studies are
required to define the role of biofeedback relaxation inter-
vention in managing postoperative pain, it is a non-invasive
and inexpensive intervention that can be considered as a
complementary treatment option for postoperative pain
control during CPM therapy.
References
American Academy of Orthopaedic Surgeons. (2010). Knee
replacement implants. Retrieved from http://orthoinfo.aaos.org/
topic.cfm?topic=A00221.
Research in Nursing & Health
Apfelbaum, J. L., Ashburn, M. A., Connis, R. T., Gan, T. J., &
Nickinovich, D. G. (2012). Practice guidelines for acute pain man-
agement in the perioperative setting: An updated report by the
American Society of Anesthesiologists Task Force on Acute Pain
Management. Anesthesiology, 116, 248–273.
Bader, P., Echtle, D., Fonteyne, V., Livadas, K., De Meerleer, G.,
Paez Borda, A., … Vranken, J. H. (2010). Guidelines on pain
management. Arnhem, Netherlands: European Association of
Urology.
Bennett, L. A., Brearley, S. C., Hart, J. A. L., & Bailey, M. J. (2004). A
comparison of 2 continuous passive motion protocols after total
knee arthroplasty. Journal of Arthroplasty, 20, 225–233. doi:
10.1016/j.arth.2004.08.009.
Benson, H., & Klipper, M. Z. (2000). Relaxation responses. New
York, NY: HarperCollins.
Bruun-Olsen, V., Heiberg, K. E., & Mengshoel, A. M., (2009). Contin-
uous passive motion as an adjunct to active exercises in early
rehabilitation following total knee arthroplasty—a randomised
controlled trial. Disability and Rehabilitation, 31, 277–283. doi:
10.1080/09638280801931204.
CDC/National Center for Health Statistics. (2010). National Hospital
Discharge Survey, 2010. Retrieved from http://www.cdc.gov/
nchs/fastats/insurg.htm.
Cheema, S., Lebovits A., & Dubois, A. (2008). Is there a role for
biofeedback in the management of acute postoperative pain?
The Journal of Pain, 9, 54. doi: 10.1016/j.jpain.2008.01.237.
Chen, F. P., Chen, T. J., Kung, Y. K., Chen, Y. C., Chou, L.F., Chen,
F. J., & Hwang, S. J., (2007). Use frequency of traditional Chinese
medicine in Taiwan. BMC Health Service Research, 7, 26. doi:
10.1186/1472–6963-7–26.
Chen, F. P., Kung, Y. Y., Chen, T. J., & Hwang, S. J. (2006). Demo-
graphics and patterns of acupuncture use in the Chinese population:
The Taiwan experience. Journal of Alternative and Complementary
Medicine, 12, 379–387. doi: 10.1089/acm.2006.12.379.
Chen, L. M., Miaskowski, C., Dodd, M., & Pantilat, S. (2008). Con-
cepts within the Chinese culture that influence the cancer pain
experience. Cancer Nursing, 31, 103–108. doi: 10.1097/01.
NCC.0000305702.07035.4d.
Chen, Y. W., & Wang, H. H. (2014). The effectiveness of acupres-
sure on relieving pain: A systematic review. Pain Management
Nursing, 15, 539–590. doi: 10.1016/j.pmn.2012.12.005.
Congdon W. (2012). Total knee arthroplasty protocol. The Brigham
and Women’s Hospital, Inc., Department of Rehabilitation Ser-
vices, Boston MA. Retrieved from http://www.brighamand-
womens.org/patients_visitors/pcs/rehabilitationservices/physical
%20therapy%20standards%20of%20care%20and%20protocols/
knee%20-%20TKA%20protocol.pdf.
Craske, M. G., & Barlow, D. H. (2006). Mastery of your anxiety and
worry. Oxford, UK: Oxford University Press, Inc..
Denis, M., Moffet, H., Caron, F., Ouellet, D., Paquet, J., & Nolet, L.
(2006). Effectiveness of continuous passive motion and conven-
tional physical therapy after total knee arthroplasty: A random-
ized clinical trial. Physical Therapy, 86, 174–185.
Durand, V. M., & Barlow, D. (2009). Abnormal psychology: An
integrative approach. Belmont, CA: Wadsworth Cengage
Learning.
Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical
power analyses using G*Power 3.1: Tests for correlation and
 BIOFEEDBACK RELAXATION FOR ACUTE POSTOPERATIVE PAIN/ WANG ET AL.
regression analyses. Behavior Research Methods, 41, 1149–
1160. doi: 10.3758/BRM.41.4.1149.
Harvey, L. A., Brosseau, L., & Herbert, R. D. (2010). Continuous pas-
sive motion following total knee arthroplasty in people with arthri-
tis. Cochrane Database Systemic Review, 17, CD004260. doi:
10.1002/14651858.CD004260.pub2.
Hjermstad, M. J., Fayers, P. M., Haugen, D. F., Caraceni, A., Hanks,
G. W., Loge, J. H., … European Palliative Care Research Collab-
orative (EPCRC). (2011). Studies comparing Numerical Rating
Scales, Verbal Rating Scales, and Visual Analogue Scales for
assessment of pain intensity in adults: A systematic literature
review Journal of Pain and Symptom Management, 41, 1073–
1093. doi: 10.1016/j.jpainsymman.2010.08.016.
Hohler S. E. (2008). Total knee arthroplasty: Past successes and
current improvements AORN Journal, 87, 143–158. doi: 10.1016/
j.aorn.2007.08.016.
Ibrahim, T., Bloch, B., Esler, C. N., Abrams, K. R., & Harper, W. M.
(2010). Temporal trends in primary total hip and knee arthroplasty
surgery: results from a UK regional joint register, 1991–2004.
Annals of the Royal College of Surgeons of England, 92, 231–
235. doi: 10.1308/003588410X12628812458572.
Kurtz, S., Mowat, F., Ong, K., Chan, N., Lau, E., & Halpern, M.
(2005).Prevalence of primary and revision total hip and knee
arthroplasty in the United States from 1990 through 2002. The
Journal of Bone and Joint Surgery, 87, 1487–1497. doi: 10.2106/
JBJS.D.02441.
Lai, Y. H., Keefe, F. J., Sun, W. Z., Tsai, L. Y., Cheng, P. L., Chiou,
J. F., & Wei, L. L. (2002). Relationship between pain-specific
beliefs and adherence to analgesic regimens in Taiwanese
cancer Patients: a preliminary study. Journal of Pain and Symp-
tom Management, 24, 415–423. doi: 10.1016/S0885-3924(02)
00509-2.
Lin, I. -M., Peper, E., & Weng, C. -Y. (2007). Current and future sta-
tus of biofeedback in Taiwan. Biofeedback, 35(4), 122–125.
Lenssen, T. A., van Steyn, M. J., Crijns, Y. H., Waltje, E. M., Roox,
G. M., Geesink, R. J., … De Bie, R. A. (2008). Effectiveness of
prolonged use of continuous passive motion (CPM), as an
adjunct to physiotherapy, after total knee arthroplasty. BMC Mus-
culoskeletal Disorders, 9, 60. doi: 10.1186/1471-2474-9-60.
Li, L., Liu, X. Q., & Herr, K. (2007). Postoperative pain intensity
assessment: a comparison of four scales in Chinese adults. Pain
Medicine, 8, 223–234. doi: 10.1111/j.1526–4637.2007.00296.x.
McCaffery, M., & Beebe, A. (1993). Pain: Clinical manual for nursing
practice. Baltimore, MD: V.V. Mosby Company.
Menzies, V., Taylor, A. G., & Bourguignon, C. (2008). Absorption: An
individual difference to consider in mind–body interventions.
Journal of Holistic Nursing, 26, 297–302. doi: 10.1177/
0898010107307456.
Narayan, M. C. (2010). Culture’s effects on pain assessment and
management. American Journal of Nursing, 110(4), 38–47. doi:
10.1097/01.NAJ. 0000370157.33223.6d.
National Center for Complementary and Alternative Medicine.
(2013). Relaxation techniques for health: An introduction
Bethesda, MD: National Institutes of Health. Retrieved from http://
nccam.nih.gov/health/stress/relaxation.htm. .
National Health Insurance Administration, Ministry of Health and
Welfare, R.O.C. (2012). NHI paid for artificial knee joints, letting
your “knee” no longer in pain. National Health Insurance
Research in Nursing & Health
49
Administration Press Releases, 237. Retrieved from: http://www.
nhi.gov.tw/epaperN/ItemDetail.aspx?DataID=3111&IsWebData=
0&ItemTypeID=5&PapersID=267&PicID=.
National Institutes of Health—Warren Grant Magnuson Clinical Cen-
ter. (2003). Pain intensity instruments. Retrieved from http://www.
webcitation.org/6Ag75MDIq.
Nestoriuc, Y., Martin, A., Rief, W., & Andrasik, F. (2008). Biofeed-
back treatment for headache disorders: A comprehensive effi-
cacy review. Applied Psychophysiology and Biofeedback, 33,
125–140. doi: 10.1007/s10484–008-9060–3 .
Nilsdotter, A. K., Toksvig-Larsen, S., & Roos, E. M. (2009). A 5 year
prospective study of patient-relevant outcomes after total knee
replacement. Osteoarthritis Cartilage, 17, 601–606. doi: 10.1016/
j.joca.2008.11.007.
OECD. (2013). Hip and knee replacement. Health at a Glance 2013:
OECD Indicators (pp. 96–97). OECD Publishing. Retrieved from
http://dx.doi.org/10.1787/health_glance-2013-en. .
Otten, R., van Roermund, P. M., & Picavet, H. S. (2010). Trends in
the number of knee and hip arthroplasties: Considerably
more knee and hip prostheses due to osteoarthritis in 2030.
Nederlands Tijdschrift voor Geneeskunde, 154, A1534.
Polomano, R. C., Punwoody, C. J., Krenzischek, D. A., & Rathmell,
J. P. (2008) Perspective on pain management in the 21st century.
Journal of Perianesthesia Nursing, 23, 4–14. doi: 10.1016/j.
jopan.2007.11.004.
Rudisile, C. (2011). The effect of continuous passive motion on knee
flexion, pain, and function after total knee arthroplasty. PT Criti-
cally Appraised Topics. Paper 25. Retrieved from http://com-
mons.pacificu.edu/ptcats/25.
Shaffer, F., & Moss, D. (2006). Biofeedback. In C.-S. Yuan, E. J.
Bieber, & B. A. Bauer (Eds.), Textbook of complementary and
alternative medicine (2nd ed.), (pp. 291–312). Abingdon, Oxford-
shire, UK: Informa Healthcare.
Tien, W-C., Kao, H-Y., Tu, Y-K., Chiu, H-C., Lee, K-T., & Shi, H-Y.
(2009). A population-based study of prevalence and hospital
charges in total hip and knee replacement. International Ortho-
paedics, 33, 949–954. doi: 10.1007/s00264–008-0612–1 .
Tsai, P. S., Chen, P. L., Lai, Y. L., Lee, M. B., & Lin, C. C. (2007).
Effects of electromyography biofeedback-assisted relaxation on
pain in patients with advanced cancer in a palliative care unit.
Cancer Nursing, 30, 347–353. doi: 10.1097/01.NCC.
0000290805.38335.7b.
Tung, W-C., & Li, Z. (2014). Pain beliefs and behaviors among
Chinese. Home Health Care Management & Practice. Advance
online publication. doi: 10.1177/1084822314547962.
Viswanathan, V., & Kidd, M. (2010) Effect of continuous passive
motion following total knee arthroplasty on knee range of motion
and function: A systematic review. NZ Journal of Physiotherapy,
38, 14–22.
Yin, H. H., Tse, M. Y., & Wong, F. K. Y. (2012). Postoperative pain
experience and barriers to pain management in Chinese adult
patients undergoing thoracic surgery. Journal of Clinical Nursing,
21, 1232–1243. doi: 10.1111/j.1365–2702.2011.03886.x.
Yucha, C., & Montgomery, D. (2008). Evidence-based practice in
biofeedback and neurofeedback. Wheat Ridge, CO: Association
for Applied Psychophysiology and Biofeedback.
50 RESEARCH IN NURSING & HEALTH

Zhang, W., Moskowitz, R. W., Nuki, G., Abramson, S., Altman R. D.,
Arden, N., … Tugwell, P. (2008). OARSI recommendations for the
management of hip and knee osteoarthritis, Part II: OARSI
evidence-based, expert consensus guidelines. Osteoarthritis
Cartilage, 16, 137–162. doi: 10.1016/j.joca.2007.12.013.
Zhou, Y., Petpichetchian, W., & Kitrungrote, L. (2011). Psychometric
properties of pain intensity scales comparing among postopera-
tive adult patients, elderly patients without and with mild cognitive
impairment in China. International Journal of Nursing Studies, 48,
449–457. doi: 10.1016/j.ijnurstu.2010.08.002.

Authors continued from first page:

Shu-Fang Vivienne Wu Heng-Hsing Tung

Associate Professor Professor
Department of Nursing Department of Nursing
National Taipei University of Nursing and Health Science National Taipei University of Nursing and Health Science
Taipei, Taiwan, ROC Taipei, Taiwan, ROC

Acknowledgements
The authors thank all the participants.

Research in Nursing & Health