540920

research-article2014
CRE0010.1177/0269215514540920Clinical RehabilitationJorge et al.

CLINICAL
Article REHABILITATION

Clinical Rehabilitation

Progressive resistance exercise in 2015, Vol. 29(3) 234­–243

© The Author(s) 2014
Reprints and permissions:
women with osteoarthritis of the sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0269215514540920

knee: a randomized controlled trial cre.sagepub.com

Renata Trajano Borges Jorge1, Marcelo Cardoso de Souza1,

Aline Chiari1, Anamaria Jones1, Artur da Rocha Correa Fernandes2,
Império Lombardi Júnior3 and Jamil Natour1

Abstract
Objective: To determine the effect of a progressive resistance exercise (PRE) program on women with
osteoarthritis (OA) of the knee.
Methods: Eligible subjects included women aged 40 to 70 years with pain between 3 and 8 on a 10-
cm pain scale. Among the 144 subjects screened, 60 met the eligibility criteria and were randomized
to the experimental group (EG) or control group (CG). Subjects in the EG participated in a 12-week
PRE program twice a week and CG remained on a waiting list for physical therapy. The PRE program
consisted of strengthening exercises for knee extensors, knee flexors, hip abductors and hip adductors,
all performed with 50% and 70% of the one-repetition maximum (1RM) using machines with free weights.
Resistance was reevaluated every two weeks. Assessments of pain, muscle strength, walking distance,
function and quality of life were performed at baseline, six weeks and 12 weeks by a blinded assessor.
Results: Twenty-nine female subjects were randomly assigned to the EG and 31 were randomly assigned
to the CG. Repeated-measures ANOVA revealed significantly better results in the EG pain (from 7.0±1.3
to 4.3±3.1 in the EG and from 7.0±1.2 to 6.6±1.5 in the CG - p<0.001), function (p<0.001), some domains
of quality of life (physical function: p=0.002; physical role limitation: p=0.002; and pain: p=0.044) and
muscle strength (extensors: p<0.001; flexors: p=0.002; and abductors: p<0.001).
Conclusion: The PRE program was effective in reducing pain and improving function, some quality of life
domains and strength in women with OA of the knee.

Keywords
Osteoarthritis of the knee, progressive resistance exercise, women, pain, function, quality of life

Received: 11 November 2013; accepted: 13 May 2014

1Universidade Federal de São Paulo, Rheumatology Division, Corresponding author:

São Paulo, SP, Brazil Jamil Natour, Division of Rheumatology, UNIFESP, Rua
2Universidade Federal de São Paulo, Department of Diagnostic Botucatu, 740, CEP: 04023-900, São Paulo, SP, Brazil.
Imaging, São Paulo, SP, Brazil Email: [email protected]
3Universidade Federal de São Paulo, Department of Human

Movement Sciences, São Paulo, SP, Brazil

Jorge et al. 235

Introduction exercise and include the hip muscles in exercise

programs, the aim of the present study was to
Osteoarthritis is considered the most common determine the effect of a progressive resistance
musculoskeletal disease, affecting 20% of the exercise program in pain of women with osteoar-
global population.1–2 Interestingly, women are thritis of the knee to facilitate the reproduction of
more affected and burdened by osteoarthritis of the the method in future studies.
knee than men.3 The vast majority of individuals
with osteoarthritis of the knee (80%) experience
pain and limited mobility, and 25% cannot perform
Methods
major activities of daily living. These figures dem- A single-blind, randomized, controlled trial was
onstrate the negative effect of osteoarthritis on carried out, women with unilateral or bilateral oste-
quality of life.4 oarthritis of the knee, based on the classification
Weakness of the quadriceps muscle is a common criteria of the American College of Rheumatology
finding among subjects with osteoarthritis of the were selected.1 A random selection process was
knee.2,5,6 According to recent studies, subjects also electronically carried out allocating the subjects to
exhibit poor hip performance,7–9 with a consequent an experimental group and a control group . Sealed
negative effect on physical function.10 Exercise, envelopes were used to ensure the confidentiality
particularly muscle strengthening, has proven effec- of the selection process. Intention-to-treat analysis
tive in restoring range of motion, providing pain was applied using the data from the last available
relief and improving the ability to perform activities evaluation.
of daily living such as walking, going up and down Subjects were selected by telephone using a
stairs and playing sports.11–13 Two recent reviews database of patients with osteoarthritis from the
show that optimal exercise programs for knee oste- Universidade Federal de Sao Paulo (Brazil). The
oarthritis should focus on improving aerobic capac- participant’s selection was done by a rheumatolo-
ity, muscle strength with or without weight-bearing gist. The following were the inclusion criteria:
and such programs have a similar effect regardless osteoarthritis of the knee based on the criteria of
of patient characteristics, including radiographic the American College of Rheumatology;1 age
severity and baseline pain.14,15 between 40 and 70 years; and pain at rest between
Progressive resistance exercise is a muscle 3 and 8 out of 10 on the visual analog scale for
strengthening modality with a gradual, progressive one or both knees. The following were the exclu-
increase in weight load that has grown in popular- sion criteria: inflammatory conditions or any
ity over the past two decades.16–18 Few studies have medical condition that prevented physical activ-
focused on the use of progressive resistance exer- ity; joint injection in the previous three months;
cise, and fewer still consider the role of the hip regular physical activity at the time; or travel
muscles in exercise programs for subjects with plans for the subsequent 12 weeks. This study
osteoarthritis of the knee. Moreover, studies fail to received approval from the Human Research
describe the modality of the strengthening exer- Ethics Committee of the Universidade Federal de
cise, the load and the increase in load employed. A Sao Paulo (Brazil).
review published in 2010 show that most studies Evaluations were carried out at baseline (T0) as
reported minimal, if any, detail regarding the inten- well as after 45 (T45) and 90 days (T90) by the
sity of effort used while performing the exercise same blinded assessor. If necessary, both knees
sessions.19 The absence of detailed protocol exer- were treated, but only the more symptomatic knee
cise, that describes the exercise used including the was evaluated. The decision to evaluate the more
intensity and dose makes difficult the implementa- painful knee was based on the worst symptom
tion of the exercise protocol. described by patient.
Due to the lack of studies that precisely describe The following socio-demographic information
the correct application of progressive resistance was collected at baseline: age, race, years of
236 Clinical Rehabilitation 29(3)

education, paid work, body mass index (BMI) and || Global improvement at the end of treat-
a radiographic grading.20 ment was measured using a Likert scale
The following outcomes were measured at each through the following question: “How
evaluation time: would you describe your health state
today?” The five response options were:
•• Primary outcome: much worse, a little worse, unchanged, a
|| Pain measured using a horizontal visual little better and much better.27
analog scale ranging from 0 cm (no pain) || Side effects – patients were questioned
to 10 cm (unbearable pain) through the regarding the presence of pain and
following question: “On a scale of zero fatigue following exercise.
to ten, with zero being the absence of
pain and ten being unbearable pain, how All subjects were instructed to take 750 mg of
would you rate the intensity of your pain acetaminophen every eight hours when experienc-
today?”21 ing pain. When pain exceeded a 7 on the visual
•• Secondary outcomes: analog scale, the subject could take 50 mg of
|| Function as measured by the Brazilian diclofenac every eight hours. Both groups received
version of the Western Ontario McMaster a chart to record the doses of drugs taken during the
Universities Osteoarthritis (WOMAC) study period for the purposes of analysis.
index, which was administered in inter- Subjects in the experimental group underwent a
view form. This index has four domains: progressive resistance exercise program that
pain (0 to 20 points), stiffness (0 to 8), included four different exercises: knee extension/
function (0 to 68) and aggregate score flexion and hip abduction/adduction using two gym
(0 to 96), for which a higher final score machines (knee flexion-extension and abduction-
denotes a poorer state.22 adduction) with free weights (supplementary mate-
|| Quality of life as measured by the rial Figure 1). The exercises were preceded by a
Brazilian version of the 36-Item Short five-minute warm-up on an exercise bicycle. The
Form Health Survey (SF-36), which initial load was based on the 1RM. The program
included the following domains: physical was structured as follows: two sets of eight repeti-
function, physical role limitation, pain, tions, the first set employing 50% of 1RM and the
general health, vitality, social aspects, second set employing 70% of 1RM. A one-minute
emotional aspects and mental health. The rest interval was given between sets. The partici-
score ranges from 0 to 100, with higher pants’ pain threshold was closely monitored
scores denoting a better quality of life.23 throughout the exercises. The exercise program
|| Walking distance as measured by the six- was performed twice a week over a 12-week period
minute walk test, which is a functional following the recommendations of the American
test calculated in meters to measure College of Sports Medicine.28 Loads used in repeti-
walking performance.24,25 tions were reevaluated every two weeks. The pro-
|| Strength as measured by the one- gram of exercise was realized by a physiotherapist
repetition maximum (1RM), in which load with 05 years of experience in rheumatology.
is added based on the subject’s pain thresh- Adherence in the experimental group was calcu-
old to calculate the maximum load the lated proportionally to the total number of exercise
individual can tolerate in one repetition. sessions (24), with attendance at a greater number
The 1RM of four exercises was evaluated of sessions denoting greater adherence, it is impor-
using the same machines on which the tant to say that all the patient that attend the train-
exercises would be performed; the subject ing also performed the exercise as prescribed (load
was seated on the machine with their spine and repetitions). In cases of the interruption or
and feet supported by the chair.26 abandonment of treatment, the data were analyzed
Jorge et al. 237

Figure 1.  Flowchart of study.

using intention-to-treat analyses in both the experi- t-test was used to determine the time in which dif-
mental group and control group.29 The subjects in ferences occurred. The effect size of the interven-
the control group remained on a waiting list and tion was also assessed by calculating Cohen’s d
received the same treatment following the end of using the means and standard deviations of both
the study. groups.

Statistical analysis Results

The significance level was set at 5%, with 80% Sixty subjects were randomly allocated to one of
power and a standard deviation (SD) of 2 points the two groups, 29 in the experimental group and
for pain (visual analog scale), requiring a mini- 31 in the control group. Patient’s inclusion occurred
mum of 26 patients in each group. Descriptive between January of 2009 and July of 2011. There
statistics (absolute and percent frequencies) were were two dropouts in the experimental group and
used for the socio-demographic information. four in the control group as a result of personal and
Student’s t-test, Mann-Whitney and chi-square health-related problems. Both of the experimental
tests were performed to test differences between group dropouts returned for reevaluation and only
groups at baseline. Repeated-measure analysis of one of the control group dropouts returned. For the
variance (ANOVA) with Bonferroni’s correction remaining cases, intention-to-treat analysis was
was performed to determine differences in the employed using the last available evaluation
outcomes between groups over time. Student’s (Figure 1).
238 Clinical Rehabilitation 29(3)

Table 1.  Clinical and socio-demographic characteristics of subjects at baseline.

EG (n=29) CG (n=31) p
Age (years) 61.7 ± 6.4 59.9 ± 7.5 0.338
Race - Caucasian (%) 69% 71% 0.866
Years of education 4.9 ± 3.5 4.1± 3.1 0.514
Paid work (%) 6.9% 35.5% 0.011
BMI 30.6 ± 5.75 31.4 ± 4.42 1.000
Radiographic evaluation: grade I:II (%) 24.1:31.0 38.7:22.6 0.423

Data presented in mean ± SD or %; EG, experimental group; CG, control group.

Only the more painful knee was evaluated. the control group had a greater number of responses
Table 1 displays the socio-demographic character- of “a little worse” (Table 4).
istics. Groups were homogeneous at baseline Regarding side effects three subjects in the
except for paid work. Table 2 displays significant experimental group reported increased pain in the
group-time interaction effects for visual analog knees following the intervention. Mean adherence
scale for pain and WOMAC pain and function, but to the exercise program was 87.5% throughout the
not for WOMAC stiffness. In the experimental study in the experimental group.
group, pain was significantly lower at T45 and a
significant improvement in function occurred at
T90.
Discussion
Table 2 displays significant group-time interac- The present study investigated the effects of a pro-
tion effects for physical function, physical role gressive resistance exercise program on women
limitation and pain at T90, with better results in the with osteoarthritis of the knee and demonstrated
experimental group. positive results regarding pain, function, some
Table 3 displays the 1RM values for the four aspects of quality of life and all strength measures
exercises performed: extension, flexion, abduction beginning in the sixth week.
and adduction. Differences were found between Individuals with osteoarthritis of the knee expe-
groups beginning at T45 for the extensors, flexors rience weakness in the hip muscles.7–9 However,
and abductors, whereas significant differences for few studies have addressed the hip muscles in reha-
the adductors were found between groups begin- bilitation programs. In an attempt to fill this gap in
ning at T0 and continuing through the end of the knowledge, the present study highlights the
study. strengthening of the hip abductor and adductor
The effect size was calculated for the parame- muscles with a gradual increase in load.
ters that were statistically different with ANOVA Only two previous studies were found involving
and a large effect size was found to visual analog hip abductor and adductor strengthening and pro-
scale for pain, WOMAC, (pain and aggregate gressive resistance exercises. Both found improve-
score), SF-36 (physical function and physical role ment in muscle strength in the experimental
limitation) and 1RM for extensors, flexors, abduc- group.29,30 However, the lack of randomization in
tors and adductors (Table 2 and 3). the trial compromises the credibility of the results
Regarding medication use the participants in the of one these studies.29
experimental group took less acetaminophen Significant differences between groups were
throughout the study (Table 2). The global improve- found for pain, with the experimental group dem-
ment evaluated at the end of treatment using the onstrating better results. In a systematic review of
Likert scale, showed the experimental group had a resistance exercises for individuals with osteoar-
greater number of responses of “a little better” and thritis of the knee, 56% of the studies evaluated
 Jorge et al.

Table 2.  Mean ± SD values for visual analog scale, WOMAC and SF-36 in both groups.

T0 T45 T90 p ES 95% CI

(ANOVA)
  EG (n=29) CG (n=31) EG (n=29) CG (n=31) EG (n=29) CG (n=31)
VAS for pain (cm) 7.0 ± 1.3 7.0 ± 1.2 5.9 ± 2.0 7.0 ±1.6 4.3 ± 3.1 6.6 ±1.5 0.001 0,95 1,48 to-0,41
  p (t test) 0.920 0.022 0.001  
WOMAC  
 Pain 9.0 ± 2.9 9.3 ± 3.3 6.2 ± 4.0 9.6 ± 2.5 4.9 ± 4.2 9.5 ± 3.2 < 0.001 1,24 1,77 to 0,67
  p (t test) 0.721 < 0.001 <0.001  
 Stiffness 3.0 ±1.9 2.7 ± 1.8 2.2 ± 1.9 2.5 ± 1.7 1.9 ± 1.9 2.1 ± 1.6 0.388  
 Function 27.7 ± 9.3 28.4 ± 10.6 23.0 ± 10.2 27.0 ± 9.7 17.3 ± 12.4 26.7 ± 10.2 < 0.001 0,83 1,35 to 0,29
  p (t test) 0.787 0.124 0.002  
  Aggregate score 39.3 ± 12.4 40.4 ± 13.9 30.9± 14.1 39.1 ± 11.7 24.1 ± 17.6 38.3 ± 12.8 < 0.001 0,93 1,45 to 0,38
  p (t test) 0.760 0.017 0.001  
SF-36  
  Physical function 39.3 ± 16.3 32.4 ± 16.0 39.1 ± 17.0 30.3 ± 16.4 49.8 ± 21.9 30.8 ± 16.8 0.002 0,98 0,43 to 1,50
  p (t test) 0.165 0.045 0.000  
  Physical role limitation 25.9 ± 36.3 22.6 ± 26.1 37.9 ±37.6 21.8 ±27.2 48.3 ±41.7 16.9 ±23.6 0.002 0,94 0,39 to 1,46
  p (t test) 0.688 0.060 0.001  
 Pain 44.9 ± 21.9 39.0 ± 15.7 49.1± 22.0 41.3±13.8 58.6 ± 25.0 41.7±20.6 0.044 0,74 0,21 to 1,25
  p (t test) 0.233 0.105 0.006  
  General health 65.4 ± 22.3 53.1 ± 23.1 62.7 ± 20.2 58.0 ± 23.4 66.1 ± 21.8 52.6 ± 21.8 0.149  
 Vitality 55.9 ± 23.1 50.0 ± 24.2 56.7 ± 22.6 49.5 ± 21.5 64.0 ± 25.2 52.4 ± 21.3 0.586  
  Social aspects 62.1 ± 28.8 55.2 ± 30.8 72.4 ± 28.6 59.7 ± 24.3 77.2 ± 28.9 57.7 ± 27.9 0.187  
  Emotional aspects 56.3 ± 44.6 44.1 ± 38.9 55.2 ± 44.8 45.2 ± 31.7 72.4 ± 39.9 49.5 ± 39.3 0.302  
  Mental health 67.9 ± 19.9 61.5 ± 21.3 72.0 ± 17.8 55.9 ± 23.7 76.4 ± 18.7 59.5 ± 21.2 0.080  
  Six minute walk text 357.1 ± 56.9 330.2 ± 55.9 355.0 ± 66.9 339.4 ± 47.5 369.5 ± 60.8 343.1 ± 54.7 0.434  
  Acetaminophen use (tablets) 9,1± 9,5 13,5 ± 8,0  
  p (t test) 0.015  

T0, baseline; T45, 45 days after baseline; T90, 90 days after baseline; EG, experimental group; CG, control group; VAS, visual analogue scale; ES, effect size; CI, confidence
interval.
239
240 Clinical Rehabilitation 29(3)

Table 3.  Load in kilograms (1RM) ± SD for four exercises.

T0 T45 T90 p ES 95% CI

(ANOVA)
  EG (n=29) CG (n=31) EG (n=29) CG (n=31) EG (n=29) CG (n=31)
1RM extensors 8.2 ± 4.8 5.9 ± 3.7 10.4 ± 6.2 5.7 ± 3.9 11.8 ± 6.1 5.7 ± 4.2
p < 0.001 1,17 0,61 to
1,70
p (t test) 0.053 0.001 0.001  
1RM flexors 6.8 ± 2.5 5.6 ± 2.9 7.4 ± 3.1 5.2 ± 2.4 8.6 ± 2.5 5.3 ± 2.1 p = 0.002 1,43 0,85 to
1,98
p (t test) 0.096 0.003 0.000  
1RM abductors 25.6 ± 8.4 21.6 ± 8.7 29.2 ± 9.5 21.0 ± 8.3 32.7 ± 19.5 ± 6.7 p < 0.001 1,52 0,93 to
10.4 2,07
p (test) 0.078 0.001 0.000  
1RM adductors 19.8 ± 5.4 16.4 ± 5.7 24.1 ± 7.9 16.7 ± 6.2 26.7 ± 8.5 16.9 ± 7.9 p < 0.001 1,20 0,63 to
1,73
p (test) 0.023 0.000 0.000  

1RM, one-repetition maximum; T0, baseline; T45, 45 days after baseline; T90, 90 days after baseline; EG, experimental group; CG,
control group; ES, effect size; CI, confidence interval.

Table 4.  Global improvement evaluated at the end of Statistically significant differences were found
treatment (Likert scale). between groups for physical function and physical
role limitation subscales of the SF-36 as well as the
EG (n=29) CG (n=31)
function subscale and aggregate score of the
Much better 6 (20,7%) 7 (22,6%) WOMAC, with better results in the experimental
A little better 15 (51,7%)* 2 (6,5%) group. In a systematic review, Lange et al.16 found
Unchanged 7 (24,1%) 9 (29,0%) significant improvements in function. In studies
A little worse 1 (3,4%) 10 (32,3%)* involving exercise strengthening of the hip mus-
Much worse 0 (0%) 3 (9,7%) cles, Foroughi et al.29 and Sled et al.30 found no
Total 29 (100%) 31 (100%) significant differences on function. However,
EG, experimental group; CG, control group; *p value < 0.05.
Bennell et al.9 studied the effects of resistance
exercise on hip muscles and found significant dif-
ferences between groups.
found statistically significant improvements in Several quality of life aspects were evaluated,
pain, but most of the studies did not use a progres- but significant differences between groups were
sive increase in load and none incorporated any hip found only regarding pain and function. Other
muscle strengthening exercise.16 studies assessing this aspect in subjects with osteo-
Bennell et al.9 employed resistance exercises for arthritis of the knee used different exercise pro-
the hip muscles in subjects with osteoarthritis of grams from those employed in the present study
the knee and found positive effects regarding pain and reported inconsistent results. In the literature it
during movement and pain measured by the is possible to find studies that found significant
WOMAC. In studies involving progressive resist- results only on the pain scale,31 or only on the men-
ance exercise for the hip muscles, Sled et al.30 also tal health scale,32 or on four scales: physical func-
found positive effects regarding WOMAC scores, tion, physical role limitation, social aspects and
whereas Foroughi et al.29 found no significant dif- mental health.6
ference between groups using the pain subscale of A significant improvement was detected in
the WOMAC. 1RM in all four muscles analyzed. A systematic
Jorge et al. 241

review regarding effects of resistance exercises In the present study, 50% and 70% of the 1RM
reported significant improvements in strength. It were used in the first and second repetition, respec-
should be noted that only a few studies included a tively, to avoid muscle damage, with a readjust-
gradual increase in load and hip muscles were not ment of the load every two weeks. Foroughi et al.29
considered in the exercise programs. Jan et al.33 used 80% of the 1RM in an exercise program, with
compared high intensity exercise (60% of 1RM) the load readjusted every two weeks. Farr et al.17
and low intensity resistance exercises (10% of used 60 to 75% of 3RM in an exercise program,
1RM), readjusting the load every two weeks and with the load readjusted based on the results of the
evaluating strength with an isokinetic dynamome- Borg scale.
ter and found no statistically significant differences In the present study, significant improvements
between groups. Foroughi et al.29 and Sled et al.30 in pain, function and strength were found in the
reported improvements in hip muscle strength. experimental group beginning in the sixth week.
However, only the study of Foroughi et al.29 include This type of intermediate evaluation between base-
only women and the studies of Sled et al.30 and Jan line and the final evaluation was not employed in
et al.33 mixed men and women, which made the any other studies found, although two of these
comparison difficult. studies reported positive results after eight weeks
According to Verdijk et al.,26 the assessment of (final evaluation).30,33 Moreover, in the present
strength using 1RM is considered a sensitive, accu- study the experimental group continued to improve
rate method that reproduces muscle contraction throughout the study until the end, which is an
patterns employed in exercise intervention pro- important finding that should be considered when
grams, as the assessment of muscle strength and deciding the length of exercise programs.
training regimens are carried out using the same The mean adherence throughout the study was
machines. In the present study, the evaluation and high, however, no correlation was found between
progressive increase in load were performed using pain and patient adherence. Sled et al.30 reported
the same equipment and testing tool -1RM. an average of 78% adherence, Foroughi et al.29
A common characteristic found in all stud- reported an average of 85% adherence in the
ies6,17,29,30,32 was the calculation of strength using experimental group and 92% in the control group
an isokinetic dynamometer, but the progressive and Bennell et al.9 reported 96% and 89% adher-
increase in load was implemented in different ence in the experimental group and control group,
ways, such as the use of 1RM,29,34 bands with dif- respectively.
ferent resistance,30 exercises with different leg At the end of the study, the experimental group
weights,17 the Borg scale,6 a determined number of took a significantly lower amount of rescue medi-
repetitions more than two consecutive times32 and cations. Despite the limitations of the assessment
other forms of load progression.9 method, the reduction in the amount of analgesic
The choice of the 1RM value and the percentage and/or anti-inflammatory medications is an impor-
of this load to be used during the exercise program tant finding in a group that is at greater risk for the
was also based on the recommendations of the use of these medications. No other similar study
American College of Sports Medicine,28,35 which has assessed this variable.
state that low to moderate intensity exercises are Greater effect size (ES) values were found for
recommended with a load of 65 to 75% of 1RM to pain function and strength. Similar results were
enhance the benefits of exercise and simultane- reported by Jan et al.33 and Lange et al.16
ously reduce the risk of injury; moreover, a gradual The present study has strengths that merit con-
increase in load is essential. However, this percent- sideration. A supervised exercise program was
age is for healthly adults and to avoid injuries in used with a gradual increase in load and the inclu-
our patients that already have an osteoarticular dis- sion of hip muscle strengthening for the treatment
ease we decided to start with a low percentage and of subjects with osteoarthritis of the knee.
progress the resistance during the study. Moreover, the subjects underwent an intermediate
242 Clinical Rehabilitation 29(3)

assessment during the study, which allowed for the

identification of precisely when change began Clinical messages
occurring. These features appear only rarely in
other studies. •• A progressive resistance exercise program
There were some side effects caused by the including hip muscles was effective in
treatment or that occurred during treatment. Three reducing pain and improving function and
individuals in the experimental group reported strength in women with osteoarthritis of
increased pain in the knees and, as a result, were the knee.
taken to the physician responsible for the care of •• The progressive resistance exercise pro-
the subjects during the study to receive effective gram did not improve walking distance,
treatment. These subjects had their medication or and this is an important problem of
its dosage modified and did not participate in the women with osteoarthritis of the knee.
exercise program for one week. Following this rest •• This progressive resistance exercise pro-
period, they returned to the program, and no further gram could be part of rehabilitation process
complaints were registered. of women with osteoarthritis of the knee,
Hoogeboom et al.36 developed a scale to eval- however other supporting treatments are
uate the validity of therapeutic exercise pro- necessary to address all problems related to
grams. They recommended that exercise osteoarthritis of the knee in women.
programs should be described in sufficient detail
to enable readers to understand how the interven-
Conflict of interest
tion was actually carried out. This scale ranges
from 0-9 with higher results showing better valid- The authors declare that there is no conflict of interest.
ity. We tried in our study to follow the scale pro-
posed by Hoogeboom et al.36 covering the five Funding
areas evaluated that were: patient selection, ther- The present study was supported by grants from
apist and setting selection, rationale, content, and Brazilian fostering agencies - Fundação de Amparo à
adherence, and are supported by evidence from Pesquisa do Estado de São Paulo (FAPESP) and
the literature, to enhance the validity of our exer- Coordenação de Aperfeiçoamento de Pessoal de Nível
cise program Superior (CAPES).
The present study had the following limita-
tions: the time of the initial diagnosis was not References
included; the medication use before the study start 1. Hochberg MC, Altman RD, Brandt KD, et al. Guidelines for the
was not recorded; only women participated what medical management of osteoarthritis. Part II. Osteoarthritis
of the knee. American College of Rheumatology. Arthritis
make impossible to extrapolate the results to other Rheum 1995; 38: 1541–1546.
genders; the control group did not undergo any 2. Wollheim FA. Osteoarthritis. Curr Opin Rheumatol 2002;
type of intervention and no follow-up was carried 14: 571–572.
out to determine the long-term impact of progres- 3. Srikant Srikanth VK, Fryer B, Math JL, Zhai G,
sive resistance exercise in patients with osteoar- Winzenberg TM, Hosmer D, et al. A meta-analysis of sex
differences prevalence, incidence,and severity of osteoar-
thritis of the knee. Further studies are needed to thritis. Osteoarthr Cartil 2005; 13: 769–781
assess long term impact in this disease and rheu- 4. Brooks PM. Impact of osteoarthritis on individuals and
matic conditions. society: how much disability? Social consequences and
In conclusion, the proposed progressive resist- health economic implications. Curr Opin Rheumatol
ance exercise program, which included hip muscle 2002; 14: 573–577.
5. Slemenda C, Brandt KD, Heilman DK, et al. Quadriceps
strengthening, was effective in reducing pain as weakness and osteoarthritis of the knee. Ann Intern Med
well as improving function, aspects of quality of 1997; 127: 97–104.
life and strength in women with osteoarthritis of 6. Baker KR, Nelson ME, Felson DT, et al. The efficacy of
the knee. home based progressive strength training in older adults
Jorge et al. 243

with knee osteoarthritis: a randomized controlled trial. J 22. Fernandes MI. Translation and validation of quality of
Rheumatol 2001; 28: 1655–1665. life questionnaire relating specifically to osteoarthritis
7. Yamada H, Koshino T, Sakai N, et al. Hip adductor mus- WOMAC (Western Ontario McMaster Universities) into
cle strength in patients with varus deformed knee. Clin the Portuguese language, PhD thesis. São Paulo: Federal
Orthop Relat Res 2001; 386: 179–185. University of São Paulo, 2003.
8. Costa RA, Oliveira LM, Watanabe SH, et al. Isokinetic 23. Ciconelli RM, Ferraz MB, Santos W, et al. Translation
assessment of the hip muscles in patients with osteoarthri- into the Portuguese language and validation of generic
tis of the knee. Clinics (Sao Paulo) 2011; 65: 1253–1259. quality of life evaluation questionnaire. SF-36 (Brasil
9. Bennell KL, Hunt MA, Wrigley TV, et al. Hip strengthen- SF-36). Rev Bras Reumatol 1999; 39: 143–150.
ing reduces symptoms but not knee load in people with 24. Enright PL and Sherrill DL. Reference equations for the
medial knee osteoarthritis and varus malalignment: a ran- six-minute walk in healthy adults. Am J Respir Crit Care
domised controlled trial. Osteoarthritis Cartilage 2010; Med 1998; 158: 1384–1387.
18: 621–628. 25. Guyatt GH, Sullivan MJ, Thompson PJ, et al. The 6-min-
10. van Dijk GM, Veenhof C, Spreeuwenberg P, et al. ute walk: a new measure of exercise capacity in patients
Prognosis of limitations in activities in osteoarthritis of the with chronic heart failure. Can Med Assoc J 1985; 132:
hip or knee: a 3-year cohort study. Arch Phys Med Rehabil 919–923.
2010; 91: 58–66. 26. Verdijk LB, van Loon L, Meijer K and Savelberg HH.
11. Jordan KP, Wilkie R, Muller S, Myers H and Nicholls One-repetition maximum strength test represents a valid
E. Measurement of change in function and disability in means to assess leg strength in vivo in humans. J Sports
osteoarthritis: current approaches and future challenges. Sci 2009; 27: 59–68.
Curr Opin Rheumatol 2009; 21: 525–530. 27. Bolognese JA, Schnitzer TJ and Ehrich EW. Response
12. Rannou F and Poiraudeau S. Non-pharmacological relationship of VAS and Likert scales in osteoarthritis
approaches for the treatment of osteoarthritis. Best Pract efficacy measurement. Osteoarthritis Cartilage 2003; 11:
Res Clin Rheumatol 2010; 24: 93–106. 499–507.
13. Silva LE, Valim V, Pessanha APC, et al. Hydrotherapy 28. Kraemer WJ and Ratamess NA. Fundamentals of resist-
versus conventional land-based exercise for the manage- ance training: progression and exercise prescription. Med
ment of patients with osteoarthritis of the knee: a rand- Sci Sports Exerc 2004; 36: 674–688.
omized clinical trial. Phys Ther 2008; 88: 12–21. 29. Foroughi N, Smith RM, Lange AK, et al. Progressive
14. Juhl C, Christensen R, Roos EM, Zhang W and Lund H. resistance training and dynamic alignment in osteoar-
Impact of exercise type and dose on pain and disability thritis: A single-blind randomised controlled trial. Clin
in knee osteoarthritis: a systematic review and meta- Biomech 2011; 26: 71–77.
regression analysis of randomized controlled trials. 30. Sled EA, Khoja L, Deluzio KJ, et al. Effect of a home
Arthritis Rheumatol 2014; 66: 622–636. program of hip abductor exercises on knee joint loading,
15. Tanaka R, Ozawa J, Kito N and Moriyama H. Efficacy of strength, function and pain in people with knee osteoar-
strengthening or aerobic exercise on pain relief in people thritis: a clinical trial. Phys Ther 2010; 90: 895–904.
with knee osteoarthritis: a systematic review and meta- 31. O’Reilly SC, Muir KR and Doherty M. Effectiveness of
analysis of randomized controlled trials. Clin Rehabil home exercise on pain and disability from osteoarthritis
2013; 27: 1059–1071. of the knee: a randomised controlled trial. Ann Rheum Dis
16. Lange AK, Vanwanseele B and Fiatarone Singh MA. 1999; 58: 15–19.
Strength training for treatment of osteoarthritis of the 32. Mikesky AE, Mazzuca SA, Brandt KD, et al. Effects of
knee: a systematic review. Arthritis Rheum 2008; 59: strength training on the incidence and progression of knee
1488–1494. osteoarthritis. Arthritis Rheum 2006; 55: 690–699.
17. Farr JN, Going SB, McKnight PE, et al. Progressive resist- 33. Jan MH, Lin JJ, Liau JJ, et al. Investigation of clinical
ance training improves overall physical activity levels in effects of high- and low-resistance training for patients
patients with early osteoarthritis of the knee: a randomized with knee osteoarthritis: a randomized controlled trial.
controlled trial. Phys Ther 2010; 90: 356–366. Phys Ther 2008; 88: 427–436.
18. Lombardi I Jr, Magri AG, Fleury AM, et al. Progressive 34. Mazini Filho ML, Rodrigues BM, Reis AC, et al. Análise
resistance training in patients with shoulder impingement do teste de uma repetição máxima no exercício supino
syndrome: a randomized controlled trial. Arthritis Rheum para predição da carga. Braz J Biomotr 2011; 4: 57–64.
2008; 59: 615–622. 35. Ratamess N, Alvar BA, Evetoch TK, et al. American
19. Iversen MD. Managing hip and knee osteoarthritis College of Sports Medicine position stand. Progression
with exercise: what is the best prescription? Ther Adv models in resistance training for healthy adults. Med Sci
Musculoskel Dis 2010; 2: 279–290. Sports Exerc 2009; 41: 687–708.
20. Kellgren JH and Lawrence JS. Radiological assessment of 36. Hoogeboom TJ, Oosting E, Vriezekolk JE, et al. Therapeutic
osteo-arthrosis. Ann Rheum Dis 1957; 16: 494–402. validity and effectiveness of preoperative exercise on func-
21. Scott J and Huskisson EC. Vertical or horizontal visual tional recovery after joint replacement: a systematic review
analogue scales. Ann Rheum Dis 1979; 38: 560. and meta-analysis. PLoS One 2012; 7: e38031.