Øiestad et al.

BMC Musculoskeletal Disorders (2023) 24:714 BMC Musculoskeletal

https://doi.org/10.1186/s12891-023-06831-x
Disorders

RESEARCH Open Access

The efficacy of strength or aerobic exercise

on quality of life and knee function in patients
with knee osteoarthritis. A multi-arm
randomized controlled trial with 1-year follow-
up
Britt Elin Øiestad1*, Asbjørn Årøen2, Jan Harald Røtterud3, Nina Østerås4, Even Jarstad5, Margreth Grotle1,6 and
May Arna Risberg7,8

Abstract
Objective To evaluate the efficacy of strength exercise or aerobic exercise compared to usual care on knee-related
quality of life (QoL) and knee function at 4 months and 1 year in individuals with knee osteoarthritis.
Methods A three-arm randomized controlled trial (RCT) compared 12 weeks of strength exercise or aerobic exercise
(stationary cycling) to usual care supervised by physiotherapists in primary care. We recruited 168 participants aged
35–70 years with symptomatic knee osteoarthritis. The primary outcome was The Knee Injury and Osteoarthritis
Outcome Score (KOOS) QoL at 1 year. Secondary outcomes were self-reported function, pain, and self-efficacy, muscle
strength and maximal oxygen uptake (VO2max) at 4 months and 1 year.
Results There were no differences between strength exercise and usual care on KOOS QoL (6.5, 95% CI -0.9 to 14),
or for aerobic exercise and usual care (5.0, 95% CI -2.7 to 12.8), at 1 year. The two exercise groups showed better
quadriceps muscle strength, and VO2max at 4 months, compared to usual care.
Conclusion This trial found no statistically significant effects of two exercise programs compared to usual care on
KOOS QoL at 1 year in individuals with symptomatic and radiographic knee osteoarthritis, but an underpowered
sample size may explain lack of efficacy between the intervention groups and the usual care group.
ClinicalTrials.gov Identifier NCT01682980.
Keywords Knee osteoarthritis, Quality of life, Exercise

4
*Correspondence: Center for treatment of Rheumatic and Musculoskeletal Diseases
Britt Elin Øiestad (REMEDY), Diakonhjemmet Hospital, Oslo, Norway
5
[email protected] Norwegian Sports Medicine Clinic, Oslo, Norway
1 6
Department of Rehabilitation Sciences and Health Technology, Oslo Department of Research and Innovation, Division of Clinical
Metropolitan University, Oslo, Norway Neuroscience, Oslo University Hospital, Oslo, Norway
2 7
Orthopedic department, Akershus University Hospital, Lørenskog, Division of Orthopedic Surgery, Oslo University Hospital, Oslo, Norway
8
Norway Department of Sports Medicine, Norwegian of School Sport Sciences,
3
Orthopaedic department, Akershus University Hospital, Lørenskog, Oslo, Norway
Norway

© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use,
sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included
in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The
Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available
in this article, unless otherwise stated in a credit line to the data.
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 2 of 12

Introduction QoL at 1 year in individuals with symptomatic knee

Knee osteoarthritis is a world-wide leading cause of pain, osteoarthritis.
disability, and reduced quality of life (QoL) [1, 2]. There The secondary objectives were to investigate the effi-
is no cure for knee osteoarthritis, but exercise interven- cacy of respectively strength exercise or aerobic exercise
tions have shown beneficial, mostly short-term (post- compared to usual care, on pain, symptoms, function
intervention) effects on pain and function compared to in activities of daily living (ADL) and sport/recreation,
no or minimal treatment, usual care, and placebo treat- self-efficacy for pain and symptoms, health-related QoL,
ments [3–9]. Exercises are currently being implemented muscle strength and cycle-specific maximal oxygen
as the first-line treatment in clinical practice across coun- uptake (VO2max) at 4 months and 1 year in individuals
tries, even though the effect sizes are at best moderate with symptomatic knee osteoarthritis.
[10, 11]. Exercise interventions involve many different
activities, such as strength and balance exercises, walking Methods
programs, cycling, or water-based exercises, which make Trial design
a good base for personalized treatment options. Strength This three-arm parallel randomized controlled trial
exercises are likely among the most important interven- (RCT) is reported according to the Extension of the
tions due to the relatively strong association between CONSORT 2010 Statement [16] and the CHAMP state-
lower knee extensor strength and increased risk of symp- ment [17]. The study was designed to compare each one
tomatic and functional deterioration in individuals with of the exercise programs to a usual care group on Knee
symptomatic knee osteoarthritis [12]. However, fewer injury and Osteoarthritis Outcome score (KOOS) QoL at
studies have reported the effects of strength training on 1 year as the primary endpoint [18]. This was described
knee-related QoL in younger populations beyond short- in the clinicaltrials.gov (NCT01682980, 11/09/2012) doc-
term post-intervention assessments [9]. The knee-related uments prior to the start of the study. The changes in the
QoL outcome has been proven to be more sensitive and trial from the study protocol are [18]: We had to stop the
responsive to detect changes in younger and more active trial before we had included the planned 207 participants
populations than pain and physical function [13] and because the recruitment rate was slow, and we lacked
represents a good complement to the pain and function the resources to conduct the recruitment. With only
outcomes. Furthermore, the OMERACT-OARSI group one person available for the recruitment job throughout
suggests that QoL is one of the core outcome domains for the study period, we decided to terminate the trial when
measurement in clinical trials of hip and knee osteoar- Covid-19 restrictions were implemented in Norway. We
thritis [14]. could not do the planned magnetic resonance imaging
Also, less is known about the efficacy of aerobic exer- (MRI) assessments and blood samples of the study par-
cises alone on knee-related QoL. Of 48 trials in a system- ticipants due to lack of funding to conduct these parts.
atic review [6], none examined moderate “single-type” The inclusion criterion for age was extended from ini-
exercises such as cycling. Another more recent system- tially 45–65 to 35–70 years of age (after inclusion of 42
atic review [15] summarizing the effects of cycling on participants) to reach out to more eligible and younger
pain, function, and stiffness in individuals with knee participants. We did not initially design the RCT to
osteoarthritis concluded that stationary cycling relieves compare the two exercise interventions as this may have
pain and improves sport function, but was not clinically required a huge sample size to detect clinically important
effective for improving stiffness, daily activity, or QoL. differences.
The authors requested more trials to clarify the effects of
stationary cycling. Participants
There is still lack of evidence on the efficacy of differ- Participants with a confirmed diagnosis of symptomatic
ent types of structured, easily available exercise programs and radiographic knee osteoarthritis were recruited from
beyond post-intervention assessment on knee-related primary care, orthopedic departments at three hospitals
QoL in younger patients with symptomatic knee osteoar- in the greater Oslo area, and from newspaper advertise-
thritis. The primary objective of this trial was to evaluate ments, from April 2013 to March 2020. We recruited
the efficacy of a standardized strength exercise program, participants with confirmed radiographic Kellgren and
or a standardized aerobic exercise program compared to Lawrence grades [19] 2 and 3 defined using the SynaF-
usual care at 1 year on knee-related QoL in individuals lexer frame (Synarc Inc, Newark, CA). The participants
with symptomatic knee osteoarthritis. We hypothesized had to fulfill 3/4 of the American College of Rheuma-
that (1) Strength exercise was more effective than usual tology (ACR) clinical criteria (stiffness < 30 min, crepi-
care on knee-related QoL at 1 year in individuals with tus, osteophytes, pain the last days of the last month)
symptomatic knee osteoarthritis, and (2) Aerobic exer- [20], being aged 35–70 years, and having no other seri-
cise was more effective than usual care on knee-related ous physical or mental illnesses preventing them from
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 3 of 12

participating in the trial (e.g. cancer under treatment phoned the participants at 6 months and 9 months for an
or unstable coronary heart disease). We excluded eli- interview of health care utilization the last three months,
gible participants who self-reported body mass index including a question regarding the frequency of physiother-
(BMI) > 35 kg/m2 because we believed they needed an apist consultations.
additional weight loss program. In addition, we excluded
those who were scheduled for surgery in the nearest 6 Primary outcome
months, those who already participated in structured, The primary outcome was knee-related QoL measured by
weekly, moderate strength training or cycling, those who the patient-reported KOOS QoL [23] at 1 year. The KOOS
had known serious musculoskeletal impairments in the is a valid and reliable self-reported questionnaire with five
lower extremities or low back, having prostheses in the sub-scores measuring pain, symptoms, activities of daily liv-
lower extremities, those with serious coronary heart dis- ing, function in sport and recreation (KOOS Sport/recre-
eases or cancer, and those who did not speak Norwegian ation), and knee-related QoL for patients with knee injuries
language. and osteoarthritis [13]. The subscales range from 0 to 100
with 0 representing the worst possible score, and 100 best
Randomization and blinding possible score.
Computer-generated randomization lists were prepared
by a biostatistician not involved in the project. Partici- Secondary outcomes measured at 4 months and 1 year
pants were randomly allocated with 1:1:1 ratio within Secondary outcomes were the KOOS subscales (0-100
block sizes of 6. A research coordinator prepared con- for pain, other symptoms, activities of daily living (ADL),
cealed envelopes from the randomization list for four Sport/recreation, and QoL (at 4 months). Other secondary
recruitment centers before recruitment (three orthopedic outcomes were knee pain the previous week (numeric rat-
departments and one for primary care involving physical ing scale (NRS) scale, 0–10), health-related QoL measured
therapy clinics/advertisements). The self-reported data by EuroQoL-5 Dimentions-5 Levels (EQ-5D-5 L) [24],
and objective outcomes assessors were blinded for group the 0-100 scale of patient-reported health status, and the
assignment. EQ-5D-5 L index. The index was calculated using the UK
value set as described by Garrett et al. [25]. Self-efficacy for
Assessments and outcomes pain (5–25) and function (6–30) was measured using the
A sheet was completed by an assessor and the partici- Arthritis Self-Efficacy Scale (ASES) [26]. The Global rating
pant at the time of recruitment including data on: age, of change scale (GRC) 7-point version was completed at
sex, self-reported height and weight, affected knee (right/ both follow-ups [27], asking “how are your knee complaints
left/both), year of diagnosis, osteoarthritis in the fam- now compared to the previous assessment”. Isokinetic knee
ily, ACR criteria, knee pain most days the last month extension strength was tested in a dynamometer (Biodex
(yes/no), scheduled surgery in any joint (yes/no), known 2000 System: Biodex Medical Systems, Shirley, NY) with
severe physical or psychological disorders, drug abuse, the participant in a sitting position with belts ensuring that
physical activity level and physical activity level index only the knee joint could move. Concentric knee extension
(type*frequency* intensity) [21], smoking (yes/no), pre- and knee flexion in a range of 0–90 degrees with five repeti-
vious injuries or surgeries in the knees or hips (type of tions at 60°/sec were tested. The assessor gave verbal feed-
injury/no injury), and known heart diseases for the par- back to all the patients to encourage maximal effort. Muscle
ticipant and their parents or siblings (yes/no). Additional strength was quantified based on peak torque in Newton
data was collected at the baseline test: objectively mea- meters (Nm) and the peak torque per kilogram body weight
sured height and weight, educational level, work status, (Nm/kg). Peak torque was reported as the highest value
and a numeric rating scale of average pain (NRS) in the among the five repetitions. VO2max was assessed using an
affected knee last week (0–10). We also assessed fron- incremental test procedure on a cycle ergometer (Monark
tal plane alignment using an inclinometer [22] and knee 839E, Sweden), after a 20-minute progressive warm-up
range of motion by a goniometer (data not shown). [28]. For this purpose, two metabolic analyzers were used;
Assessments were performed before random group allo- a Sensor Medics VMax29 with a mixing chamber (Vyaire
cation (baseline), and at post-intervention (4 months), and Medical, Höchberg, Germany), or a Vyntus CPX with a
at 1 year. The participants started with a warm-up on a sta- breath-by-breath system (Vyaire Medical, Höchberg, Ger-
tionary cycle. Then the VO2max test was conducted before many), where the same analyzer was used at pre- and post-
the isokinetic muscle strength tests. All participants had a test for correct comparison. During the test procedure, the
5-10-minute break between the cycle and muscle strength workload was increased by 25 Watts every 30 s to a supra-
tests. In the end, the participant completed the patient- maximal workload and expected volitional exhaustion
reported outcomes. This procedure was used for all the tests within ~ 4–6 min. The cadence was customized individu-
to ensure consistent order of the physiological tests. We also ally and was increased from ~ 50–75 revolutions per minute
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 4 of 12

to ~ 90–110 revolutions per minute at peak workload. The and pain during and after the exercise session (NRS, 0–10).
main criterion for achievement of VO2max was the classic Adequate adherence to the exercise interventions was
leveling off of oxygen uptake (VO2), despite increased work- defined as completing 80% of the total number of sessions
load [29]. Secondary criteria used to validate the attainment prescribed (2 sessions per week*12 weeks = 24 sessions).
of VO2max and indicate maximal effort during the incremen-
tal test was a plateau in VO2, despite increased pulmonary Strength exercise program
ventilation [30], a peak pulmonary respiratory gas-exchange The strength exercise program was an individual, super-
ratio of > 1.10 [31], a rate of perceived exertion of ≥ 17 on vised program based on a previously developed exercise
the BORG6 − 20 scale [30], and visible exhaustion of the sub- program for knee patients [38], including balance exercises,
ject [32]. For confirmation of a satisfying test procedure, at and resistance exercises (Appendix File 2). The physiothera-
least two of the five criteria mentioned above had to be met. pist individualized the program according to the patients`
VO2max, reported in mL·kg− 1·min− 1 in the present study, impairments (pain, swelling, muscle strength, and neu-
was calculated as the mean of the two highest consecutive romuscular control). The dose for the strength exercises
30-second VO2 measurements [33, 34]. Maximal heart rate was planned according to ACSM`s guidelines for strength
(HRmax) was estimated by peak heart rate achieved during progression in healthy adults [37] and an intensity of 8–12
the incremental test + 5 beats·min− 1 [35] by wearing a heart repetitions maximum (RM). The program consisted of a
rate monitor (Polar S210, Polar Electro Oy, Kempele, Fin- variety of exercises for balance (neuromuscular control) and
land). A cardiovascular examination was performed prior six exercises for muscle strength for the following muscle
to the test to exclude those with severe and unstable coro- groups: quadriceps, hamstring, hip abductors and extensors,
nary heart diseases, and an electrocardiogram assessment and calf muscles. Prior to each exercise session, a warm-up
was included of all participants > 50 years of age during the on a stationary bike or a treadmill was performed for 5 min.
incremental test. Each patient’s neuromuscular function was decisive for
how and when to advance the neuromuscular exercises, for
Exercise programs and usual care instance, from standing on both legs to one leg or using dif-
The exercise interventions are described according to the ferent surfaces such as foam mats, wobble boards, or bosu
Consensus on Exercise Reporting Template (CERT) [36] balls. For the strength exercises, the patients were encour-
(Appendix File 1). The intervention period started as soon aged to follow the “2 + principle”, where the weight load was
as possible after the baseline assessment with individual increased on the next exercise session when the patient was
follow-up by a physiotherapist with previous clinical experi- able to perform at least two more repetitions than planned
ence with osteoarthritis patients. The physiotherapists were on the last set [39].
located at clinics near the participants’ homes. The physio-
therapists received one oral explanation of the project and Aerobic exercise program
the two exercise programs, and received a print of the pro- The aerobic exercise program was conducted on a station-
grams, from our research coordinator before meeting their ary cycle and based on guidelines for training parameters
first participant. The physiotherapists could contact our in people with pain associated with osteoarthritis [40]. The
research coordinator for questions at any time point dur- participants were told to cycle 2–3 times per week for 12
ing the intervention period. The physiotherapist clinics were weeks, including a warm-up for 10 min, then 30 min on
differently equipped, but all had equipment for leg-press moderate intensity (70–80% HRmax) which we considered
and leg-extension exercises, and all had stationary bicycles. to be moderate loading on the knee joint), and finish with
Both interventions started with a two-week preparation 5 min on low intensity. The participants were told to have
phase to adapt to the program. Then both groups were told a 2-week preparation phase with low intensity and shorter
to exercise 2–3 times per week for 12 weeks (at least 2 ses- sessions to avoid “too-much-too-soon” complaints.
sions per week supervised by the physiotherapist and the
third could be a home session) according to the American Usual care
College of Sports Medicine`s (ACSM) guidelines for exer- Participants randomized to the usual care group were told
cise in untrained people [37]. The participants were told not to live as they usually did but to avoid starting a new exercise
to change their usual physical activities during the 12-week program involving structured strength exercise or cycling
intervention, but this was not systematically recorded until the 4-month follow-up was completed. The partici-
other than self-reporting of physical activity level. The pants in this group could get access to the exercise program
interventions involved exercises only, but the PTs were not after the post-intervention test was conducted, but < 5 par-
instructed to refrain from giving advice on lifestyle changes, ticipants in the usual care group asked for the programs.
including continuing the exercise program post-interven-
tion. The participants completed a training diary including
details about type of exercise, frequency, intensity, duration,
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 5 of 12

Fig. 1 Participant flow during the trial. *Final sample excluded seven participants who withdrew their informed consent to participate (strength exercise:
n=3, aerobic exercise: n=2, usual care: n=2)

Sample size Statistical methods

The sample size was calculated to detect a suggested clini- The statistical analyses were pre-planned prior to par-
cally important difference on the KOOS QoL of 10 points ticipant enrolment and published at clinicaltrials.gov
[13] (β = 0.2, two-sided α = 0.05) with a standard deviation (NCT01682980) before we revealed the group alloca-
(sd) of 20 points. This estimation gave 63 in each group, in tion. Descriptive baseline characteristics are presented for
total of 189 participants. We planned for a total of 69 in each each group separately as number (n) and percentage (%),
group (n = 207), including a drop-out rate of 10%. mean and standard deviations, or median and minimum-
maximum (min-max) values depending on the type of the
variable or its distribution. All outcomes were compared
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 6 of 12

between each intervention group and the usual care group Ethical considerations
at the 4-month and 1-year follow-ups. To test the hypoth- All participants signed an informed consent prior to the
eses for the primary outcome at 1 year, intention-to-treat baseline assessment. The Regional Ethical Committee in the
mixed linear model with restricted maximum likelihood Health Region South-East in Norway approved the study
(REML) solution was applied using KOOS QoL data as protocol (REK 10/223), and the Data Inspectorate at Oslo
the dependent variable. Because 7 participants withdrew University Hospital approved the study.
their informed consents, we had to exclude these from our
dataset. Consequently, a modified intention-to-treat analy- Results
sis was conducted. The baseline score for KOOS QoL were The trial randomized 168 participants: 55 to strength exer-
included as covariate, the participant’s ID variable as a ran- cise, 56 to aerobic exercise, and 57 to usual care. Seven par-
dom effect, and intervention group, time, and group*time ticipants withdrew their informed consent (due to lack of
as fixed effect factors. The statistical analyses for second- time) and 19 participants showed up only on the baseline
ary outcomes were conducted using the same approach as assessment. The participant flow and reasons for loss to fol-
for the primary outcome. No adjustments for multiplicity low-up are shown in Fig. 1.
were applied. Statistical Package for Social Sciences (IBM© Participants’ baseline characteristics are presented in
SPSS© Statistics version 27) was used for the statistical anal- Table 1. The groups were balanced on baseline characteris-
yses. Cost-effectiveness analyses will be reported in a sepa- tics as compared to the usual care group, except for more
rate paper. men and smokers in the usual care group. Sixty-eight phys-
iotherapists treated the participants in the two exercise
intervention groups. Training diaries were delivered by 60%
in the strength exercise group (33/55) and 57% in the aero-
Table 1 Baseline characteristics of the study participants bic exercise group (32/56). Of those who delivered diaries,
(n = 161) 77% (33 of 43 in the strength exercise group) and 80% (32
Characteristics Strength Aerobic Usual of 40 in the aerobic exercise group) completed ≥ 80% of the
exercise exercise care prescribed training sessions. Physical activity level and most
(n = 54) (n = 53) (n = 54) frequent activity type at the three time-points are presented
Sex, men, n (%) 24 (44) 25 (47) 30 (56) in Appendix File 3. Descriptive data outcomes are given in
Age (years), mean (sd) 57.6 (6.6) 57.3 57.8 Table 2.
(7.1) (7.4)
Body mass index, mean (sd) 28.9 (4.3) 29.4 28.4
(4.4) (4.1)
Efficacy of strength exercise or aerobic exercise compared
Smoking, yes, n (%) 2 (4) 6(12) 13 (24) to usual care at 1 year
Education: >4 years of college or 15 (28) 15 (28) 12 (22) There were no statistically significant differences between
university, n (%) strength exercise group and usual care group (6.5, 95% CI
Occupational status, n (%)* -0.9 to 14), or between aerobic exercise group and usual
• Working 46 (94) 44 (88) 41 (80) care group (5.0, 95% CI -2.7 to 12.8) for KOOS QoL at
• Sick leave/Retired 3 (6) 6 (12) 10 (20) the 1-year follow-up (Table 3; Fig. 2) beyond the adminis-
Self-reported knee osteoarthritis*, n tration of the supervised interventions.
(%)
• One knee 31 (60) 39 (75) 32 (60) Secondary outcomes for strength exercise group vs. usual
• Both knees 21 (40) 13 (25) 21 (40)
care group
Any previous injuries - involved knee, 26 (48) 30 (57) 16 (30)
n (%) At the 4-month follow-up, the strength exercise group
Any previous injuries - uninvolved 8 (15) 13 (25) 15 (28) had statistically significant better scores than the usual
knee care group on quadriceps strength (14.3, 95% CI 5.3 to
Osteoarthritis in parents or siblings, 26 (50) 20 (41) 23 (43) 23.3 Nm and 0.21, 95% CI 0.11 to 0.31 Nm/BW), and
n (%) VO2max (1.9, 95% CI 0.85 to 3.0 mL·kg− 1·min− 1)(Table 3).
Other pain/injuries (back, hip, ankle), 11 (20) 16 (30) 10 (19) At the 1-year follow-up, the strength exercise group
n (%) had statistically significant better results than usual care
Self-reported known heart disease, 15 (28) 19 (36) 10 (19) group for self-efficacy for pain (2.1, 95% CI 0.2 to 4.0)
n (%)
(Table 3).
Physiotherapy consultations post-
intervention, n (%)
• 4–6 months 11 (22) 15 (33) 12 (27) Secondary outcomes for aerobic exercise vs. usual care
• 7–9 months 9 (18) 11 (26) 17 (38) The aerobic exercise showed better KOOS symptoms
• 10–12 months 5 (12) 9 (24) 14 (37) score (7.4, 95% CI 0.4 to 14.4), better quadriceps strength
N = number, sd = standard deviation, *data had missing values (10.1, 95% CI 0.9 to 19.3 Nm and 0.15, 95% CI 0.05 to
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 7 of 12

Table 2 Descriptive values for the primary and secondary outcome measures
Outcomes Strength exercise Aerobic exercise Usual Care
KOOS QoL (0-100)
• Baseline 37 (19) 33 (15) 37 (20)
• 4 months 47 (22) 42 (20) 39 (23)
• 1 year 47 (25) 43 (20) 40 (26)
KOOS Pain (0-100)
• Baseline 57 (17) 55 (17) 52 (20)
• 4 months 63 (22) 62 (20) 55 (22)
• 1 year 64 (24) 60 (22) 53 (28)
KOOS Symptoms (0-100)
• Baseline 64 (18) 57 (15) 57 (20)
• 4 months 68 (20) 63 (18) 55 (20)
• 1 year 70 (21) 62 (18) 58 (25)
KOOS ADL (0-100)
• Baseline 66 (20) 64 (21) 61 (21)
• 4 months 71 (24) 71 (24) 63 (21)
• 1 year 72 (27) 69 (22) 58 (30)
KOOS Sport (0-100)
• Baseline 31 (22) 27 (19) 27 (24)
• 4 months 42 (27) 33 (22) 30 (26)
• 1 year 37 (27) 33 (32) 33 (32)
Pain last week (0–10)
• Baseline 4.9 (2.1) 4.6 (1.9) 5.3 (2.3)
• 4 months 3.9 (2.8) 3.7 (2.2) 4.5 (2.4)
• 1 year 4.0 (2.4) 3.7 (2.2) 5.1 (2.9)
EQ-5D-VAS (0-100)
• Baseline 67 (16) 62 (16) 64 (18)
• 4 months 70 (19) 67 (22) 62 (19)
• 1 year 66 (19) 66 (19) 64 (21)
EQ-5D-5L index
• Baseline 0.774 (0.159) 0.736 (0.191) 0.725 (0.201)
• 4 months 0.779 (0.180) 0.765 (0.225) 0.747 (0.219)
• 1 year 0.774 (0.194) 0.767 (0.195) 0.696 (0.247)
Self-efficacy for pain (5–25)
• Baseline 18 (4.6) 17 (4.0) 16 (5.1)
• 4 months 18 (4.8) 18 (4.8) 17 (4.8)
• 1 year 19 (5.2) 18 (5.1) 16 (4.1)
Self-efficacy for symptoms (6–30)
• Baseline 24 (3.6) 23 (4.1) 22 (4.0)
• 4 months 23 (5.5) 23 (5.1) 22 (5.5)
• 1 year 23 (4.5) 22 (5.5) 20 (5.5)
Quadriceps strength involved knee (Nm)
• Baseline 122 (51) 116 (40) 108 (42)
• 4 months 133 (53) 129 (42) 108 (46)
• 1 year 125 (46) 129 (36) 112 (48)
Quadriceps strength involved knee (Nm/kg)
• Baseline 1.4 (0.5) 1.3 (0.3) 1.3 (0.4)
• 4 months 1.6 (0.5) 1.5 (0.3) 1.2 (0.4)
• 1 year 1.6 (0.5) 1.5 (0.3) 1.3 (0.5)
Hamstrings strength involved knee (Nm)
• Baseline 68 (33) 63 (27) 62 (28)
• 4 months 72 (36) 72 (25) 63 (27)
• 1 year 67 (29) 71 (23) 64 (29)
Hamstrings strength involved (Nm/kg)
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 8 of 12

Table 2 (continued)
Outcomes Strength exercise Aerobic exercise Usual Care
• Baseline 0.8 (0.3) 0.7 (0.3) 0.7 (0.3)
• 4 months 0.8 (0.3) 0.8 (0.2) 0.7 (0.3)
• 1 year 0.7 (0.3) 0.8 (0.2) 0.8 (0.3)
Maximal oxygen consumption (VO2max)
• Baseline 28.2 (7.3) 27.4 (4.9) 29.4 (6.5)
• 4 months 29.1 (7.5) 30.1 (5.9) 28.0 (6.6)
• 1 year 29.6 (8.4) 26.9 (4.1) 27.4 (6.7)
Mean values (standard deviation) are presented. KOOS; Knee injury and osteoarthritis outcomes core, QoL, knee-related quality of life; ADL, activities of daily living;
EQ-5D-5 L, EuroQoL-5 Dimensions 5 Levels, VAS, visual analogue scale: 0 best, 10 worst; Nm, newton meter; kg, kilograms (body weight); VO2max was reported in
mL*kg*min− 1. Numbers at 4 months for KOOS QoL: strength training (n = 49), stationary cycling (n = 42), usual care (n = 45). Numbers At 1 year for KOOS QoL: strength
training (n = 45), stationary cycling (n = 39), usual care (n = 41)

0.26 Nm/BW), and better VO2max (3.0, 95% CI 1.9 to 4.1 better results at the 4-month follow-up for quadriceps
mL·kg− 1·min− 1) at the 4-month follow-up as compared strength and VO2max. This indicate that the participants
to the usual care group (Table 3). responded physiologically to the exercise programs.
The GRC showed that 65% (at 4 months) and 39% (at Our results are most likely influenced by the smaller
1 year) in the strength exercise group reported a little bit sample size than what we intended to include, but the
better, much better, or fully recovered knee complaints. wide CIs might also reflect that the participants have
In the aerobic exercise group, the corresponding propor- conducted and responded individually to the interven-
tions were 62% (at 4 months) and 42% (at 1 year), and in tions. Future studies should investigate more thoroughly
the usual care group 38% (at 4 months) and 31.5% (at 1 who is responding to exercise interventions and prob-
year) (Table 4). able explanations behind clinically important improve-
ments or differences. Furthermore, the lack of efficacy at
Adverse events and trial amendments 1 year might relate to adherence to the exercises both in
There were no adverse events during testing or the exer- the intervention period, but also in the post-intervention
cise programs except one participant who experienced period. The interventions might not have been optimal
high blood pressure and EKG irregularities during the for a heterogenous group of patients with knee osteo-
baseline VO2max cycle test. This was inspected by a cardi- arthritis with some having previous injury, some were
ologist. The participant withdrew the informed consent. overweight, and they had different experience with mod-
Another participant experienced swelling during the erate intensity training. The physiotherapists were told to
cycling, withdrew, and received total knee replacement. tailor the intervention within the frame of the program,
but we have no information on how this was conducted
Discussion other than information from the training diary. We did
In this RCT we aimed to investigate the efficacy of not collect data on information the physiotherapists gave
strength exercise or aerobic exercise compared to usual to the participants during and at the end of the interven-
care on KOOS QoL at 1 year in individuals with symp- tion, but we assessed health care utilization at 6 months,
tomatic knee osteoarthritis. Our secondary outcomes 9 months, and 12 months (physiotherapy consultations
included several functional and physiological outcomes presented in Table 1). A systematic review with meta-
at 4 months and 1 year. We did not detect statistically sig- analysis including data from 77 RCTs of hip and knee
nificant between-group differences for knee-related QoL osteoarthritis patients confirmed exercise benefits for
at 1 year, but our results had wide CIs. A recent system- pain, function, QoL, and performance tests at 8 weeks,
atic review evaluating the effects of exercise for patients but with a gradually decreasing effect over time to no
with knee osteoarthritis [41] summarized results from better than usual care at around 9 months [9]. The meta-
four studies using KOOS QoL as an outcome measure. In analysis included all types of exercise which preclude
contrast to our study, they did not include 1-year results, direct comparison to our study with strength or cycling
but the short-term post-intervention results detected interventions. Our exercise groups showed around 10%
an improvement in QoL. Our study shows inconclusive better quadriceps muscle strength and VO2max at the
results due to the wide CIs, which means that the popula- 4-month follow-up compared to usual care, but no fur-
tion estimate could be a clinically important effect, or it ther improvement was seen at the 1-year follow-up. This
could be no effect. The same were seen for most of the could indicate that the participants did not continue the
secondary outcomes. However, the secondary outcome structured quadriceps strength exercises and aerobic
analyses showed that the strength exercise group and exercise beyond the intervention period.
the aerobic exercise group had statistically significant
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 9 of 12

Table 3 Mean difference between the intervention groups and the usual care group (UC) (95% CI).
Outcomes Strength exercise vs. UC Aerobic exercise
vs. UC
KOOS QoL (0-100)
• 4 months 6.7 (-0.5 to 14.0) 4.7 (-2.8 to 12.2)
• 1 year 6.5 (-0.9 to 14.0) 5.0 (-2.7 to 12.8)
KOOS Pain (0-100)
• 4 months 2.4 (-4.6 to 9.4) 3.2 (-3.9 to 10.4)
• 1 year 5.7 (-1.4 to 12.9) 3.2 (-4.2 to 10.6)
KOOS Symptoms (0-100)
• 4 months 6.3 (-0.5 to 13.2) 7.4 (0.4 to 14.4)*
• 1 year 6.2 (-0.9 to 13.3) 5.1 (-2.1 to 12.4)
KOOS ADL (0-100)
• 4 months 2.3 (-4.0 to 9.5) 3.5 (-3.9 to 10.9)
• 1 year 7.3 (-0.1 to 14.6) 5.8 (-1.8 to 13.4)
KOOS Sport (0-100)
• 4 months 8.4 (-0.1 to 16.9) 2.1 (-6.7 to 10.9)
• 1 year 2.4 (-6.4 to 11.2) 0.5 (-8.6 to 9.5)
Pain last week (0–10)
• 4 months -0.4 (-1.2 to 0.4) -0.3 (-1.1 to 0.6)
• 1 year -0.9 (-1.7 to -0.02) -0.9 (-1.8 to 0.03)
EQ-5D-VAS (0-100)
• 4 months 6.4 (-0.8 to 13.6) 6.4 (-1.1 to 13.9)
• 1 year 2.5 (-5.0 to 9.9) 5.4 (-2.4 to 13.2)
EQ-5D-5L index
• 4 months 0.00 (-0.07 to 0.07) 0.01 (-0.06 to 0.08)
• 1 year 0.02 (-0.05 to 0.1) 0.06 (-0.02 to 0.13)
Self-efficacy for pain (5–25)
• 4 months 0.8 (-1.03 to 2.7) 1.0 (-1.0 to 2.9)
• 1 year 2.1 (0.2 to 4.0)* 1.7 (-0.3 to 3.7)
Self-efficacy for symptoms (6–30)
• 4 months 0.7 (-1.4 to 2.7) 0.6 (-1.5 to 2.7)
• 1 year 2.0 (-0.2 to 4.1) 2.0 (-0.12 to 4.2)
Quadriceps strength (Nm)
• 4 months 14.3 (5.3 to 23.3)* 10.1 (0.9 to 19.3)*
• 1 year -1.1 (-11.4 to 7.2) 3.2 (-7.7 to 14.1)
Quadriceps strength (Nm/kg)
• 4 months 0.21 (0.11 to 0.31)* 0.15 (0.05 to
0.26)*
• 1 year 0.04 (-0.08 to 0.16) 0.04 (-0.08 to 0.16)
Hamstrings strength (Nm)
• 4 months 3.3 (-2.3 to 9.0) 3.9 (-1.9 to 9.7)
• 1 year -1.6 (-8.0 to 4.9) 2.4 (-4.4 to 9.3)
Hamstrings strength (Nm/kg)
• 4 months 0.05 (-0.02 to 0.12) 0.07 (-0.00 to 0.14)
• 1 year 0.00 (-0.08 to 0.9) 0.04 (-0.05 to 0.12)
VO2max (mL·kg− 1·min− 1)
• 4 months 1.9 (0.85 to 3.0)* 3.01 (1.9 to 4.1)*
• 1 year 0.85 (-0.5 to 2.2) 0.24 (-1.1 to 1.6)
KOOS; Knee injury and osteoarthritis outcomes core, QoL, knee-related quality of life; ADL, activities of daily living; EQ-5D-5 L, EuroQoL-5 Dimensions 5 Levels,
VAS, visual analogue scale: 0 best, 10 worst; Nm, newton meter for involved knee; kg, kilograms (body weight); VO2peak, voluntary maximal oxygen consumption.
Numbers at 3 months and 1 year varies for the three groups for the different vary. Mixed linear models are adjusted for baseline value of the outcome. *p < 0.05
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 10 of 12

Fig. 2 Knee injury and osteoarthritis outcome score (KOOS) knee-related quality of life (QoL) at three time points

Table 4 Global rating of change in knee complaints

Categories Strength exercise Aerobic exercise Usual care
4 months (n = 46) 1 year (n = 44) 4 months (n = 42) 1 year (n = 38) 4 months (n = 45) 1 year (n = 38)
Completely recovered 1 (2) 0 1 (2) 0 0 2 (5)
Much better 12 (26) 8 (18) 6 (14.5) 6 (16) 5 (11) 7 (18.5)
A little bit better 17 (37) 9 (21) 19 (45.5) 10 (26) 12 (27) 3 (8)
No change 9 (19.5) 16 (36) 7 (17) 11 (29) 20 (44) 13 (34)
A little bit worse 3 (6.5) 8 (18) 7 (17) 9 (24) 5 (11) 7 (18.5)
Much worse 4 (9) 3 (7) 1 (2) 2 (5) 1 (2.5) 4 (11)
Worse than ever 0 0 1 (2) 0 2 (4.5) 2 (5)
Numbers (%) are presented for change in knee complaints from the previous follow-up, between baseline and 4 months and between 4 months and 1 year

Limitations and strengths because this score has been found to have a greater respon-
The main limitation of this study is that we did not con- siveness in younger populations compared to other instru-
duct a feasibility and pilot study before we initiated the ments such as WOMAC and the SF-36 [13]. We have no
trial, thus we had not estimated recruitment rate before- data on potential eligible patients seeking health care for
hand. The main reasons for the slow recruitment rate were knee osteoarthritis in the institutions and hospitals over
lack of time to recruit in the primary health care and that the recruitment period. We have not recorded reasons
only one person recruited participants. Multi-center stud- for ineligibility because we received patients from clini-
ies and adequate resources used on the recruitment pro- cal practice and advertisements. The physical activity level
cess should be emphasized to successfully conduct similar data indicated that the participants were active before the
RCTs. We decided to terminate the trial early and chose intervention. However, their main activity was walking, and
to do that when the COVID-19 pandemic hit Norway in participants who reported structured strength training or
March 2020. The estimates for our outcomes had wide CIs, cycling regularly were not included. Our experiences from
and with a larger data set, the results may have been differ- the recruitment conversations were that the participants
ent. The clinically important difference for the KOOS QoL generally overreported their physical activity level, particu-
with 10 points might not be correct as it has been suggested larly the intensity. Low physical activity was also confirmed
to be 8 points for “somewhat better” and 15.6 points for a by the fact that most participants expressed that the cycling
great deal better [42], but these limits are not well estab- VO2max test was unexpectable heavy. We excluded partici-
lished. Furthermore, KOOS QoL might be a challenging pants with a BMI > 35 because we believe an intervention
outcome for measuring efficacy after exercises (for instance for that group needs to include a weight reduction interven-
by influencing pain and function more directly than QoL), tion in addition to exercise. We also excluded eligible partic-
even though it has been more sensitive for changes in this ipants above 70 years because we aimed to target a younger
population. The KOOS QoL was used as primary outcome osteoarthritis population. The exercise interventions lasted
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 11 of 12

around 90 to 180 min per week, which may have been too Supplementary Information
little to improve KOOS QoL and knee function; however, to The online version contains supplementary material available at https://doi.
org/10.1186/s12891-023-06831-x.
the best of our knowledge, at the time when we designed the
study, no studies had documented the best dosage beyond Appendix 1
the recommendations given by WHO guidelines of 150 min
Appendix 2
of exercise per week. Juhl et al. reported in 2014 [6] that an
Appendix 3
optimal exercise program for improving pain and disability
in knee osteoarthritis should be supervised and conducted
three times per week. However, the review did not include Acknowledgements
We are grateful to research coordinator at the Norwegian Sports Medicine
KOOS QoL. Clinic (Nimi) Kristin Bølstad and PT Marte Lund for all help with administration
The strength of this study is that we have included rela- of the trial, testing and participant flow throughout the project period. Thanks
tively young study participants from primary and sec- to all the collaborating physiotherapy institutes, Department of Radiology
at Oslo University Hospital by Kristin Dehli, Kathrine Lamark, and radiologist
ondary care and conducted the study in clinical practice. Ragnhild Gunderson, Oslo University Hospital, and to Lovisenberg Hospital
We analyzed data with a mixed-effect linear model that and Akershus University Hospital for recruiting participants. We would also like
includes participants regardless of one missing follow-up, to thank all the study participants.
and includes both fixed and random factors. This study was Authors’ contributions
designed at a time point when user involvement was not a Britt Elin Øiestad, Margreth Grotle, and May Arna Risberg designed the trial
mandatory part of trials, however, the strength exercise pro- and wrote the funding application. Britt Elin Øiestad and May Arna Risberg
developed the interventions and testing protocols. Britt Elin Øiestad recruited
gram was developed collaboration with specialists in sport and tested participants, conducted statistical analyses, and was first author
physical therapy with high competence in treating patients of the paper. Asbjørn Årøen, Jan Harald Røtterud and Nina Østerås recruited
with knee osteoarthritis. Furthermore, according to the sys- participants, and contributed with interpretation of the data. Even Jarstad
planned the VO2max exercise and testing protocol and tested all participants.
tematic review by Luan et al. [15] few studies have evaluated All authors critically reviewed the data analyses, revised, and approved
the effect of aerobic exercise, such as cycling exercise on the final manuscript. All authors agree to be accountable for all aspects
QoL in patients with knee osteoarthritis. To our knowledge, of the work in ensuring accuracy or integrity of any part of the work are
appropriately investigated and resolved.
none have tested direct maximal oxygen consumption.
Funding
Clinical implications The Research Council of Norway and the Norwegian Physiotherapy Fund.
This study showed no effect of exercise compared to Data availability
usual care on knee-related QoL at 1 year. Secondary out- The dataset for replication analyses of this paper may be available in an
comes showed that strength and aerobic exercise pro- anonymous format according to GDPR. Please contact the corresponding
author for requests.
grams improved physiological measures at 4 months,
but not at 1 year. Importantly, cycling 2–3 times a week
Declarations
at moderate intensity showed a 10% improved VO2max
which could be important for general health and life- Competing interests
style diseases in this population. Longer follow-up with The authors declare no competing interests.
supervision from physiotherapists might be needed to Ethics approval and consent to participate
maintain the short-term, post-intervention physiological All participants signed an informed consent prior to the baseline assessment.
effects. Future studies might be more personalized add- The Regional Ethical Committee in the Health Region South-East in Norway
approved the study protocol (REK 10/223), and the Data Inspectorate at
ing behavioral change and self-management strategies to Oslo University Hospital approved the study. The research was conducted in
people who struggle with exercise adherence, and future accordance with the Declarations of Helsinki.
studies should investigate which patients respond to
Consent for publication
exercise interventions. Not applicable.

Conclusion Received: 24 April 2023 / Accepted: 25 August 2023

This trial found no statistically significant effects of two
exercise programs compared to usual care on KOOS
QoL at 1 year in individuals with knee osteoarthritis. An
underpowered sample size may explain lack of efficacy
between the intervention and the usual care groups and References
1. Long H, Liu Q, Yin H, Wang K, Diao N, Zhang Y, Lin J, Guo A. Prevalence trends
imprecise results. Secondary outcomes showed improved of site-specific osteoarthritis from 1990 to 2019: findings from the Global
quadriceps strength and VO2max at 4 months in both Burden of Disease Study 2019. Arthritis & rheumatology (Hoboken, NJ) 2022.
groups. 2. Hunter DJ, Bierma-Zeinstra S. Osteoarthritis. Lancet.
2019;393(10182):1745–59.
3. Kolasinski SL, Neogi T, Hochberg MC, Oatis C, Guyatt G, Block J, Callahan
L, Copenhaver C, Dodge C, Felson D, et al. 2019 American College of
Øiestad et al. BMC Musculoskeletal Disorders (2023) 24:714 Page 12 of 12

Rheumatology/Arthritis Foundation Guideline for the management of 23. Roos EM, Roos HP, Ekdahl C, Lohmander LS. Knee injury and osteoarthritis
Osteoarthritis of the Hand, hip, and knee. Arthritis & Rheumatology (Hoboken outcome score (KOOS)--validation of a swedish version. Scand J Med Sci
NJ). 2020;72(2):220–33. Sports. 1998;8(6):439–48.
4. Whittaker JL, Truong LK, Dhiman K, Beck C. Osteoarthritis year in review 2020: 24. Herdman M, Gudex C, Lloyd A, Janssen MF, Kind P, Parkin D, Bonsel G, Badia X.
rehabilitation and outcomes. Osteoarthr Cartil. 2021;29(2):190–207. Development and preliminary testing of the new five-level version of EQ-5D
5. Fransen M, McConnell S, Harmer AR, Van der Esch M, Simic M, Bennell KL. (EQ-5D-5L). Qual Life Res. 2011;20(10):1727–36.
Exercise for osteoarthritis of the knee: a Cochrane systematic review. Br J 25. Garratt AM, Hansen TM, Augestad LA, Rand K, Stavem K. Norwegian popula-
Sports Med. 2015;49(24):1554–7. tion norms for the EQ-5D-5L: results from a general population survey. Qual
6. Juhl C, Christensen R, Roos EM, Zhang W, Lund H. Impact of exercise type and Life Res. 2022;31(2):517–26.
dose on pain and disability in knee osteoarthritis: a systematic review and 26. Lorig K, Chastain RL, Ung E, Shoor S, Holman HR. Development and evalu-
meta-regression analysis of randomized controlled trials. Arthritis & Rheuma- ation of a scale to measure perceived self-efficacy in people with arthritis.
tology (Hoboken NJ). 2014;66(3):622–36. Arthritis Rheum. 1989;32(1):37–44.
7. Dean BJF, Collins J, Thurley N, Rombach I, Bennell K. Exercise therapy with or 27. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascer-
without other physical therapy interventions versus placebo interventions for taining the minimal clinically important difference. Control Clin Trials.
osteoarthritis –systematic review. Osteoarthr Cartil Open. 2021;3(3):100195. 1989;10(4):407–15.
8. Verhagen AP, Ferreira M, Reijneveld-van de Vendel EAE, Teirlinck CH, Runhaar 28. Foss O, Hallen J. The most economical cadence increases with increasing
J, van Middelkoop M, Hermsen L, de Groot IB, Bierma-Zeinstra SMA. Do we workload. Eur J Appl Physiol. 2004;92(4–5):443–51.
need another trial on exercise in patients with knee osteoarthritis?: no new 29. Taylor HL, Buskirk E, Henschel A. Maximal oxygen intake as an objective
trials on exercise in knee OA. Osteoarthr Cartil. 2019;27(9):1266–9. measure of cardio-respiratory performance. J Appl Physiol. 1955;8(1):73–80.
9. Goh SL, Persson MSM, Stocks J, Hou Y, Lin J, Hall MC, Doherty M, Zhang W. 30. Edvardsen E, Hem E, Anderssen SA. End criteria for reaching maximal oxygen
Efficacy and potential determinants of exercise therapy in knee and hip uptake must be strict and adjusted to sex and age: a cross-sectional study.
osteoarthritis: a systematic review and meta-analysis. Annals of Physical and PLoS ONE. 2014;9(1):e85276.
Rehabilitation Medicine. 2019;62(5):356–65. 31. Goedecke JH, Gibson ASC, Grobler L, Collins M, Noakes TD, Lambert EV.
10. Skou ST, Roos EM. Good life with osteoArthritis in Denmark (GLA:D™): Determinants of the variability in respiratory exchange ratio at rest and dur-
evidence-based education and supervised neuromuscular exercise delivered ing exercise in trained athletes. Am J Physiology-Endocrinology Metabolism.
by certified physiotherapists nationwide. BMC Musculoskelet Disord. 2000;279(6):E1325–34.
2017;18(1):72. 32. Esfarjani F, Laursen PB. Manipulating high-intensity interval training: effects
11. Holm I, Pripp AH, Risberg MA. The active with OsteoArthritis (AktivA) physio- on VO2max, the lactate threshold and 3000 m running performance in
therapy implementation model: a patient education, supervised Exercise and moderately trained males. J Sci Med Sport. 2007;10(1):27–35.
Self-Management Program for patients with mild to moderate osteoarthritis 33. Tonnessen E, Sylta O, Haugen TA, Hem E, Svendsen IS, Seiler S. The road to
of the knee or hip Joint. A National Register Study with a two-year Follow-Up. gold: training and peaking characteristics in the year prior to a gold medal
J Clin Med 2020, 9(10). endurance performance. PLoS ONE. 2014;9(7):e101796.
12. Patterson BE, Girdwood MA, West TJ, Bruder AM, Oiestad BE, Juhl C, Culvenor 34. Sandvei M, Jeppesen PB, Stoen L, Litleskare S, Johansen E, Stensrud T,
AG. Muscle strength and osteoarthritis of the knee: a systematic review and Enoksen E, Hautala A, Martinmaki K, Kinnunen H, et al. Sprint interval running
meta-analysis of longitudinal studies. Skeletal Radiol 2022. increases insulin sensitivity in young healthy subjects. Arch Physiol Biochem.
13. Roos EM, Lohmander LS. The knee injury and osteoarthritis outcome score 2012;118(3):139–47.
(KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 35. Ingjer F. Factors influencing assessment of maximal heart rate. Scand J Med
2003;1:64. Sci Sports. 1991;1(3):134–40.
14. Smith TO, Hawker GA, Hunter DJ, March LM, Boers M, Shea BJ, Christensen 36. Slade SC, Dionne CE, Underwood M, Buchbinder R, Beck B, Bennell K,
R, Guillemin F, Terwee CB, Williamson PR, et al. The OMERACT-OARSI core Brosseau L, Costa L, Cramp F, Cup E, et al. Consensus on Exercise Reporting
domain set for measurement in clinical trials of hip and/or knee osteoarthri- Template (CERT): modified Delphi Study. Phys Ther. 2016;96(10):1514–24.
tis. J Rhuematol. 2019;46(8):981–9. 37. Garber CE, Blissmer B, Deschenes MR, Franklin BA, Lamonte MJ, Lee IM,
15. Luan L, Bousie J, Pranata A, Adams R, Han J. Stationary cycling exercise for Nieman DC, Swain DP. American College of Sports M: American College of
knee osteoarthritis: a systematic review and meta-analysis. Clin Rehabil. Sports Medicine position stand. Quantity and quality of exercise for develop-
2021;35(4):522–33. ing and maintaining cardiorespiratory, musculoskeletal, and neuromotor
16. Juszczak E, Altman DG, Hopewell S, Schulz K. Reporting of Multi-Arm parallel- fitness in apparently healthy adults: guidance for prescribing exercise. Med
group randomized trials: extension of the CONSORT 2010 Statement. JAMA. Sci Sports Exerc. 2011;43(7):1334–59.
2019;321(16):1610–20. 38. Stensrud S, Roos EM, Risberg MA. A 12-week exercise therapy program in
17. Mansournia MA, Collins GS, Nielsen RO, Nazemipour M, Jewell NP, Altman middle-aged patients with degenerative meniscus tears: a case series with
DG, Campbell MJ. A CHecklist for statistical Assessment of Medical Papers 1-year follow-up. J Orthop Sports Phys Ther. 2012;42(11):919–31.
(the CHAMP statement): explanation and elaboration. Br J Sports Med. 39. Eitzen I, Moksnes H, Snyder-Mackler L, Risberg MA. A progressive 5-week
2021;55(18):1009–17. exercise therapy program leads to significant improvement in knee func-
18. Oiestad BE, Osteras N, Frobell R, Grotle M, Brogger H, Risberg MA. Efficacy of tion early after anterior cruciate ligament injury. J Orthop Sports Phys Ther.
strength and aerobic exercise on patient-reported outcomes and structural 2010;40(11):705–21.
changes in patients with knee osteoarthritis: study protocol for a randomized 40. Nelson ME, Rejeski WJ, Blair SN, Duncan PW, Judge JO, King AC, Macera CA,
controlled trial. BMC Musculoskelet Disord. 2013;14:266. Castaneda-Sceppa C. Physical activity and Public Health in older adults:
19. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann recommendation from the American College of Sports Medicine and the
Rheum Dis. 1957;16(4):494–502. American Heart Association. Med Sci Sports Exerc 2007, 39(8).
20. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, Christy W, Cooke 41. Raposo F, Ramos M, Lucia Cruz A. Effects of exercise on knee osteoarthritis: a
TD, Greenwald R, Hochberg M, et al. Development of criteria for the classifica- systematic review. Musculoskelet Care. 2021;19(4):399–435.
tion and reporting of osteoarthritis. Classification of osteoarthritis of the knee. 42. Singh JA, Luo R, Landon GC, Suarez-Almazor M. Reliability and clinically
Diagnostic and therapeutic Criteria Committee of the american Rheumatism important improvement thresholds for osteoarthritis pain and function
Association. Arthritis Rheum. 1986;29(8):1039–49. scales: a multicenter study. J Rhuematol. 2014;41(3):509–15.
21. Kurtze N, Rangul V, Hustvedt BE. Reliability and validity of the international
physical activity questionnaire in the Nord-Trondelag health study (HUNT)
population of men. Bmc Med Res Methodol. 2008;8:63. Publisher’s Note
22. Hinman RS, May RL, Crossley KM. Is there an alternative to the full-leg Springer Nature remains neutral with regard to jurisdictional claims in
radiograph for determining knee joint alignment in osteoarthritis? Arthritis published maps and institutional affiliations.
Rheum. 2006;55(2):306–13.