Trial design
This single blind, randomized, and controlled trial, which was conducted at Razi University rehabilitation center in Kermanshah, Iran, lasted for eight weeks starting in February and ending in May, 2019. The assessors who measured variables for patients were blinded about the group allocation. This study included three steps: 1. Pre-intervention measurements 2. 8-week aquatic and TRX exercises for case groups, and the control group which just received drug regimens by a rheumatologist, and 3. Post-intervention measurements (Fig 1).
Participants
A total of 200 knee osteoarthritis’ female patients were studied for eligibility among which 36 individuals were included in the present study. Of the 36 participants, 12 patients were allocated to the aquatic exercise, twelve to TRX exercises, and the rest to the control group. However, 111 patients didn’t meet the eligibility criteria and 53 declined to participate. The participants of the present study consisted of individuals from two groups: individuals who had regular referral from rheumatologist, and patients who waited for common physiotherapy in rehabilitation centers.
The criteria for including participants in the study were as follows: (1) having the age of more than 40 years, (2) acquiring the American College of Rheumatology clinical criteria for knee OA (Altman et al., 1986), (3) meeting the Kellgren and Lawrence radiographic disease severity scale ≥ II, and (4) self-reporting knee instability. Despite the abovementioned conditions, the participants were excluded in case they (1) had strokes, (2) faced uncontrolled hypertension, (3) were unable to walk without assistant instruments, (4) had received other treatment interventions in the past three months, (5) were suffered obesity (BMI>40kg/m2), (6) suffered from neuromuscular diseases like multiple sclerosis or Parkinson, (7) had lower extremity fracture, (8) afflicted with concurrent hip osteoarthritis, (9) waited for arthroplasty, (10) or cardiovascular diseases [7, 21]. Further, the patients who had received injection in the 6-month-period adjacent to the study, or had had surgical procedures were excluded from the study. Although the patients included in this study had bilateral arthritis, only the knee in which they reported symptom of instability was assessed. It is worthy to add that all the participating patients received the same drug regimens including Meloxicam 7.5 milligram, Glucosamine Sulfate 750 milligram, and Calcium-D in daily manner.
The G. Power (Ver. 3.1, Heinrich Heine University, Düsseldorf, Germany) software was used to estimate the minimum sample size. According to the reported results of previous researches and based on the test power of 0.90, the effect size of 0.63 and the significance level of 0.05, the minimum sample size was determined to be 36 [22]. Finally, the participants were randomly distributed into three groups, including TRX (n=12), aquatic exercises (n=12), and control (n=12) (Table 2).
Study interventions
TRX exercises
The TRX ® Rip Trainer ™ (China) was used for performing the exercises. TRX training was conducted by a TRX specialist, while an assistant coach also collaborated to prevent the patients from doing wrong exercises. The safety points were checked by a trainer in every session to avoid injuries before the exercises started. The TRX straps hanging down from anchor point, and Suspension AnchorTM were adjustable to execute various exercises. Exercises were designed based on the patients’ motion limitations like knee flexion and extension. Furthermore, TRX exercises were started at their most convenient forms while they gradually, but progressively, turned to be more and more difficult. The difficulty level of exercises increased through a step by step process by 1) Narrowing the base of support which increased the difficulty by reducing the stability 2) Changing the angle of pull, in other words, the farther the person was from the vertical, the greater the resistance3) employing the pendulum in ground exercises in which the feet were placed in the Suspension Trainer and the hands were off the ground (head or back was on the ground). The gravity center in relationship to the perpendicular gravitational pull determined the difficulty. 4) and adding a handle which can increase the difficulty level of exercises. Each session time length was divided as: 5-10 min of each was allocated to introducing the sessions’ regarded exercises and their related correct techniques, 5-10 min were devoted to warm-up, which mainly included stretching exercises, and the remaining 40-50 min were regarded for performing the TRX exercises. Participants who were recognized with wrist pain through conducting the planks’ tests were allowed to put their forearm on the ground in order to prevent the wrist pain increasing. The TRX exercises protocol lasted for 60 minutes each session and was performed three times a week for eight weeks, 24 sessions altogether. The majority of the exercises were focused on the core muscles, hip abductors, and leg muscles strengthening (Table 1) [23].
Table 1. insert here
Aquatic exercises
Aquatic exercise intervention was performed for eight weeks, three times a week, 24 sessions in total, with each session lasting for exactly 90 minutes. In other words, each participant was required to attend 24 sessions of rehabilitation with 90 minutes of duration for each session during the conducting phase of the study. The water temperature was approximately 32 ° C (89° F), and the minimum water depth was considered1.3 meter. The water based exercises protocol included: 10 minutes of warm-up along with walking (forward, backward, and sidewalk), and also stretching exercises for lower extremity muscles (quadriceps, hamstrings, triceps surae, abductors and adductors of hip, and gluteal muscles), 20-minute strength exercises with elastic band and sandbag (gluteus, adductors and abductors of hip, quadriceps, hamstrings, and triceps surae muscles); 20 minutes of aerobic exercises (stationary running or deep water-running); 20 minutes of step training and proprioceptive exercises; 10 minutes of core exercises, and finally 10 minutes of cool down. Based on the previous study findings, we selected exercises of the current study with the purpose of function, pain, and balance improvements [24, 25]. The aquatic exercises protocol was supervised by a certified physiotherapist in the pool (table 2).
Table 2. insert here
Randomization and Blinding
Participants were randomized by the use of Random Number Generator Software (Research Randomizer, version 3.0), and also were allocated to three groups using Sequentially Numbered Opaque Sealed Envelopes (SNOSE) concealed allocation method. A physiotherapist who did not involve in the data collection and evaluation of the outcomes did the random allocation sequence, and enrolled as well as assigned participants to groups by random (allocation ratio 1:1:1).
The assessors of this research were blinded about the exercises and interventions assigned to the groups, but there was no possible way for blinding the subjects to training as well as the statistician towards the groups and their assigned exercises.
Study outcomes
The current study was planned based on authors’ previous study and literature results [2,6,7]. In this line, the effect of 8-week aquatic and TRX exercises on factors that could be affected by KI including WOMAC stiffness subscale, balance, pain, KI, quadriceps strength and knee flexion ROM were evaluated.
Knee pain
Knee pain intensity was measured by employing a 10-cm visual analog scale (VAS) with the scoring range of 0 to10cm, in which “0” represented absence of any pain, “1” minimal pain, and “10” extreme or intolerable pain. To assess the intensity of knee pain of the participants, they were asked the following question: “how much pain do you have during your daily activities?” The VAS was used to measure the intensity of participants’ subjective pain prior to and after the interventions. A good reliability and validity has been reported for the VAS (ICC=0.92) [26].
WOMAC stiffness
WOMAC is a reliable and valid instrument in the literature (ICC:0.80). The stiffness subscale consisted of two items based on the five-point Likert scale of 0 indicating no symptoms to 4 indicating extreme symptoms. The maximum score based on the scale was 8, (two four- point items) and the total range was from 0 to 8, with the zero indicating no symptom and 8 indicating the worst symptoms [27].
Berg Balance Scale (BBS)
Berg Balance Scale, which consisted of 14 different tasks, was used to assess balance in sitting and standing position and in transfer. Each motor task was rated by the use of a 5-point scale ranging from 0 to 4. The total score ranges from 0 to 56, where 56 represents normal balance. The test-retest reliability for the BBS was reported to be excellent (ICC= .71 to .99) [28].
Knee instability
Self-reported knee instability was evaluated according to giving way, and also shifting evidence, during the last month by Felson’s questionnaire [29]. Knee instability severity was graded based on the numerical scale (0 to 5) in response to the following question. The question was “What degree of giving way, buckling, or shifting of the knee would affect your daily routine activity?” The ratings were as follows: 5 = “I have no symptom”, 4 = “I have symptom, but it does not affect my ADL”, 3 = “Symptoms affect my ADL slightly”, 2 = “symptoms affect my ADL moderately”, 1 = “symptoms affect my ADL strongly”, 0 = “symptoms prevent me to perform all my everyday activities” [25]. The test-retest reliability of this self-report rating of KI was estimated by the use of an intra-class correlation coefficient (ICC =0.72) [30].
Knee Flexion ROM
The Bubble inclinometer device (Fabrication Enterprises, Inc, White Plains, NY, USA) was used to measure Knee flexion ROM. The subjects were placed in prone position. Then the inclinometer was placed at the back of the tibia. The test was conducted on the limb which was more affected. Knee flexion was stopped in end-range of passive motion and further movement was restricted by pain. Three trials were recorded and the average of the three values was used for analysis [31].
Knee extensors strength
Maximal isometric strength of the knee extensors (quadriceps muscle) was measured using the Baseline Pull-Push Dynamometer (Model 12-0343, Fabrication Enterprises Inc., NY, USA). This digital dynamometer measures the force up to a maximum of 199.9 kg. Measurements were performed at 80˚- 90˚ of knee flexion. The instrument was calibrated according to the instructions, before any measurement. The patients were seated in a comfortable position with the backrest angled at 100˚. The shin pad was placed 2 cm above the medial and lateral malleoli. The instrument shaft remained horizontal to the anterior aspect of the mid shaft of tibia and horizontal to the posterior aspect over the musculotendinous junction of calf muscles. Subjects were then asked to remain at that position while pushing against the dynamometer. Also, the subjects were required to push against the gauge pad as hard as possible when given the appropriate command. All measurements were performed with the limb segment in a position that was with gravity eliminated. Resting times between trials were approximately 60 seconds. Each contraction was held for six seconds. The peak force was recorded and average of records was considered as the quadriceps strength [32].
Ethical Considerations
This study's protocols were reviewed and approved by the research ethics committee of the Medical Sciences University of Kermanshah in Iran (Registration no.: IR.UMMS.REC.1397.718). The study’s protocol was also registered in the Iranian Registry of Clinical Trials (Registration no.: IRCT20181222042070N1). All the tests and measurements were carried out at the Sport Rehabilitation Laboratory of Razi University, Iran. As well, all the participants were provided with related informed consent forms which were both completed and signed by participants in person.
Statistical analysis
We analyzed balance, Pain, WOMAC (stiffness), knee flexion ROM, quadriceps strength, and self-reported knee instability variables before and after the 8-week aquatic and TRX exercises.
First, we used Shapiro-wilks and Leven’s test for assessing the normal distribution of data, and also the variances homogeneity. When variances normality and homogeneity tests were confirmed, the data were considered to be parametric. Consequently, demographic and baseline parameters were analyzed by the one-way analysis of variance (ANOVA). Additionally, Tukey’s post hoc test was used for the pairwise comparisons. In order to compare the changes of each dependent variable over the time (t0=pretest, and t1=posttest) and between groups, the variables were analyzed by employing mixed-model repeated measures ANOVA using time and group as factors time ⁎ group (2⁎3). In the presence of significance, Tukey’s post hoc test pairwise comparisons were used. Also, we used Paired samples t-test for pretest to posttest assessing in each group. Statistical analysis was performed using statistical software, SPSS version 20.0 (IBM SPSS, Armonk, NY, USA). Statistical significance was determined at p-values less than 0.05. All results were reported as the mean ± standard deviation.
Fig 1. Insert here