Effects of backward walking exercise using lower body positive pressure treadmill on knee symptoms and physical function in individuals with knee osteoarthritis: a protocol for RCT

Demographic information pertaining to patient’s age, sex, and education level will be collected. Clinical variables include OA diagnosis type and severity, having other chronic conditions (e.g., diabetes mellitus and high blood pressure), prescribed medications, history of falls during the last 12 months, average pain level in the past month, body measures [height, weight, and body mass index (BMI)], and physical activity (minutes per week) will be obtained.

The 11-point NPRS measures pain severity from 0 to 10, where 0 indicates "no pain" and 10 indicates "worst imaginable pain." The NPRS has excellent internal consistency (Coefficient alpha = 0.84) [19] and has been validated to evaluate pain severity [20]. Minimal clinically important difference (MCID) in individuals with osteoarthritis is from 1 to 1.7 points or a change of 15–27.9% [21, 22]. At baseline, participants will rate their pain over the past 24 h using the NPRS. The NPRS will be taken before each session, throughout gait training, and after six weeks.

The KOOS assesses knee pain, symptoms, and level of disability during the last week [23]. It has five subscales: Pain; Symptoms; Function in ADLs; Sports and Recreation; and Quality of Life. Each subscale has a normalized score from 0 to 100, where 0 represents extreme symptoms and 100 represents no symptoms. The test is valid, reliable, and responsive to change for people with knee OA [24]. The minimal detectable change (MDC) is from 8 to 10 points [25]. A validated Arabic version of KOOS will be utilized for this study [26].

An isokinetic dynamometer (Biodex Model 4Pro, N.Y. 11,967, U.S.A) will be used to assess isometric knee flexion and extension strength of both sides [27]. The hip angle will be at 85° of hip flexion and the knee angle will be at 60° of knee flexion [28]. Both legs will be tested, and the trial order will be randomized. Patients will be instructed to exert as much effort as they could for five seconds during each test. We will explain the test procedure to the participants and provide 5 min warm-up cycling before the test. To avoid fatigue, 60-s rest time between trials will be provided. The testing session will be directed by the same investigator, who will verbally encourage each participant to perform their best. The peak torque obtained over the 3 repetitions will be recorded. Peak torque measurement is considered the most valid method to assess the muscle strength in individuals with knee OA [29].

The FTSST is an indicator of postural control, lower limb muscle strength, and balance [30]. Participants will stand up from a 45-cm chair with their arms over their chest and sit down five times as fast as they can without contacting the chair back. Start a timer at "go" and stop it when the participant's buttocks touch the chair on the fifth repeat. Participants can practice one trial before data collection. The FTSST has an excellent test–retest reliability (ICC = 0.960) in individuals with OA [31], and construct validity in people with rheumatoid arthritis [30].

The 3MBWT’s field begins and ends with marked colored tape. The participant will stand with their heels at the level of the front edge of the tape. Participants will be instructed to walk backwards 3 m when the investigator says “go” and stop when the participant passes the end-line tape. The participant can look back during the test. The test will only be administered once, and the time spend in completing the test will be recorded using a stopwatch. The investigator will walk behind the participant for safety. Participants will be advised to walk as fast as possible to the end line.

The TUG evaluates an individual’s functional mobility by asking them to stand up, walk toward a cone placed of 3 m from the chair at a normal speed, turnaround 180 degrees, walk back 3 m, and sit down [32]. The TUG has an excellent reliability and validity in people with knee OA [33]. The participants will be allowed to have one-practice trials. The third TUG trial will be recorded and documented.

The FSST is designed to assess dynamic balance and coordination. Participants will step forwards, sideways, and backwards as fast as they could over 4 canes that shaped plus sign without touching the canes. Start in square 1 facing square 2. Participants will follow the following order: square number 2, 3, 4, 1, 4, 3, 2, and 1. Participants can use walking aid during the test. Two trails will be performed, and the best trial will be documented. The trial will be repeated if the participant unsuccessfully completed the sequence. The time will be recoded when the first foot makes contact with square 2 and stopped when the last foot makes contact with square 1. The test has an excellent reliability and validity in individuals with hip OA [34].

Participants' stability and balance are evaluated using the FRT by measuring how far they are able to lean forward from a standing position. Participants extend their arm with a closed fist and leaning forward as possible without taking a step. On the yardstick, the beginning position will be marked at the 3rd metacarpal proximal edge. The highest achievable distance will be marked as ending position. The distance between the beginning and ending positions will be computed. One-practice trials will be permitted. The participant will complete two trials. The analysis will utilize the average of the two trials. The FRT has high intra-rater reliability (ICC = 0.98) and validity (r = 0.71) [35].

The 10-MWT evaluates walking speed over a brief period of time with or without the use of an assistive device. The participant will walk 10 m with an acceleration and deceleration zones for first and last 2 m. Only the 6 m distance in between will be used for scoring. The 10-MWT has excellent test–retest reliability with different populations [36, 37]. The 10-MWT has MCID of 0.13 m/s for significant meaningful change in older adult [37]. Participants will perform the test twice at comfortable speed and twice at fast speed, and the average of the results speeds will be utilized for the analysis separately.

6MWT measures aerobic endurance through distance walking during 6 min. Participants may take as many time as desired for standing rest; however, the investigator must continue timing. The number of standing rests and the duration of each standing will be recorded. Assistive device or braces are allowed and will be documented.

The PHQ-9 is a questionnaire for evaluating depression symptoms. It comprises nine items, each of which asks the participants to score the severity of their depression symptoms on a 4-point Likert scale over the last two weeks ranging from 0 “Not at all” to 3 “Nearly every day”. A total score of 5–9 indicates minimal depression symptoms, 10–14 indicates moderate symptoms, 15–19 indicates substantial symptoms, and score of 20 or higher indicates severe depression [39, 40]. This tool is reliability and validity in previous investigation [41]. This tool has been translated and validated into different languages including Arabic language [42].

The RAPA is a subjective survey that consists of nine questions that examines the physical activity of elderly. The physical activity is classified into three categories (light, moderate, and vigorous) along with textual and graphical descriptions [43]. Arabic language of RAPA measure has been translated and validated [44].

In a ratio of 1:1, the participants will be allocated randomly into an experimental group (EG, n = 13) or a control group (CG, n = 13), using an online randomization website (https://​www.​graphpad.​com/​quickcalcs/​randomize1.​cfm), a research assistant who is not involved in the intervention or data collection will generate the randomization.

The random assignment for each participant will be placed in an anonymous envelope. Envelopes will be prepared by an independent research assistant, who will then conceal information for the evaluator investigator. Following completing the baseline evaluation, the independent research assistant will draw an envelope and notify a trained physical therapist of group assignment for each participant.

This study is a single-blinded investigation. The group assignment will be concealed from the evaluator.

After signing the informed consent form, participants will fill out an intake form about demographic data and clinical variables. Additionally, they will be evaluated on inclusion/exclusion criteria. Participants meeting the inclusion criteria will be recruited in the study and assessed on the primary and secondary outcome measures. For test order, participants will be randomly assigned to one of four groups. A counterbalanced approach will be used to manage the order effects of learning and fatigue associated with the administered tests.

Participants will be randomized on the first day of the intervention to either a EG or CG. All participants will receive manual therapy and therapeutic exercise (Table 1) as described previously [12]. LBPP (AlterG Anti-Gravity Treadmill, Fremont, CA, USA) will be used for both forward and backward walking exercise (Fig. 3). All participants will have three sessions a week for 6 weeks (18 sessions) for up to 1 h/session. For the CG, gait training will be trained on forward walking using LBPP. The EG will be trained on backward walking using LBPP. For both groups, progression of gait training time will be instructed to gradually increase up to 30 min over the 6 weeks period. The air pressure inside chamber will be set to achieve pain free or max pain relief. Treadmill speed will be continuously assessed and increased gradually depending on the patient’s capacity. All participants will have 5 min of warm-up and cool-down exercise including heel raise, and hamstring and gastrocnemius-soleus stretches. All interventions and activities will be formed with a supervision of licensed physical therapists. Physical therapists will be trained on delivering the intervention and reporting the needed data each session. Before beginning the study, the evaluator will receive training on data collection and measuring all outcome measures.

G Power software analysis version 3.1.9.2 (University of Kiel, Kiel, Germany) was used to determine the sample size using pain level changes following intervention from a prior study [12]. With a small effect size of 0.29, α level of 0.05, and power of 0.80, the required sample size is 22 participants (11 participants per group). With a 20% attrition rate, there will be a total of 26 participants for total sample (13 participants in each group).