Here, based on a clinician’s point-of-view, we propose a two-model lower body positive pressure (LBPP) protocol (walking and squatting models) in addition to a clinical, functional assessment methodology, including details for further encouragement of the development of non-drug surgical intervention strategies in knee osteoarthritis patients. However, we only present the effect of LBPP training in improvement of pain and knee function in one patient through three-dimensional gait analysis. The exact, long-term effects of this approach should be explored in future studies.
Here, based on a clinician’s point-of-view, we propose a two-model lower body positive pressure (LBPP) protocol (walking and squatting models) in addition to a clinical, functional assessment methodology, including details for further encouragement of the development of non-drug surgical intervention strategies in knee osteoarthritis patients. However, we only present the effect of LBPP training in improvement of pain and knee function in one patient through three-dimensional gait analysis. The exact, long-term effects of this approach should be explored in future studies.
Knee osteoarthritis (OA) is a progressive degenerative joint condition and a major cause of pain and locomotor disability in people all over the world1. Knee OA is characterized by osteophyte and cyst formation, narrow joint spacing, and subchondral bone sclerosis2. These pathological changes make it difficult to perform essential activities of daily living such as walking, squatting, and going up and down stairs3. However, physical activity is recommended as an essential component of first-line knee OA management4. Exercise intervention for knee OA rehabilitation is affected by several factors: (1) limited knee joint movement caused by pain and minor knee structural changes; (2) muscle atrophy associated with maintaining knee stability and a decrease in muscle strength5; and (3) the above reasons lead to a reduction in exercise and an increase in body mass index (BMI), which further increases the burden on the knees, thus creating a vicious cycle6.
In response to the above mentioned issues, the body weight-supported training system (BWSTT) has gradually addressed bone and joint disease-related rehabilitation7. In recent years, one of the emerging body weight-supported training technologies is called the lower body positive pressure (LBPP) treadmill7. This technology uses a waist-high inflatable balloon to achieve positive lower limb pressure and accurately adjust the air pressure to regulate body weight with the aim of achieving weight reduction. The system is also equipped with a running platform that can concurrently perform treadmill-related activities under the control of body weight8. Meanwhile, the pressure generated in the inflated enclosure provides a lifting force against the body. Because the pressure is only slightly above atmospheric pressure and is evenly distributed, the force on the lower body is almost imperceptible. Thus, the LBPP running platform provides a higher level of comfort and is more suitable for long-term training compared with the traditional BWSTT9. Peeler et al. performed an LBPP treadmill intervention on 32 knee OA patients and showed that the LBPP treadmill can effectively relieve knee pain, improve daily life functions, and produce an increase in thigh muscle strength10. The potential mechanism might be related to achievement of effective knee joint activity while reducing knee joint torque11. On the other hand, since the age of onset of knee OA patients is mostly over 45 years old12, onset may also be associated with cardio-pulmonary diseases. Studies have shown that LBPP allows people to achieve walking as exercise with relatively low heart rate, blood pressure, and oxygen consumption and achieve safer and more effective aerobic exercise than full-weight flat walking; this type of walking is another advantage of LBPP when compared with traditional BWSTT13.
However, due to the relatively new application of this system to knee OA intervention, the relatively few existing studies have greatly limited the clinical application of this technology in knee OA rehabilitation. The LBPP protocol proposed in this article aimed to explore the clinical non-drug and surgical knee OA treatment using the LBPP treadmill.
The clinical project was approved by the Medical Ethics Association of the Fifth Affiliated Hospital of Guangzhou Medical University and has been registered at the China Clinical Trial Registration Center (No. ChiCTR1800017677 and entitled “Effect and Mechanism of Anti-gravity Treadmill on Lower Limb Motor Function in Patients with Knee Osteoarthritis”).
1. Recruitment
2. Pre-training evaluation
3. LBPP training
NOTE: An anti-gravity treadmill (Table of Materials) was used for this LBPP training protocol and shown in Figure 1. For patient safety, a therapist is required to set up the patient in the LBPP and supervise the whole treatment process.
4. Post-training evaluation
NOTE: The same therapist completes each patient's pre- and post-evaluation.
5. 3D gait analysis data processing
We show results from a knee OA patient, who was a 60-year-old female (BMI = 22.9) undergoing “more than 3 years of knee osteoarthritis” and severe pain when she was walking (visual analog scale [VAS] = 8/10) and participated in a 2-week LBPP training program at our facility. During the entire intervention, the patient did not take any painkillers to relieve knee pain. The radiological image of her knee joints and the results of clinical function assessments are shown in Figure 3 and Table 1.
The 10 MWT decreased from 4.1 s at pre-training to 3.3 s at post-training. The TUG test decreased from 9.1 s at pre-training to 8.2 s at post-raining. After two weeks of LBPP training, patients showed an improvement in total WOMAC scores (15 versus 8), pain subscales (8 versus 3), stiffness subscales (1 versus 0), and function subscales (6 versus 5). The total VAS pain scores or the knee flex-extension AROM did not change after two weeks of treatment.
The gait parameter results are shown in Figure 4. The right swing phase (%height) increased from 40.75 at pre-training to 41.51 at post-training (Figure 4A). The left swing phase (%height) decreased from 41.11 at pre-training to 40.33 at post-training (Figure 4B). The right stride length (%height) decreased from 77.00 at pre-training to 74.10 at post-training (Figure 4C). In contrast, the left stride length (%height) increased from 74.1 at pre-training to 75.68 at post-training (Figure 4C). The mean velocity (%height) increased from 74.44 at pre-training to 74.97 at post-training (Figure 4D). The cadence (steps/min) increased from 117.2 at pre-training to 119.8 at post-training (Figure 4E). The step width decreased from 0.08 m at pre-training to 0.06 m at post-training (Figure 4F).
Knee joint movement trajectories in the frontal, sagittal, and transverse planes are shown in Figure 5. Both right and left trajectories of the knees’ AROMs were closer to normal reference values at post-training than at pre-training, especially during the swing phase of knee AROM in the sagittal plane.
The results of tight muscle EMG activities are shown in Figure 6. The mean root-mean-square (RMS) of the left long head biceps femoris muscles, left rectus femoris, and left semitendinosus increased from 0.160 ± 0.069, 0.130 ± 0.054, and 0.259 ± 0.138 mV, respectively, at pre-training to 0.194 ± 0.136, 0.317 ± 0.215, and 0.315 ± 0.204 mV, respectively, at post-training (Figure 6A). The mean RMS of the right long head biceps femoris muscles, right rectus femoris, and right semitendinosus increased from 0.160 ± 0.022, 0.136 ± 0.013, and 0.259 ± 0.021 mV, respectively, at pre-training to 0.234 ± 0.018, 0.206 ± 0.009, and 0.438 ± 0.017 mV, respectively, at post-training (Figure 6C). The peak RMS of the left long head biceps femoris muscles, left rectus femoris, and left semitendinosus increased from 0.342 ± 0.094, 0.256 ± 0.245, and 0.528 ± 0.197 mV, respectively, at pre-training to 0.540 ± 0.032, 0.797 ± 0.116, and 0.784 ± 0.074 mV, respectively, at post-training (Figure 6B). The peak RMS of the right long head biceps femoris muscles, right rectus femoris, and right semitendinosus increased from 0.388 ± 0.078, 0.286 ± 0.036, and 0.855 ± 0.055 mV, respectively, at pre-training to 0.576 ± 0.098, 0.390 ± 0.024, and 1.300 ± 0.140 mV at post-training, respectively (Figure 6D).
The patient claimed that she was satisfied with the whole LBPP training process without any discomfort and would like to accept another session in the future.
Figure 1: Diagram of the LBPP setup and LBBP training protocol.
(A) Walking model; (B) squatting model; (C) The protocol and parameter setting of the LBPP training. AROM = active range of motion, BW = body weight. Please click here to view a larger version of this figure.
Figure 2: Example of definition of the right foot initial contact with the floor (green vertical lines) and right toe off (blue vertical line).
The knee flexion-extension angle (green) and the ankle dorsi-plantarflexion angle (red) are shown. R = right. Please click here to view a larger version of this figure.
Figure 3: The digital radiography image of the knee OA patient at pre-training. Please click here to view a larger version of this figure.
Figure 4: The spatial-temporal parameters of the patient with knee at pre and post LBPP training intervention.
(A) The percentage of right stand phase (dark green) versus swing phase (light green) in the gait cycle. (B) The percentage of left stand phase (dark red) versus swing phase (light red) in the gait cycle. (C) The stride length (%height) of right side (green) versus left side (red). Panels (D), (E), and (F) show mean velocity (%height/s), cadence, and step width, respectively. Please click here to view a larger version of this figure.
Figure 5: Knee joint movement trajectories in gait cycle at frontal, sagittal and transversal plane.
The knee movement trajectory of a normal subject as the normal reference (grey) is also shown, which refers to the motion capture system (Table of Materials). Please click here to view a larger version of this figure.
Figure 6: The synchronized EMG activity of the patient with knee OA in the gait cycle at pre and post LBBP training intervention.
Panels (A) and (C) show the mean RMS of muscle activity in biceps femoris caput longus, rectus femoris and semitendinosus, respectively; panels (B) and (D) show the peak RMS of muscle activity in biceps femoris caput longus, rectus femoris and semitendinosus, respectively. RMS = root mean square; mV = microvolt; L = left; R = right. Please click here to view a larger version of this figure.
Clinical Assessment | Pre-training | Post-training |
10MWT (SPP) | 4.1 s | 3.3 s |
TUG | 9.1 s | 8.2 s |
WOMAC-pain | 8 | 3 |
WOMAC-stiffness | 1 | 0 |
WOMAC-functionality | 6 | 5 |
NRS (pain at rest) | 0 | 0 |
Knee flex-extension AROM | Left: 0°−130° | Left: 0°−130° |
Right: 0°−130° | Right: 0°−130° |
Table 1: Clinical assessment results.
10MWT = 10-meter walk test; SSP = self-selected pace; TUGT = timed up and go test; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index; NRS = the numerical rating scale; AROM = active range of motion; s = second.
We proposed an LBPP treadmill intervention protocol, which includes both clinical assessment and treatment models, for the rehabilitation of lower extremity motor function in knee OA. Meanwhile, in response to the clinical symptoms and knee OA dysfunction, the treatment model includes not only a training section for walking in the LBPP protocol but also an innovative squatting training section, which aims to solve the daily dysfunction due to thigh muscle weakness and squatting difficulties in knee OA patients. To the best of our knowledge, this protocol is the first to include a squatting exercise regimen with un-weighting technology in knee OA patients.
The design of this protocol was based on five main points. First, pain and the resulting break are the main problem of patients with knee OA. This protocol aims to explore the potential effect of an anti-gravity treadmill for increasing the amount of exercise by reducing knee load and pain during exercise in knee OA patients22. Therefore, the inclusion criteria focused on knee OA patients with knee pain when walking, squatting, and/or kneeling. Second, WOMAC and KOOS are both widely used in clinics to assess the physical function of knee OA patients. WOMAC is used to assess symptoms associated with the condition of patients with OA of the knee and hip (five pain items, two stiffness items, and 17 joint function items) and to reflect the severity and therapeutic effects of arthritis16. The KOOS is a self-administered instrument used for evaluating knee-associated problems including knee joint injury and OA (42 items in five subscales: pain, other symptoms, activities of daily living, sports and recreation)17. Moreover, the EQ-5D is used to assess the general condition of patients, which includes five dimensions (mobility, self-care, usual activities, pain/discomfort, anxiety/depression)18. Although this protocol is mainly focused on pain and physical function of mild to moderate knee OA patients, the KOOS and EQ-5D for a comprehensive evaluation of health is optional and recommended. Third, the LBPP training session consists of the walking and squatting modules. The walking module focuses on improving walking function and knee activity, and the squatting module focuses on the enhancement of the tight muscle strength23. It is noticeable, however, that retropatellar arthritis should be excluded from our LBBP training protocol due to tight anatomical structures (mis-tracking of the patella through the femoral groove) and the squatting-induced heavy physical loading pressure, which can aggravate pain24. Meanwhile, if the patient cannot tolerate the squatting training model, only the walking mode is performed. Fourth, gradual warm-up and cool-down periods are important for better adaptation with high exercise intensity at the beginning of exercise session and restoration of full body weight slowly before stopping the exercise session. Lastly, in our protocol, the frequency of alter-gravity treadmill training is six times per week for two weeks, but the training frequency can be adjusted according to the specific situation of the patient and their Medicare payments, such as one session of treatment with two to three times per week for three−four weeks.
Comparing the results at pre-training and the 2-week post-training, which is presented in the representative results section, the functional improvement was mainly reflected in three aspects. First, the improvement in walking ability, which is reflected in the decrease in time cost of the 10 MWT and TUG tests (the reduction of TUG also indicates a reduction in the risk of falling) (Table 1) in addition to improvement in 3D gait analysis parameters, including an increase in the mean velocity (%height) and cadence and decrease in step width (Figure 4). Second, an increase in muscle strength in thigh muscles, including the rectus femoris, semitendinosus, and long head biceps femoris on both sides (Figure 6). Third, a reduction in knee pain (although the overall NRS pain score was not apparent at pre-training under resting conditions, the patient complained that the main pain was induced during functional activities, such as walking or climbing up and down stairs). Moreover, after two weeks of the LBPP training, the WOMCA assessment showed a significant reduction in pain during functional exercise (Table 1). Additionally, the results collected from the 3D gait motion analysis system at pre and post LBPP training sessions were consistent with the results of the clinical evaluation scales in our study. It is worth noting that the active knee joint mobility did not improve significantly before and after treatment, but the 3D gait motion analysis showed that both sides of knee joint movement trajectories were closer to the normal reference in the sagittal plane at post-training than at pre-training (Figure 5). Meanwhile, the patient has no restriction in AROM, no resting pain. This could explain why knee ROM did not change.
We must address certain limitations in this article. First, this article aims to provide a protocol for anti-gravity treadmills in knee OA patients based on our past clinical experience and previous research reports10,11,22. However, our findings are only valid in this case report (due to the lack of objective evaluation methods in our past clinical applications, such as 3D gait analysis and the conventional control group). The clinical efficacy of this approach requires further investigation. Second, neither the protocol nor the case report involved multiple sessions or follow-up. Considering the irreversibility and progress of the knee OA disease, we recommend that this cohort should be followed up as part of future studies.
The authors have nothing to disclose.
This study was funded by Guangzhou Medical University (Grant Number 2018A053).
AlterG Anti-Gravity Treadmill M320 | AlterG Inc, Fremont, CA, USA | 1 | LBBP training |
BTS Smart DX system | Bioengineering Technology System, Milan, Italy | 2 | Temporospatial data collection |
BTS FREEEMG | Bioengineering Technology System, Milan, Italy | 3 | Surface EMG data collection |
BTS SMART-Clinic software | Bioengineering Technology System, Milan, Italy | 4 | Data processing |