- Design
This was a randomized trial submitted in 06/08/2021 (NCT04046042). The participants were randomized into two groups by the nephrologist of the HD Unit: Group A exercised within the first two hours of HD and Group B exercised within the last two hours of the HD session. The study was carried out in the HD Unit of Hospital de Manises (Spain). Patient recruitment was open from September 2019 with follow-up until April 2023. The inclusion criteria were being on chronic HD treatment for 3 months or more, remaining clinically stable, and providing informed consent. The exclusion criteria were having suffered from myocardial infarction in the 6 weeks running up to the study, unstable angina, amputation of lower limbs above the knee with no prosthesis, cerebral vascular disease like stroke or transitory ischemia, musculoskeletal or respiratory alterations that worsen with effort or unable to perform functional physical tests.
- Intervention
The intervention with the intradialysis VR exercise program was implemented by the HD unit staff and supervised by a physiotherapist. The intradialysis exercise session was implemented for at least 6 months and started with a 5-minute warm-up, followed by the VR exercise session, depending on each patient’s fitness level. Group A and B exercised at different times as described above. To develop the VR exercise program, we used the adapted version of the Treasure Hunt game, a non-immersive VR system designed with a game format [11] [27]. In Treasure Hunt, the player must play by raising a lower limb and alternating the movement of both limbs across the screen. The difficulty level was adapted to the participants’ results. The hardware we used was a standard personal computer and monitor screen and Ms Kinect®, a movement-detection camera. At the beginning of the session, Treasure Hunt allows the therapist to define the VR session by selecting the number of exercise sub-sessions and their duration, as well as the sub-session rest periods, that were kept at 1 minute. The game difficulty was adaptive, and so the system automatically increased or decreased the level of difficulty depending on each person’s results. Once the program was finished, a 5-minute cool-down including stretching was performed. The intensity was registered through the rate of perception scale (RPE) and the session was adapted so that participants reported a rate of perceived exertion (RPE) between 12 and 15 (6 to 20 RPE scale). If the session was too easy (6–11) or too hard (16–20), the physiotherapist re-adjusted the number of sub-sessions and/or their duration, with a maximum of 6 sub-sessions, lasting 6 minutes for each sub-session. Adherence was defined as the percentage of sessions the participant performed from the total number of sessions offered.
- Measurements
Descriptive variables were recorded from the electronic medical record of each patient.
The hemodynamic control variables were Systolic blood pressure (SBP), Diastolic blood pressure (DBP), and Heart rate (HR). These variables were recorded every 30 minutes during all the HD sessions. These variables were compared in two ways. The first comparison was, in each HD session, variables recorded during the Rest time vs the Exercise time. The second comparison assessed the differences between the hemodynamic control variables (SBP, DBP and HR) during the Exercise time for Group A vs B.
The events of hypotension, hypertensive crisis, and cardiac arrhythmias were recorded during the HD sessions and are shown as the total number of events divided by the total number of HD sessions. Intradialysis hypotension was considered according to the KDOQI criteria: (SBP pre-HD – minimum intraHD SBP) ≥ 20mmHg and symptoms attributable to hypotension, and according to the Fall20Nadir90 criteria [30] (SBP preHD – SBP minimum intradialytic SBP) ≥ 20 mmHg and minimum intradialytic SBP < 90 mmHg. Hypertensive crises were considered for men SBP greater than 210 mmHg and 190 mmHg for women [28]. Clinically arrhythmia was defined as any of ventricular tachycardia > 130 beats/min lasting for ≥ 30 seconds, bradycardia of < 40 beats/min sustained for > 6 seconds, asystole for > 3 seconds, or patient-reported symptomatic events that were validated as being associated with clinically relevant arrhythmia by a serious event committee [29]. Events were recorded in two moments Baseline (events in the month before the start of the program) and Exercise (events during the exercise sessions). We studied in whole sample differences for events in Baseline vs Exercise and differences in Group A vs B for events during Exercise time.
In addition, chronic blood pressure control and antihypertensive drugs prescribed were studied in all patients who exercised for 6 months or more. Two moments were compared: Baseline (month before the start of the program) vs Final (last month of the exercise program). SBP, DBP, HR and mean arterial pressure (MAP) were determined before all HD sessions. MAP was calculated with the formula MAP = [SBP + 2 (DBP)] / 3 [33].
Dialysis dose and molecule rebound were calculated in one HD session (second weekly session out of three) without exercise (Rest) and in another session with intradialysis Exercise in consecutive weeks.
The dialysis dose was measured by second-generation KtV Daugirdas:
Kt/V Daug = - ln ((C2/C1) - (0.008*T)) + (4–3.5 * (C2/C1)) * UF/P
(C1: pre-dialysis Blood Urea in mg/dL; C2: post-dialysis Blood Urea in mg/dL; T: Session duration time in minutes UF: volume of removed ultrafiltrate in liters; P: post-dialysis weight of a subject in kilogram)
Urea, creatinine, potassium, and phosphorus rebound was calculated by the formula (34): Rebound (%) = 100 x (CR-C2) / C2
(C2: blood sample taken at the end of session; CR: blood sample taken 30 min after the session ended)
The nutrition and metabolism variables were recorded in two moments, Baseline (starting month of the program) Final (last month of the exercise program). The recorded variables were serum albumin, serum hemoglobin, serum potassium, serum phosphorus, serum calcium, serum ferritin, serum bicarbonate, serum triglycerides, serum LDL cholesterol and serum HDL cholesterol. We compared these results in two groups, control and exercise groups. The control group included participants who participated for less than 3 months in the clinical trial and dropped out voluntarily (patients who dropped out due to an incident pathology were excluded from the control group). The measures were taken at least 6 months from the start of the exercise program. The exercise group included participants in the clinical trial who exercised for, at least, 9 months.
- Statistical analysis
The sample size was estimated using the GRANMO sample size calculator, Version 8.0 April 2012 of the Program of the Girona Heart Registry (REGICOR), IMIM, Barcelona https://www.datarus.eu/en/applications/granmo/
Accepting an alpha risk of 0.05 and a power of 0.8 in a two-sided test 30 subjects are necessary in the first group and 30 in the second group to recognize statistically significant a difference greater than or equal to 10 mmHg in SBP. The common standard deviation is assumed to be 18 mmHg and the correlation coefficient between the initial and the final measurement is assumed to be 0.8. A drop-out rate of 30% has been anticipated. The expected results have been based on results obtained in previous studies[11] [14].
Data are presented as mean, standard deviation for continuous variables, and relative and absolute frequencies for categorical variables.
The analysis of the hemodynamic control measures (SBP, DBP and HR) was done using a mixed linear regression model that compares differences depending on the moment of the exercise during the session (Group A vs. Group B). Differences between the hemodynamic control variables (SBP, DBP and HR) during the Exercise time for Group A vs B were compared with a linear regression.
Events of hypotension, hypertensive crisis, and cardiac arrhythmias in Baseline vs Exercise for the whole sample were compared with the mixed linear regression. Differences in Group A vs B for events during Exercise time were compared with the linear regression.
Chronic blood pressure control and antihypertensive drugs prescribed (baseline vs final) were studied in all patients and compared with mixed linear regression.
Analysis of dialysis dose, nutrition, lipid metabolism, anemia, and bone mineral disease were compared baseline vs final in the control and exercise group with a mixed regression analysis.
Values of P < 0.05 were considered statistically significant. All analyses were performed using R software (version 4.2.2).