Physical fitness and health‐related quality of life in pediatric renal transplant recipients: An interventional trial with active video gaming

Pediatric renal transplant recipients are at increased risk for cardiovascular diseases, one contributing factor is reduced cardiorespiratory fitness. The purpose was to evaluate cardiorespiratory fitness, motor coordination, muscle strength, daily physical activity, and health‐related quality of life and to find out, if active video gaming is effective for improving these items in this patient population.


| INTRODUC TI ON
Renal transplantation is the preferred treatment for pediatric endstage renal disease and has improved life expectancy and quality of life of these young patients. However, patients who underwent renal transplantation during childhood are at increased risk for metabolic and cardiovascular diseases such as obesity, dyslipidemia, diabetes mellitus, and hypertension, probably at least partly due to insufficient regular physical activity and reduced fitness. [1][2][3][4][5][6][7] For example, in an investigation of Akber et al, 1 only 10% of the male and 5% of the female patients with chronic kidney disease (7-20 years of age) met the recommendations for these age groups of 15 000 and 12 000 steps per day, respectively. 8 Lifestyle modifications including an increase of physical activity are of utmost importance for improving the outcome after pediatric renal transplantation. 9,10 To date, only few studies investigated the beneficial effects of an exercise intervention in renal transplanted subjects, [10][11][12][13] and to the best of our knowledge, only one study focused on an increased physical activity and cardiorespiratory fitness in TX. 10 Recently, the use of active video games has been reported to be a feasible and effective method to improve regular physical activity, physical fitness, cardio-metabolic health, and health-related quality of life in children and adolescents with various health problems. [14][15][16][17][18][19] Active video gaming can also be used for homebased exercise programs. For TX who often live at some distance to the renal transplantation outpatient clinic, home-based exercise intervention might be better suited than center-based programs.
In overweight adult patients with chronic kidney disease, a homebased exercise program was as effective as center-based training for the improvement of physical fitness and health-related quality of life. 20 The primary aim of the present study was to investigate the cardiorespiratory fitness, muscle strength, motor coordination, daily physical activity, and health-related quality of life of TX in comparison with CON (cross-sectional study). The secondary aim was to find out whether a home-based exercise intervention with the use of active video gaming is an appropriate tool for improving physical capacity and health-related quality of life in this patient population. Therefore, a subgroup of the cross-sectional study was subjected to an exercise program with the Wii fit console of Nintendo ® . We hypothesized that 6 weeks of regular exergaming could initiate an increased general daily physical activity and F I G U R E 1 Flowchart of recruiting and analyzing pediatric renal transplant recipients and healthy controls. n, sample size; t1, first baseline testing; t2, second baseline testing (immediately prior to intervention); t3, testing after active video game intervention; yr, years of age would improve the impaired fitness in these young renal transplanted patients.  (Figure 1).

| Healthy controls
Thirty-three CON (25 males and 8 females; CON) from various local schools agreed to participate in the first part of this trial (cross-sectional study) and completed the exercise tests, the activity screening, and answered the PedsQL ™ 4.0 questionnaire. The principle aim was to match two healthy controls with one renal transplanted patient. Matching criteria were sex, pubertal stage, attended type of school, and regular physical activity per week, which were assessed by a semi-quantitative telephone interview. Although two healthy control subjects could not be recruited for each pediatric renal transplant recipient (TX, n = 20; CON, n = 33), the groups were well comparable regarding the matching criteria.
Based on the information about regular physical activity provided by the participants and their parents within a personal interview, the study participants were stratified into the following subgroups: (a) no physical activity besides physical education, (b) low (or no regular) physical activity, and (c) moderate physical activity, that is, more than once per week.
The study was approved by the local Ethics Committee (S-043/2015) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all parents/guardians, with assent from patients when appropriate for their age.

| Study design
All participants were subjected to cycling or treadmill spiroergometries, motor coordination, and handgrip strength tests. To minimize habitual effects, these tests were performed twice (t1 and t2).
For spiroergometry data, mean values of the two baseline measurements t1 and t2 were used for statistical analyses. Because of expected learning effects for motor coordination and maximal handgrip strength, only the results of t2 were used as baseline values for TX and CON, respectively. Furthermore, at t1, activity of daily life (steps per hour) was evaluated with a SenseWear ® device (BodyMedia ® ) and health-related quality of life was assessed with PedsQL ™ 4.0. In the TX Wii subgroup, all examinations were repeated after the Wii fit intervention (t3) described below.
At each investigation date (t1-t3), the participants' weight and height were measured. Weight, height, and BMI were converted to Z-score values related to age-and gender-specific means and SD of European reference populations. 22,23 Systolic and diastolic blood pressures were derived from casual blood pressure measurements by sphygmomanometer and converted to age-and gender-specific Z-score values based on reference data from a large national trial. 24 Furthermore, a 12-lead resting ECG was recorded. In the TX cohort, serum creatinine for the eGFR 25 was determined within the routinely performed blood tests at the renal transplantation outpatient clinic. No blood samples were taken from CON.

| Spiroergometry
To assess cardiorespiratory fitness, a spiroergometry was performed on an electronically braked cycle ergometer (Corival, Lode BV Medical Technology) or in small children on a motorized treadmill (Quasar med, h/p/cosmos ® ). On the cycle ergometer, the exercise test started with 20 W and load was increased by 10 W every minute until volitional exhaustion, followed by 5 minutes of recovery with 20 W. The ergometer seat could not be adapted for small children (ie, body height <135 cm; TX, n = 3; TX Wii , n = 1; CON, n = 5). Therefore, these small children performed a walking treadmill test at an individually chosen speed starting with a treadmill inclination of 1%, which was increased by 1% every minute until volitional exhaustion.  ). 26 One TX patient did not tolerate the spirometry face mask and thus completed a treadmill ergometry instead. In TX Wii , spirometry data were available for 11 of 13 patients because of two measurement errors at t3.

| Motor coordination test
Motor coordination level was checked with a motor coordination test for children and adolescents comprising the four motoric tasks balancing backwards, jumping on one leg, jumping side to side, and shifting platforms. All tasks were executed according to the procedure described in detail by Kiphard

| Maximal handgrip strength
Handgrip strength was measured with a hydraulic Jamar ® hand dy-

namometer (Sammons Preston). Participants executed the test while
standing straight and bending the operating arm at a 90° angle. Arms were held tight to the trunk. The handle could be adjusted to participants' hand size, so most of the tested children and adolescents used the smallest settings 1 and 2, and some of the elder participants used setting 3. Maximal handgrip strength was determined out of three maximal attempts with the non-dominant forearm. Maximal handgrip strength could not be measured in one TX patient.

| Daily physical activity
Daily physical activity was recorded for at least 5 consecutive days at t1 and for TX Wii additionally at t3 using a SenseWear ® -device (BodyMedia ® ). The SenseWear ® armband was placed at the upper arm of the non-dominant side and was prepared individually before recording. Only measurements with at least two consecutive days of recording and at least 12 hours of patients being awake, respectively, were included for data evaluation. Recorded data were evaluated using the SenseWear ® Professional software (version 8.1).
The mean values of steps per hour, instead of steps per day, were calculated because of the wide inter-individual variation in the participants' time of being awake. To identify outlier days, participants were asked to keep an activity diary while wearing the SenseWear ®device. One TX patient refused to wear the SenseWear ® armband.

| Health-related quality of life
To determine health-related quality of life, TX and CON were asked to answer the PedsQL ™ , version 4.0 Generic Core Scales at baseline (t1) and for TX Wii additionally at t3. The PedsQL ™ 4.0 consists of 23 items to assess physical, emotional, social, and school functioning of children (8-12 years), adolescents (13-18 years), and young adults (18-25 years). The total maximum score could be 100 for the whole questionnaire and every subscale. 28 In this regard, a higher score corresponds to a higher level of health-related quality of life.
Because one of the patients had serious problems in understanding the PedsQL ™ 4.0 items linguistically and substantively, only 19 questionnaires could be evaluated to identify the child/teen reported overall score.

| Wii fit intervention
For the second part of the investigation, a Nintendo ® Wii game console including a balance board and the Wii fit DVD were allocated to the renal transplanted subgroup TX Wii for 6 weeks. The young patients were instructed to practice 3 × 30 minutes per week at home and were asked to record their heart rate, measured by a provided heart rate monitor (Polar FS1C) as well as their perceived exertion using a Borg Scale 29 after each exercise bout.
Furthermore, they were contacted regularly via email or phone by the investigators and were asked to return the activity protocol.
Intervention compliance was rated based on the completeness and returning of the practicing protocols. Compliance was interpreted as good, if the patients exercised more than half of the required practicing time.

| Statistical analysis
All statistical tests were processed using SPSS Statistics for Comparability of TX and CON concerning the matching criteria sex, pubertal stage, attended type of school, and regular physical activity was tested using the chi-square test. t Tests for independent samples were calculated to check significant differences between TX and CON. Because normal distribution was missing for absolute V O 2peak and overall score PedsQL ™ 4.0, differences were tested using the Mann-Whitney U test. To assess effects from pre-to post-intervention in TX Wii , t tests for dependent samples were calculated for all parameters except absolute V O 2peak (Wilcoxon's test). Because of small sample sizes, no statistical tests were calculated for subgroups relating to regular physical activity and intervention compliance. As this is an exploratory trial, P-values have no confirmatory value, but will be considered significant (P < .05) in descriptive manner. Table data are presented as mean ± SD. Box plots concerning the comparison of TX and CON show medians and whiskers as minimum to maximum.
Floating bars, to present the relative changes from pre-to postintervention in TX Wii , show lines as median and the bars range from minimum to maximum.

| Anthropometric and health-related data
Anthropometric and health-related data for TX, TX Wii , and CON are presented in Table 1. There were significant differences between TX and CON concerning the height Z-score (P = .003) as well as BMI (P = .029) and BMI Z-score (P = .043), while all other anthropometric and health-related parameters were similar in both groups. All TX patients received immunosuppressive medication, 17 of 20 glucocorticoids. Additionally, 17 of the kidney-transplanted patients were treated with antihypertensive drugs and 8 of them with beta blockers.

| Compliance during exercise intervention
Based on the heart rate records and the information about perceived exertion, only 5 patients out of the 13 TX Wii performed the intervention exercises as required and were identified as well compliant. Heart rate during aerobic and strengthening exercises with the Wii console of the well-compliant subgroup was approximately 65% of the peak heart rate measured in the spiroergometry tests before the exercise intervention. Overall perceived exertion rated with a Borg Scale 29 was 10.9 (range 8.5-12.5) which is defined as a fairly light effort.

| Cross-sectional study
In TX, relative V O 2peak was significantly reduced by 31% on average ( Table 2) and relative peak power output showed a significant average reduction by 37% compared with CON (1.9 ± 0.5 W·kg −1 vs 3.0 ± 0.5 W·kg −1 ; P < .001; Figure 2A and 2B). A significant lower (21%) V E peak was observed in TX compared with CON (Table 2). An average RER peak of 1.11 in TX and CON indicated a high effort in both groups toward volitional exhaustion. Peak heart rate was significantly lower in TX than in CON (Table 2). Furthermore, significant differences were observed between TX and CON for V O 2 and heart rate at the ventilatory thresholds ( Table 2). While V O 2 related to V O 2peak was similar in both groups at these thresholds, V O 2 per kg body weight at V T1 and V T2 was significantly reduced in TX compared with CON by 35% and 32%, respectively. Also, heart rate at V T1 and V T2 was significantly lower in TX than in CON. Regarding the participants' regular physical activity per week, there were tendencies for V O 2peak and peak power output to be increased in subjects with a higher physical activity level in both groups (Figure 2A and 2B).

| Exercise intervention
After 6 weeks of active video gaming, absolute values of V O 2peak and peak power output remained unchanged ( Figure 3A and 3D). Related to body weight, there was even a significant average reduction by 7% in V O 2peak (Table 2). Peak values for RER, V E, and heart rate did not change significantly. V O 2 related to V O 2peak at both ventilatory thresholds was similar before and after intervention. However, a significant decrease in V O 2 at V T2 and a tendency for decrease (P = .076) in V O 2 at V T1 were observed after the exercise intervention by 7% and 11% on average, respectively. Heart rate at V T2 also tended toward a decrease (P = .085; Table 2). With regard to the compliance in TX Wii , a decrease in V O 2peak and peak power output, especially related to body weight, was observed in patients with low compliance, whereas V O 2peak and peak power output was maintained in complying patients ( Figure 3A, 3D).

| Cross-sectional study
TX showed significantly reduced levels of motor coordination compared with CON (P < .001). The mean total MQ of TX (59.7 ± 17.5) corresponded to a disordered motor coordination level, while CON reached normal motor coordination levels (MQ total = 105.8 ± 14.9).
With regard to regular physical activity per week, TX with higher activity levels showed higher levels of motor coordination than inactive patients, but did not yet reach motor coordination levels of the CON subgroups ( Figure 2C).

| Exercise intervention
Regardless of the intervention compliance, there was no change of motor coordination after 6 weeks of active video gaming ( Figure 3B).

| Cross-sectional study
No significant difference between TX and CON was found for maximal handgrip strength. There was a tendency for a greater handgrip strength in moderately active subjects in both groups ( Figure 2D).

| Exercise intervention
After intervention handgrip strength remained unchanged in all of TX Wii ( Figure 3E).

| Cross-sectional study
On average, CON did 229 steps per hour more than TX, the difference being significant (P = .001). However, no tendency could be observed regarding the self-reported physical activity per week in both groups ( Figure 2E).

| Exercise intervention
After 6 weeks of active video gaming, daily physical activity of TX Wii was significantly (P = .043) increased by 121 steps·per hour (approximately 31%; Figure 3C). The average improvement of daily physical activity could be observed for both, the low and the good complying patients.

| Cross-sectional study
The total score of child-reported health-related quality of life was significantly lower in TX than in CON (75.0 ± 14.9 vs 85.2 ± 7.6; P = .017). In TX, a high inter-individual variability concerning healthrelated quality of life was observed. However, it was lowest in patients with no regular physical activity per week. Health-related quality of life was comparable for all activity subgroups of CON ( Figure 2F).

| Exercise intervention
TX Wii showed no significant improvement in overall health-related quality of life score after 6 weeks of active video gaming ( Figure 3F).
In contrast to the low-compliant subgroup, the complying patients showed an average increase of 7% in health-related quality of life after intervention.  home-based exergaming intervention 19 as well as for 10 weekly facilitated exergaming sessions. 14 However, in the TX of the present study, compliance was low, probably at least partly due to disordered motor competence.

| D ISCUSS I ON
To our knowledge, the present study is the first that investigated motor coordination in TX. Motor coordination was found to be on the lowest measurable level according to Kiphard and Schilling,27 two levels below the normal values of the CON. The substantial limitations in motor competence could have caused an enormous mental effort to correctly perform the required exercises and might also provide an explanation for the low heart rate (65% of peak heart rate) during the aerobic exercises of the active video game sessions.
Differences in the execution of exercises in active video games were observed between healthy weight and obese children causing less energy expenditure and therefore smaller cardio-metabolic benefits in obese children. 32 It is very likely that the TX of the present study were discouraged by the high coordinative effort, and this high mental and coordinative effort might also have intensified the concern to injure the kidney by physical activity. 31 In accordance, a high dropout rate has previously been reported for programs intending to increase physical activity of children with chronic kidney disease. 33  The association between health-related quality of life and physical activity levels is well known. 37  be not appropriate to improve these items in TX. However, the active video game intervention provided a sufficient stimulus for an increase in daily physical activity. Considering the low compliance in the present study, high dropout rates in previous studies 33

ACK N OWLED G M ENTS
We thank the renal transplanted and healthy participants, their parents, and the medical professionals at the renal transplantation outpatient clinic as well as the principals of the local schools for their support in the recruiting and examination process of the present study. Furthermore, we thank Prof. Gerhard Huber and Dr Markus Buchner (Institute of Sports and Sport Science, University of Heidelberg) for providing the motor coordination test materials and the Jamar ® hand dynamometer. No financial support was received for the present study.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interest to report.