Effects of intermittent hypoxia-hyperoxia on mobility and perceived health in geriatric patients performing a multimodal training intervention: a randomized controlled trial

Forty-one geriatric patients between 64 and 92 years participated in this stratified, randomised and double-blind study. The study was performed at the Geriatric Day Clinic in Klagenfurt (Carinthia, Austria), this is a semi-stationary facility in the house of geriatrics offering elderly people the possibility of rehabilitation with the goal: to enable the elderly to live as independently as possible in health at home. All patients of the Geriatric Day Clinic suffer from several different diseases. The most frequent diagnoses of the study participants were arterial hypertension, condition after surgery of a total endoprosthesis, partially after fractures due to falls, osteoporosis, degenerative spinal diseases, arthrosis, dementia development, atrial fibrillation, heart failure, coronary heart disease, diabetes mellitus, renal insufficiency, gastrointestinal diseases and depression. The patients were randomly assigned to the hypoxic group (HG) and the normoxic group (NG). The process of inclusion, randomization, stratification, training program and outcome analysis is presented in Fig. 1, the baseline characteristics in Table 1. A more detailed description of the participants and methods is provided in our previous paper [10]. All comorbidities, therapies, and interventions of medical care were documented in written documents and also electronically stored in the intranet of the hospital.

At the end, data from 34 patients who successfully completed the entire study program were available for the outcome analysis (Fig. 1).

The study was approved by the local Ethics Committee (EK-Nr.: A 09/14) and performed in accordance with the ethical standards of the Declaration of Helsinki in 1975.

Data are presented as means ± standard deviation (SD) or proportions. Unpaired t-tests (normally distributed data) and the Wilcoxon-Mann-Whitney U-test (not normally distributed data) were used to compare baseline data between groups as well as different changes (delta pre-post) between groups. Fisher exact test was used to compare proportions. Pearson or Spearman correlation analysis was performed to test relationships as demonstrated in 3.2.

Mobility was measured by the values of the TMT and the Barthel-Index and by the time needed for the TUG. Before and after the intervention, there were no significant differences between groups. After the intervention there was an improvement in all test results, but again with no difference between the two groups. IHHT did not lead to a significant additional improvement.

Testing the perceived health with the EQ-5D revealed no significant difference between the groups at the beginning. The increase in the level of perceived health was somewhat higher in the HG but did not reach statistical significance compared to the NG (Table 2).

Improvements in mobility and the decreased fall risk within the overall group were significantly correlated with an improvement in the EQ VAS, as shown in Fig. 2a and b. There was a significant negative correlation between the differences of the TUG and the EQ VAS (Fig. 2a) and a correlation between the differences of the TMT and the EQ VAS (Fig. 2b), which just failed to be statistically significant, with a p-value of 0.057, but showed a clear trend. Even the outliers do not influence these results. There was no significant correlation between EQ VAS and the Six-Minute-Walk-Test (r = − 0.23, p = 0.09), between EQ VAS and the Clock-Drawing-Test (r = − 0.048, p = 0.4) and between EQ VAS and the Dementia Detection Test (r = 0.12, p = 0.24) as shown in Fig. 3. A more detailed prescription of these test results can be found in our previous article [10].

A slight but not significant decrease in systolic blood pressure, triglycerides and erythrocytes was observed within groups while the decrease in diastolic blood pressure was statistically significant within both groups. A significant decrease in resting heart rates was only found within the NG and a significant decrease in total cholesterol, HDL and LDL levels was observed in the HG. Arterial oxygen saturation and TNF alpha increased significantly after the intervention also only within the HG. None of these parameters changed differently between groups [10].

The risk of falling is especially high at values ​​below 19 in the TMT, as reported by Tinetti [26] and the “Kompetenz-Zentrum-Geriatrie” in Germany report [33]. This is also true for both groups before intervention, with a mean value below 19. After the intervention, both groups revealed a mean value greater than 19, which indicates a change from a high to a medium risk for falls [26]. A TMT score of 11 or less is even predictive of day hospital patients having a history of recurrent falls [34]. The TMT tests predictive and reactive balance mechanisms [35]; as so many body systems are needed to maintain balance [36], Thomas et al. recommend the TMT as a potential screening tool to identify patients at risk for falls in their pilot study. The scores of the TMT showed significant differences between fallers and non-fallers in this retrospective study [34]. Also, the TUG is claimed to predict falls, as other studies have reported, the shorter the time the better the result [37, 38]. A TUG with > 12 s is one of the most evidence-supported functional measures to determine individual risk of future falls, as reported by Delbaere et al. [39]. This reveals that our study participants, with a mean TUG of more than 12 s in both groups before intervention, have a high risk of falls. The TUG has almost as high a sensitivity and specificity as the TMT and contains components such as sit-to-stand, gait and turning, which are important aspects of postural control and are functionally important because many falls occur while walking [34]. The causes of falls are diverse and complex, but IHHT, which has positive effects on memory and exercise tolerance [10, 21], seems not to be an appropriate therapy for fall reduction. We were not able to confirm the hypothesis that IHHT would significantly contribute to expected improvements in the MTI. This study demonstrates improved mobility in both groups through MTI. However, since we did not include a control group, no valid statement can be made.

In our study, IHHT did not improve perceived health assessed with the use of EQ VAS. Various factors influence the subjective perception of quality of life. Perceived social support and marital status are associated with improved psychological health and QoL [40], as well as a higher socio-economic position [41]. All of these factors can hardly be influenced by therapy, but the geriatric day clinic also tries to offer help in these cases with consultations by social workers and psychologists. Being free of physical disability seems to be one of the most important factors that influence QoL [3]. Wahrendorf et al. found that there is also a strong association between functional limitations and changes in quality of life; in their study, being free from functional limitations led to a significant increase in QoL [42]. Also in our study, the improvements in mobility and the decrease in fall risk within the overall group were significantly correlated with an improvement in perceived health (Fig. 2). Vellas at al. observed a strong association between fear of falling and a decrease in mobility and QoL in their study [43]. Falls and a fear of falling in turn contribute to restricted activity as a strategy to reduce the perceived risk of subsequent falls [44], which closes the vicious circle. Even depression is related with a higher risk of falls [45]. However, IHHT did not positively affect perceived health more than MTI alone in the present study with geriatric patients. In contrast to mobility, there is no relationship in our study between improvements in exercise tolerance measured with the 6MWT and perceived health assessed with EQ VAS (Fig. 3), which shows that endurance does not influence QoL sufficiently and measurably. It seems that enhancements in mobility, gait and balance directly influence the way in which patients can easily cope with their everyday life and reduces the risk of falls. This, in turn, leads to an increase in the perceived health [42, 43]. Dementia, on the other hand, does not seem to reduce the quality of life for the affected patients; only the affected caregivers show a strong reduction in their own quality of life [46]. In our study, there was no statistical significant correlation between perceived health and cognitive performance (Fig. 3). However, as IHHT leads to an improvement in cognitive performance [10] and as Jing et al. showed that at least caregivers profit from an enhanced cognitive performance of elderly people [46]. IHHT may positively affect elderly patients in several ways [10, 18, 21], but this seems not to be true regarding mobility and perceived health.

First, it is not possible to have the exact same MTI for all patients because of the multi-morbidity of geriatric patients; a special, individual training program has to be created for each person. Second, as muscle weakness is an important risk factor for falls, we did not measure the strength of the lower extremities. Third, we had no control group without intervention to make a valid statement on the effect of MTI on perceived health and fall risk. Fourth, much effort and time of trained supervisors is needed for the individual MTI and IHHT to reach sufficient compliance by elderly people.