Identifi cation of susceptibility to acute mountain sickness by detecting vascular tone using a photoplethysmographic sensor

Xiao-min LUO, Lei WANG Lei YANG

Identifi cation of susceptibility to acute mountain sickness by detecting vascular tone using a photoplethysmographic sensor

Xiao-min LUO1,2, Lei WANG1, Lei YANG1

1.Tibet Branch, Beijing Genomics Institute, Lhasa 850032, China;
2. Healthcare Dept., Beijing Genomics Institute, Shenzhen 518083, China

Objecti ve:Vascular tone had shown the potenti al suscepti bility to acute mountain sickness (AMS), however the detailed tendency has not been studied.Methods:Vascular tone, SpO2and Rate pressure product (RPP) were studied in seventeen healthy subjects before and aft er rapid ascent from sea level to 3658 m. Human acute mountain sickness was evaluated by the Lake Louise Score (LLS).Results:Nine of the seventeen parti cipants were diagnosed with AMS. On initi al exposure, there was a signifi cant decrease in vascular tone between subjects with and without AMS. Signifi cance was also found in the decrease of SpO2before and aft er rapid ascent but the diff erences between subjects with and without AMS did not reach signifi cance during the initi al phase. Conclusions: Vascular tone on initi al exposure in response to rapid ascent is a possible sign of suscepti bility to AMS.Conclusion:measurement of vascular tone using a wearable sensor throughout the acute phase response will provide numerical values of pathophysiology throughout the development of AMS.

vascular tone; refl ecti on index; acute mountain sickness; suscepti bility; photoplethysmographic sensor

Introduction

AMS is a pathological effect of high altitude on humans caused by acute exposure to low partial pressure of oxygen at high altitude. It commonly occurs above 2 400 meters [1, 2]. Ascent to high altitudes requires adaptation to a hypoxic and hypobaric environment, while failure to adapt results in AMS [3-6].

Pulse oximetry measuring SpO2has been a portable and useful tool for the diagnosis of AMS. However, there are still some cases in published literature reporting questionable results or failure in predicting AMS using pulse oximetry only [7-9]. In additional to SpO2, rapid ascent to altitude induces acute response in physiological variables including blood pressure (BP) and heart rate (HR) as well as vascular tone [10]. Previous studies indicated that changes in BP and endothelial function may be important factors in the pathogenesis of clinical syndromes occurring at high altitude [11-13], including AMS, high altitude cerebral oedema (HACE) and high altitude pulmonary oedema (HAPE). However, the published results of these studies in normotensive subjects regarding the response of BP to initial exposure to hypobaric hypoxia have also shown discrepancies. An increase in BP was reported by some investigators [12, 14-18] while a slight reduction [11, 19, 20] and no change in BP [21, 22] were also found by others. In this situation, more objective measures are needed to improve the assessment of physiological condition which is currently a limitation at altitude, particularly in remote settings with minimal detection ability. Hence, the importance of vascular tone gained our attention. Encouragingly, our preceding studies had already demonstrated that the eff ect of acute response to hypoxia in vascular tone had been represented by a temporary reduction in the photoplethysmography (PPG) derived refl ection index (RI) and AMS occurrence was in association with a relatively blunted and slow fall in vascular tone during acute phase [23, 24], which suggested that vascular response may be one of the determinants of susceptibility to AMS, at least a valuable addition to existing diagnosis of AMS.But a detailed tendency of RI has not been recorded due to limited time points of measurement included in previous studies at altitude. Th us, a new wearable device for continuous monitoring of RI was therefore proposed in which a modified version of software specifi cally for PPG dataset processing of long-term continuous recording would be embedded. In this pilot study, we evaluated the diff erence in changes of RI as well as relations with SpO2and RPP between subjects with and without AMS and discussed these fi ndings for further investigation.

Methods

Human hypoxic procedure

Seventeen male healthy lowlanders (median age 25.94±3.65 years, range 21–32) were enrolled in the study. Participants were prospectively recruited among offi ce workers screened free of medical illness planning a trip from sea level to Lhasa (3 658 m), Tibet by air. None of them had been to high altitude region over 2 400 meters within three months. Th is study was approved by the institutional review b oard of Beijing Genomics Institute. All subjects included in the study gave full informed consent to all procedures.

Detections of RI, SpO2, BP and HR

Arterial vascular tone was quantified by RI derived from digital volume pulse (DVP) obtained using PPG technique. Typical DVP waveform and the algorithm for the determination of RI are depicted in Figure 1. The equation for RI is expressed as: RI=b/aX100%. The PPG derived RI has been verified repeatedly and used as an indicator of activation of endothelial function as well as acute feedback responses to hypoxia which adequately increases oxygen supply by decreasing vascular tone and increasing vascular fl ow [13, 25, 26].

Four data points were obtained for the whole group including one point of baseline before ascent and three points during the fi rst three hours at altitude, e.g. before taking off at sea level for P0 as baseline, 30 min aft er landing at Lhasa airport for P1, 120 min aft er landing at hotel for P2, 180 min aft er landing at hotel for P3. All the measurements were performed aft er a rest no less than 10 min in the seated position. All subjects were nonsmokers. Furthermore, subjects avoided stimulants such as tea and coff ee for at least two hours prior to the measurements. During measurement, the subjects were told to calm down and breathe normally. In addition to RI and SpO, HR, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were also recorded in all subjects immediately aft er PPG measurement at each time point. Mean arterial pressure (MAP) was calculated from SBP and DBP as: MAP=DBP+1/3×(SBP-DBP) [27, 28]. The RPP, calculated as the product of HR and MAP, has been used as a rough estimate of myocardial oxygen demand [29].

AMS scores were derived from self-completed LLS Questionnaires at 6 hours and 18 hours separately aft er landing. Subjects were considered to have AMS if the score on a single questionnaire given was three or more, including headache, either 6 or 18 hours after arrival. Th e subjects were then divided into AMS group and no AMS group accordingly.

RI was detected using PPG technique with a fi nger clip sensor applied to right index fi nger of subjects. The sensor was driven by HC2180-D research platform (Comperson Biotechnology Co., Ltd., Beijing, China) in which the calculation of RI was conducted automatically pulse to pulse [23, 24, 30-33]. Th e data of the RI obtained from all single waveforms within each 3 min test was averaged to represent the value of the designed time point. Th e tendency of the value is the focus in this study as the fi rst three hours aft er rapid ascent may include major changes related to acute phase response.

During PPG measurement, SpO2was simultaneously recorded using a pulse oximeter MD300C2 (Beijing Choice Electronic Tech Co., Ltd., Beijing, China) in the opposite hand. Aft er 3 min testing, the subjects did not move or speak. HR, SBP and DBP were then measured in the brachial artery of the right arm using a digital sphygmomanometer UA-771 (A & D Electronics Co., Ltd., Shenzhen, China).

Statistical analysis

Data was expressed as mean ± SD, differences between AMS and no AMS as well as between time points were analyzed using T Test in and between groups respectively. Th e statistical analyses were performed using the SPSS 19.0. Data for RI during the first three hours after rapid ascent was also analyzed using repeated measures analysis of variance (ANOVA).

Results

Subjects involved in this study were divided into AMS group and no AMS group later at altitude in accordance with LLS. Th e data of participants which might influence arterial vascular tone and BP was analyzed and there was no significant differenceobserved in body height, weight, body mass index (BMI), and age between groups (Tab.1).

Fig. 1 The normalized waveform and its dPPG/dt (first derivative) obtained using PPG. Exhibiting: (A) A diastolic peak (B) A inflection point.

As shown in Table 2, SpO2decreased continuously in both AMS and no AMS group during the fi rst three hours after rapid ascent to altitude from P1 through P3. The difference of SpO2between AMS and no AMS was not obvious in the three time points at altitude and no signifi cance was found. In contrast, RPP increased continuously in both AMS and no AMS group during the first three hours after rapid ascent. The increase of RPP in subjects with AMS was relatively higher than subjects without AMS but the diff erence reached signifi cance only at P1 and P3 rather than P2.

In order to observe the tendency of the arterial vascular tone during initial phase response and capture temporal characterization of the diff erences, we compared and analyzed the parameters of the three time points after arrival separately. RI decreased gradually within three hours aft er arrival at altitude in all subjects either with or without AMS compared to the baseline obtained at sea level (from P0: 0.562±0.038 to P1: 0.522±0.042 P<0.01 by t-test, P2: 0.491±0.058 P<0.001 by t-test and P3: 0.479±0.055 P<0.001 by t-test). As shown in table 2, a more profound fall was associated with no AMS group and the diff erence of RI between AMS and no AMS was getting obvious from P1 to P2 and reached signifi cance at P3. At the time point of 30 min after arrival, RI was signifi cantly lower than baseline in both groups although almost no difference was found between AMS and no AMS (0.524±0.047 vs 0.520±0.038; P=0.834 by t-test). Comparatively at the time of 120 min aft er arrival, the RI decreased further in all subjects and the difference between AMS and no AMS was observed although not significant (0.512±0.051 vs 0.468±0.059; P=0.115 by t-test). At the time of 180 min after arrival, a further reduction in RI was found but more profound in subjects without AMS. The difference between AMS and no AMS then got more distinct and reached signifi cance (0.502±0.043 vs 0.454±0.048; P=0.044 by t-test). Furthermore, the analysis of the data with ANOVA confi rmed that RI was more signifi cant in subjects without AMS (n=8, F=37.71, P<0.001) than subjects with AMS (n=9, F=19.28, P<0.001) during the first three hours after rapid ascent. The typical wave trains of a subject at P0 and P3 are shown in Figure 2.

Tab. 1 Data of the study population.

In addition, there was no signifi cant diff erence in baseline values of RI between subjects with AMS and without AMS (0.559±0.036 vs 0.566±0.041; P=0.700 by t-test). In subjects without AMS, RI significantly decreased from baseline to a lower level (from 0.566±0.041 to 0.454±0.048, P<0.001) at the time of 180 min aft er landing. In subjects with AMS, RI also signifi cantly decreased from baseline to a lower level (from 0.559±0.036 to 0.502±0.043, P<0.01) but the absolute values of subjects with AMS remained relatively higher than subjects without AMS (0.502±0.043 vs 0.454±0.048, P<0.05). Th ese fi ndings indicated the potential value of RI in the pathophysiology of AMS.

Discussion

Based on the results of AMS score obtained aft erward from the subjects, 9 of 17 were diagnosed with AMS at either 6 or 18 hours after arrival at altitude. The subjects were then divided into two groups, AMS group and no AMS group. As seen in the result, both groups showed a gradual fall in RI during the initial phase response but the fall was more profound in no AMS as documented in our previous studies [23, 24]. However, the diff erences between AMS and no AMS in SpO2and RPP showed uncertainty.

During the 3 min period of continuous monitoring, although transient lower value of RI was observed in some subjects with AMS, the average value within the period was in general higher than no AMS. Even though only a small group of samples were included in this study, the result confi rmed ourprevious fi nding that subjects without AMS was significantly lower in RI than those with AMS during the initial phase of response [23, 24]. Meanwhile we observed a less signifi cant variation of RI in AMS as represented by a relatively lower F value. It also indicated that the transient value of RI may not be reliable in diagnosis of AMS. With such a small samples, we were not able to indicate the role of this feature a possible sign of responsive instability in this pilot study, however this result suggested that RI had been implicated in the pathophysiology of AMS and more featured responses might be revealed with the dataset obtained from continuous monitoring of variations in RI.

Tab. 2 Values of RPP, SpO2and RI at altitude of 3 658 min.

In 2012, C.J. Boos et al. reported that acute hypoxia decreased RI significantly (baseline vs. 150 minutes) in a hypobaric chamber at the simulated altitude equivalent to 4 800 m (15 748 ft ) which was in general consistent with our observations of RI at altitude [11]. Th e fall in RI was smaller in our result at altitude of 3658m indicating that the amplitude of initial response in vascular tone might be altitude dependent. The higher ascend, the more response induced. However, an important distinction is that the aforementioned study was performed in a hypobaric chamber in which the barometric pressure was reduced evenly at a rate of 1 219 m (4 000 ft) per minute until a final simulated altitude reached. The situation of traveling to high altitude region by airline could be diff erent and more factors might be included, such as physical and sympathetic activities more or less. In this study, we had taken these into consideration in our method. A resting time no less than 10 min before test was required to minimize the eff ects other than altitude as we aimed to identify the differences in RI specifically between AMS and no AMS.

Although a small sample size is involved in this study, our data appears to suggest that acute hypoxia induced by rapid ascent leads to differential effects along the vascular tone which may be related to susceptibility of AMS, e.g. no change or smaller fall in RI during initial phase response to acute hypoxia was in certain degree associated with subjects prone to AMS as per results of LLS Questionnaires obtained aft erward, meanwhile greater fall in RI during initial phase response was associated with AMS resistant subjects. In addition to the fall in RI which has been used as an indicator of reciprocal change in endothelial function, the amplitude of increase in RPP, an indicator of oxygen demand, was also noted in the results. Among 10 subjects exhibited relatively greater fall in RI, two of them were also diagnosed with AMS and they were found eventually in association with more profound increase in RPP at altitude. Th is would be partly explained as an insuffi cient accommodative response in endothelial function to a greater myocardial oxygen demand following acute hypoxia and the relationship between RI and RPP might be a factor of susceptibility as well.

In future work, we will record the variation and tendency in RI of longer period and explore its role in susceptibility to AMS, as well as onset and severity. The impact of physical or sympathetic activities which is also critical in development of AMS will be evaluated separately. Therefore, the development of a new wearable sensor that can record PPG derived RI all day and night in a person’s natural environment is proposed. Compared to other non-invasive techniques for the determination of vascular tone including applanation tonometry or ultrasound, movements of the subject or changes in pressure applied to the skin are of minor importance using PPG technique [34]. With the RI algorithm embedded MCU integrated within a pulse oximetry, the variation of RI throughout the acute phase response will be fully recorded and the big data will provide preclinical and clinical evidence forpathophysiology of AMS.

Fig. 2 The typical wave trains of a subject. (A) at P0 and (B) at P3.

In summary, vascular tone on initial exposure in response to rapid ascent is a possible sign of susceptibility to AMS and playing a role in the pathophysiology of AMS. Continuous measurement of vascular tone throughout the acute phase response will provide numerical values of pathophysiology throughout development of AMS as well as valuable addition to existing diagnosis of AMS, thus the development of a new wearable sensor that can record both SpO2and PPG derived RI continuously is strongly recommended.

Acknowledgements

The study was supported by a grant from the International Science & Technology Cooperation Program of China (2010DFB32940).

Declaration of Interest

Th e authors report no confl icts of interests.

Reference

1. Maggiorini M, Buhler B, Walter M, et al. Prevalence of acute mountain sickness in the Swiss Alps[J]. BMJ, 1990, 301(6756): 853–855.

2. Muza SR, Beidleman BA, Fulco CS. Altitude preexposure recommendations for inducing acclimatization[J]. High Alt Med Biol, 2010, 11(2): 87-92.

3. Hackett PH, Roach RC. High-altitude illness[J]. New Engl J Med, 2001, 345(2): 107-114.

4. Roach RC, Hackett PH. Frontiers of hypoxia research: acute mountain sickness[J]. J Exp Biol, 2001, 204(Pt 18): 3161-3170.

5. Imray C, Wright A, Subudhi A, et al. Acute mountain sickness: pathophysiology, prevention, and treatment[J]. Prog Cardiovasc Dis, 2010, 52(6): 467-484.

6. Gomersall T. Acute mountain sickness[J]. Scottish Univ Med J, 2012, 1(1): 98-103.

7. Chen HC, Lin WL, Wu JY, et al. Change in oxygen saturation does not predict acute mountain sickness on jade mountain[J]. Wilderness Environ Med, 2012, 23(2): 122-127.

8. Leichtfried V, Basic D, Burtscher M, et al. Diagnosis and prediction of the occurrence of acute mountain sickness measuring oxygen saturation—independent of absolute altitude[J]? Sleep Breath, 2016, 20(1): 435-442.

9. Wagner DR, Knott JR, Fry JP. Oximetry fails to predict acute mountain sickness or summit success during a rapid ascent to 5 640 meters[J]. Wilderness Environ Med, 2012, 23(2): 114-121.

10. Kanstrup IL, Poulsen TD, Hansen JM, et al. Blood pressure and plasma catecholamines in acute and prolonged hypoxia: effects of local hypothermia[J].J Appl Physiol, 1999, 87(6): 2053-2058.

11. Boos CJ, Hodkinson P, Mellor A, et al. The effects of acute hypobaric hypoxia on arterial stiffness and endothelial function and its relationship to changes in pulmonary artery pressure and left ventricular diastolic function[J]. High Alt Med Biol, 2012, 13(2): 105-111.

12. Rhodes HL, Chesterman K, Chan CW, et al. Systemic blood pressure, arterial stiff ness and pulse waveform analysis at altitude[J]. J R Army Med Corps, 2011, 157(1), 110-113.

13. Morganti A, Giussani M, Sala C, et al. Effects of exposure to high altitude on plasma endothelin-1 levels in normal subjects[J]. J Hyperten, 1995, 13(8): 859-865.

14. Barthelemy JC, Lacour JR, Roche F, et al. Elevated nocturnal blood pressure assessed by ambulatory automatic monitoring during a stay at high altitude[J]. Eur J Appl Physiol Occup Physiol, 1995, 70(3): 258-262.

15. Basu CK, Gautam RK, Sharma RP, et al. Metabolic responses during initial days of altitude acclimatization in the eastern Himalayas[J]. Int J Biometeorol, 1996, 39(3): 133-138.

16. Bernardi L, Passino C, Spadacini G, et al. Cardiovascular autonomic modulation and activity of carotid baroreceptors at altitude[J]. Clin Sci (Lond), 1998, 95(5): 565-573.

17. Drinkwater BL, Kramar PO, Bedi JF, et al. Women at altitude: cardiovascular responses to hypoxia[J]. Aviat Space Environ Med, 1982, 53(5): 472-477.

18. Rupwate RU, Chitaley M, Kamat SR. Cardiopulmonary functional changes in acute acclimatisation to high altitude in mountaineers[J]. Eur J Epidemiol, 1990, 6(3): 266-272.

19. Korner PI. Circulation adaptation in hypoxia[J]. Physiol Rev, 1959, 39(4): 687-730.

20. Stenberg J, Ekblom B, Messin R. Hemodynamic response to work at simulated altitude, 4 000 m[J]. J Appl Physiol, 1966, 21(5): 1589-1594.

21. Aigner A, Berghold F, Muss N. Investigations on the cardiovascular system at altitudes up to a height of 7 800 meters[J]. Z Kardiol, 1980, 69(9): 604-610.

To page 528

doi 10.13459/j.cnki.cjap.2016.06.003

Xiao-ming LUO, Principal Investigator, Tibet Branch, Beijing Genomics Institute, 189 Jinzhu Road, Lhasa 850032, China. Tel/Fax：86-891-6618661; E-mail: luo. xiaomin@139.com.

2016-07-23; accepted 2016-10-24