Principal Component Analysis and Risk Factors for Acute Mountain Sickness upon Acute Exposure at 3700 m

Abstract

Objective: We aimed to describe the heterogeneity in the clinical presentation of acute mountain sickness (AMS) and to identify its primary risk factors.

Methods: The participants (n = 163) received case report form questionnaires, and their heart rate (HR), oxygen saturation (SpO2), echocardiographic and transcranial Doppler variables, ability to perform mental and physical work, mood and psychological factors were assessed within 18 to 22 hours after arriving at 3700 m from sea level (500 m) by plane. First, we examined the differences in all variables between the AMS-positive and the AMS-negative groups. Second, an adjusted regression analysis was performed after correlation and principal component analyses.

Results: The AMS patients had a higher diastolic vertebral artery velocity (Vd; p = 0.018), a higher HR (p = 0.006) and a lower SpO2. The AMS subjects also experienced poorer sleep quality, as quantified using the Athens Insomnia Scale (AIS). Moreover, the AMS population exhibited more negative mood states, including anxiety, depression, hostility, fatigue and confusion. Five principal components focused on diverse aspects were also found to be significant. Additionally, more advanced age (p = 0.007), a higher HR (p = 0.034), a higher Vd (p = 0.014), a higher AIS score (p = 0.030), a decreased pursuit aiming capacity (p = 0.035) and decreased vigor (p = 0.015) were risk factors for AMS.

Conclusions: Mood states play critical roles in the development of AMS. Furthermore, an elevated HR and Vd, advanced age, elevated AIS sores, insufficient vigor and decreased mental work capacity are independent risk factors for AMS.

Fig 1. Flow diagram of the statistical analyses.

An adjusted logistic regression was performed to identify independent risk factors for AMS (i.e., factors for which the p value was less than 0.05 in univariate analyses and the correlation coefficient was greater than 0.5 in principal component analysis).

Fig 2. Mood states of the 150 subjects.

Fig 3. Distributions of the cognitive functions.