Accuracy and Reliability of Pulse O2 Saturation Measured by a Wrist-worn Oximeter

Abstract

This study aims to evaluate the accuracy of the Garmin Forerunner 245 heart rate (HR) and pulse O2 saturation (SpO2) sensors compared with electrocardiogram and medical oximeter, from sea level to high altitude. Ten healthy subjects underwent five tests in normoxia and hypoxia (simulated altitudes from 3000 to 5500 m), consisting in a 5-min rest phase, followed by 5-min of mild exercise. Absolute error (±10 bpm for HR and ±3% for SpO2, around criterion) and intraclass correlations (ICC) were calculated. Error rates for HR remained under 10%, except at 3000 m, and ICCs evidenced a good reliability between Garmin and criterion. Overall SpO2 was higher than criterion (P<0.001) with a >50% error rate (>80% above 4800 m), and a poor reliability with criterion. The Garmin device displayed acceptable HR data at rest and exercise for all altitudes, but failed to provide trustworthy SpO2 values, especially at high altitude, where a pronounced arterial O2 desaturation could lead to acute mountain sickness in hypoxia-sensitive subjects, and its life-threatening complications; moreover, readings of overestimated SpO2 values might induce trekkers into further hazardous behavior by pursuing an ascent while being already at risk. Therefore, its use to assess SpO2 should be proscribed in altitude for acclimatization evaluation.

Validity of Peripheral Oxygen Saturation Measurements with the Garmin Fēnix® 5X Plus Wearable Device at 4559 m

“…SpO 2 readings from wrist-worn devices might be influenced by skin tone, and this factor was not assessed in our study. However, there was no effect of skin tone on any SpO 2 variables in the study of Hermand et al [10]. Also, future firmware versions released by Garmin could change the accuracy of SpO 2 measurements, and thus affect the conclusions of this study.…”

“…However, in Lauterbach's study, only Bland-Altman analysis was used to assess validity. More recently, Hermand et al evaluated the accuracy of the Garmin Forerunner 245 SpO 2 sensor in 10 healthy participants at simulated altitudes from 3000-5500 m [10] and applied more comprehensive statistical methods to assess device validity, including ICCs. The device failed to provide trustworthy SpO 2 values, yielding an ICC of less than 0.280 over all the altitudes studied.…”

“…Several recent commercially available smartwatches, including the Garmin Fēnix ® 5X Plus (GAR), include sensors for SpO 2 measurement. Not surprisingly, they are increasingly being used by mountaineers with the intention of monitoring their risk of altitude illness [10], even though this usage has never been rigorously tested or confirmed.…”

“…While several studies have already investigated the validity and reliability of smartwatchderived physiological biomarkers, such as heart rate and energy expenditure [11,12], data on the validity and reliability of SpO 2 measurements are scarce [10,13,14]. Of note, there is a complete lack of data obtained in a high-altitude field setting.…”

Smart Electrically Assisted Bicycles as Health Monitoring Systems: A Review

“…Several commercial wrist-worn products measure pulse oximetry using photoplethysmography while maintaining an affordable cost. Few researchers have investigated the accuracy of such devices for dynamic cycling, but recent studies from Schiefer et al [105] and Hermand et al [106] invite a careful consideration of these devices, especially when the pulse oximetry variable is of interest.…”