The Dynamic Response of Sweat Chloride to Changes in Exercise Load Measured by a Wearable Sweat Sensor

Abstract

Wearable sensors enable the monitoring of an individual's sweat composition in real time. in this work, we recorded real-time sweat chloride concentration for 12 healthy subjects in three different protocols involving step changes in exercise load and compared the results to laboratory-based analysis. the sensor results reflected the changes in exercise load in real time. On increasing the exercise load from 100 W to 200 W the sweat chloride concentration increased from 12.0 ± 5.9 to 31.4 ± 16 mM (mean ± SD). On decreasing the load from 200 W to 100 W, the sweat chloride concentration decreased from 27.7 ± 10.5 to 14.8 ± 8.1 mM. The half-time associated with the change in sweat chloride, defined as the time at which the concentration reached half of the overall change, was about 6 minutes. While the changes in sweat chloride were statistically significant, there was no correlation with changes in sweat rate or other physiological parameters, which we attribute to intra-individual variation (SD = 1.6-8.1 mM). The response to exercise-induced sweating was significantly different to chemically-induced sweating where the sweat chloride concentration was almost independent of sweat rate. We speculate that this difference is related to changes in the open probability of the CFTR channel during exercise, resulting in a decrease in reabsorption efficiency at higher sweat rates. The most effective method of thermal regulation during exercise is sweating, however, excessive water and electrolyte losses can contribute to dehydration and electrolyte imbalances 1-4. Measurement of electrolyte loss during sweating usually involves absorption patches, collection devices (e.g. Macroduct), or plastic bags to collect samples to be sent to a laboratory for analysis 4-8. Since sweat collection times are generally in the range of 5 to 90 minutes, depending on the study 8 , laboratory analysis has limited capability for elucidating real-time, dynamic changes during exercise. Recent advances in wearable sweat sensors 9-13 have enabled real-time measurement of water and electrolyte loss during exercise. In addition, continuous measurement enables detection of transient changes that could be masked by the sweat collection time in conventional measurements. Finally, wearable sensors require very small volumes for measurement in comparison to standard laboratory-based tests. For example a sweat test for diagnosis of cystic fibrosis requires a minimum sweat volume of 15 µL 14 whereas a sweat sensor can detect sweat volumes less than 1 μL 15. Electrolytes regulate fluid balance in plasma and tissues and are involved in cell signaling 16 , and electrolyte imbalance can lead to a wide range of medical conditions 17,18. Chloride and sodium are the most abundant electrolytes and their concentrations are closely correlated 19. In this work, to assess the feasibility of using a wearable sensor to evaluate transient changes in real-time, we performed a systematic study of the dynamic response of sweat chloride to step changes in exercise load. We p...

Electron transfer studies of a conventional redox probe in human sweat and saliva bio-mimicking conditions

“…There are a numerous distributions that propose utilizing human sweat and salivation rather than blood, for assessing metabolites present in sweat and salivation towards the development of electrochemical biosensors 17 – 20 . Many literature reports that deal with utilizing change in viscosity 24 , pH 25 , measurement of stress biomarkers like Cortisol 26 and concentration of chloride ion 27 in human body fluids are available. These studies clearly demonstrate the possibility of using sweat and saliva for health care diagnostics.…”

A Capacitive Sweat Rate Sensor for Continuous and Real-Time Monitoring of Sweat Loss

“…In the example shown here (Figure B), the volume of the microfluidic channel was 31.5 μL: total length ( l ) = 350 mm, width ( w ) = 1 mm, and thickness ( d m ) = 90 μm. Typical sweat rates for healthy individuals vary from 0.1 to 1 μL min –1 cm –2 ; therefore, with these dimensions and a sweat collection area of 0.5 cm 2 , the sensor can be used for more than 60 min. Incorporating a dye into the channels allows sweat filling of the sensor to be easily visualized (Figure C).…”

“…Monitoring of electrolyte loss requires measurement of electrolyte concentration (either Cl – or Na + ) and sweat rate. Wearable sensors for the measurement of sweat concentration are relatively advanced, − and hence the development of wearable sweat rate sensors has the potential to enable integrated measurement of electrolyte loss in real time.…”

Fluidic Patch Device to Sample Sweat for Accurate Measurement of Sweat Rate and Chemical Composition: A Proof-of-Concept Study

“…Sweat sodium and chloride are sweat rate-dependent as they are reabsorbed in the straight duct of the sweat gland before appearing on the skin surface . When the sweat secretion rate exceeds the reabsorption rate, the concentrations of these ions increase linearly with the sweat rate. , To evaluate whether these well-known relations could be determined using the proposed fluidic patch, we examined the relationship between the sweat rate, sweat sodium, and chloride (Figure ). A strong relationship was observed ( R 2 = 0.76 and 0.66, respectively).…”

“…Thus, the filling time of each reservoir often exceeds 10 min, especially at low sweat rates (<0.5 mg/(cm 2 ·min)), and in some cases, dependent on varying sweat rates. Another new trend has been sweat sampling in wearable sensors. − However, when minute-order sampling resolution is required, many of them risk the contamination of “old” and “newly” secreted sweat around the sensor unit and the unwanted averaging effect of sweat solute concentrations, as again the fluidics is driven by the slow sweat gland pumping. In summary, there is still the need for devices that can accurately and easily collect sweat for analysis, which can be studied for real-time physiological investigations to advance our knowledge of sudomotor function.…”