Issues in Continuous 24-h Core Body Temperature Monitoring in Humans Using an Ingestible Capsule Telemetric Sensor

Cathriona R. Monnard, Elie-Jacques Fares, Julie Calonne, Jennifer L. Miles-Chan, Jean-Pierre Montani, Dominique Durrer, Yves Schutz, Abdul G. Dulloo

There is increasing interest in the use of pill-sized ingestible capsule telemetric sensors for assessing core body temperature (Tc) as a potential indicator of variability in metabolic efficiency and thrifty metabolic traits. The aim of this study was to investigate the feasibility and accuracy of measuring Tc using the CorTemp(®) system. Tc was measured over an average of 20 h in 27 human subjects, with measurements of energy expenditure made in the overnight fasted state at rest, during standardized low-intensity physical activity and after a 600 kcal mixed meal. Validation of accuracy of the capsule sensors was made ex vivo against mercury and electronic thermometers across the physiological range (35-40°C) in morning and afternoon of 2 or 3 consecutive days. Comparisons between capsule sensors and thermometers were made using Bland-Altman analysis. Systematic bias, error, and temperature drift over time were assessed. The circadian Tc profile classically reported in free-living humans was confirmed. Significant increases in Tc (+0.2°C) were found in response to low-power cycling at 40-50 W (~3-4 METs), but no changes in Tc were detectable during low-level isometric leg press exercise (<2 METs) or during the peak postprandial thermogenesis induced by the 600 kcal meal. Issues of particular interest include fast "turbo" gut transit with expulsion time of <15 h after capsule ingestion in one out of every five subjects and sudden erratic readings in teletransmission of Tc. Furthermore, ex vivo validation revealed a substantial mean bias (exceeding ±0.5°C) between the Tc capsule readings and mercury or electronic thermometers in half of the capsules. When examined over 2 or 3 days, the initial bias (small or large) drifted in excess of ±0.5°C in one out of every four capsules. Since Tc is regulated within a very narrow range in the healthy homeotherm's body (within 1°C), physiological investigations of Tc require great accuracy and precision (better than 0.1°C). Although ingestible capsule methodology appears of great interest for non-invasively monitoring the transit gut temperature, new technology requires a reduction in the inherent error of measurement and elimination of temperature drift and warrants more interlaboratory investigation on the above factors.