An Individualized, Perception-Based Protocol to Investigate Human Physiological Responses to Cooling

Abstract

Cold exposure, a known stimulant of the thermogenic effects of brown adipose tissue (BAT), is the most widely used method to study BAT physiology in adult humans. Recently, individualized cooling has been recommended to standardize the physiological cold stress applied across participants, but critical experimental details remain unclear. The purpose of this work was to develop a detailed methodology for an individualized, perception-based protocol to investigate human physiological responses to cooling. Participants were wrapped in two water-circulating blankets and fitted with skin temperature probes to estimate BAT activity and peripheral vasoconstriction. We created a thermoesthesia graphical user interface (tGUI) to continuously record the subject's perception of cooling and shivering status during the cooling protocol. The protocol began with a 15 min thermoneutral phase followed by a series of 10 min cooling phases and concluded when sustained shivering (>1 min duration) occurred. Researchers used perception of cooling feedback (tGUI ratings) to manually adjust and personalize the water temperature at each cooling phase. Blanket water temperatures were recorded continuously during the protocol. Twelve volunteers (ages: 26.2 ± 1.4 years; 25% female) completed a feasibility study to evaluate the proposed protocol. Water temperature, perception of cooling, and shivering varied considerably across participants in response to cooling. Mean clavicle skin temperature, a surrogate measure of BAT activity, decreased (-0.99°C, 95% CI: -1.7 to -0.25°C, P = 0.16) after the cooling protocol, but an increase in supraclavicular skin temperature was observed in 4 participants. A strong positive correlation was also found between thermoesthesia and peripheral vasoconstriction (ρ = 0.84, P < 0.001). The proposed individualized, perception-based protocol therefore has potential to investigate the physiological responses to cold stress applied across populations with varying age, sex, body composition, and cold sensitivity characteristics.

Keywords: brown adipose tissue; cold exposure; humans; shivering; supraclavicular skin temperature; thermoregulation; vasoconstriction.

Figure 1

Participant's setup used for an individualized, perception-based cooling protocol (A). Two water-circulating blankets connected to a Blanketrol® III hyper-hypothermia system are secured around the participant. Diagram of the skin temperature probe locations (human body outline adapted from a free graphics library: http://clipart-library.com) (B). Skin temperature was measured in the supraclavicular region (red circle), on the anterior portion of the forearm (green circle), and on the distal end of the middle finger (gray circle). A laptop computer logs water temperature and runs a thermoesthesia graphical user interface (tGUI) program. The tGUI records feedback from an external keypad (C) attached to the participant's right hand (positioned underneath the blankets during cooling) and updates a visual display projected above the patient bed in real-time. Schematic of the perception-based cooling protocol water temperature adjustment guide (blue and red text) overlay on the tGUI visual interface (D). During the protocol, the participant uses two buttons to move the green slider along the continuum of temperature ratings. The slider's position determines how to alter the current water temperature set point (TN during thermoneutral (red text) or T during cooling (blue text) phases; see section Individualized, Perception-Based Cooling Protocol for details). A third button can be pressed to move the orange circle and indicate if shivering is active (“Yes”) or inactive (“No”).

Figure 2

Water temperature (°C; A) and subjective perception of cooling ratings (arbitrary units; B) exhibit a wide range of variability between subjects at three physiological and temporally distinct markers: vasoconstriction index (>4°C gradient between forearm and finger skin temperature; blue circles), onset of shivering (first self-reported shiver event; green circles), and sustained shivering (self-reported shiver event with a duration >1 min; red circles) in the perception-based cooling protocol. The centerline in each box indicates the mean value, and the top and bottom of the box mark the 95% confidence intervals. Tukey's post-hoc pairwise tests were performed following a linear mixed-effects analysis with repeated measures to assess statistical comparisons.

Figure 3

Self-reported shiver patterns (onset (red), 50% of total shiver events (green), and sustained shivering (>1 min duration; blue)) differed across participants during the individualized, perception-based cooling protocol.

Figure 4

Clavicle (°C; A), forearm (°C; B), finger (°C; C), and peripheral vasoconstriction (forearm–finger, °C; D) skin temperatures correlated with subjective perception of cooling ratings. Relative clavicle, forearm, and finger skin temperatures are expressed as the change (Δ) in temperature from the end of the thermoneutral phase. Least squares best-fit lines are overlaid on the temperature data, which were linearly interpolated and averaged among participants. Small upward fluctuations in the temperature gradient (most notable in the finger skin temperature) were the result of two volunteers indicating a warmer perception rating when water temperature was adjusted (i.e., a brief stop of circulating cold water). Correlation (ρ) between temperature and perception was evaluated with the Spearman rank test.

Figure 5

Example of the water temperature set points (°C, blue), subjective perception of cooling ratings (arbitrary units, orange), and self-reported shiver events (gray dots) (A) and clavicle (°C, red), forearm (°C, green line), and finger (°C, gray line) skin temperature gradients (B) recorded from a healthy, young male during a perception-based cooling protocol. During the protocol, the set point of the water temperature (blue line) was adjusted at the end of each 10-min cooling phase (vertical gray lines) according to the subject's thermal perception rating (orange line; see section Individualized, Perception-Based Cooling Protocol for details). The protocol terminated when the subject reported sustained shivering (>1 min in duration; gray dots).

Figure 6

Similarity fractions (arbitrary units; see equations 2 and 3 for details) were calculated to evaluate the utility of surface electromyography (EMG; red) recordings of trapezius and sternocleidomastoid muscle activity and subjective events captured with the thermoesthesia graphical user interface (tGUI; blue) to quantify shivering (A). The centerline in each box indicates the mean value, and the top and bottom of the box mark the 95% confidence intervals. Individual examples highlight good agreement between EMG and tGUI shiver events number and timing (B) and illustrate instances of poor specificity (C) and sensitivity (D) of EMG to measure whole body shivering. Brief shiver events may be difficult to visualize on plots (B–D) due to the resolution of the time scale.