Eur J Appl Physiol (2012) 112:20252034 DOI 10.

1007/s00421-011-2173-z

ORIGINAL ARTICLE

Development of a perceptual hyperthermia index to evaluate heat strain during treadmill exercise
Michael Gallagher Jr Robert J. Robertson Fredric L. Goss Elizabeth F. Nagle-Stilley Mark A. Schafer Joe Suyama David Hostler

Received: 25 March 2011 / Accepted: 8 September 2011 / Published online: 23 September 2011 Springer-Verlag 2011

Abstract Fire suppression and rescue is a physiologically demanding occupation due to extreme external heat as well as the physical and thermal burden of the protective garments. These conditions challenge body temperature homeostasis and results in heat stress. Accurate eld assessment of core temperature is complex and unreliable. The present investigation developed a perceptually based hyperthermia metric to measure physiologic exertional heat strain during treadmill exercise. Sixty-ve (28.9 6.8 years) female (n = 11) and male (n = 54) reghters and non-reghting volunteers participated in four related exertional heat stress investigations performing treadmill exercise in a heated room while wearing thermal protective clothing. Body core temperature, perceived exertion, and

thermal sensation were assessed at baseline, 20-mins exercise, and at termination. Perceived exertion increased from baseline (0.24 0.42) to termination (7.43 1.86). Thermal sensation increased from baseline (1.78 0.77) to termination (4.50 0.68). Perceived exertion and thermal sensation were measured concurrently with body core temperature to develop a two-dimensional graphical representation of three exertional heat strain zones representative of a range of mean body core temperature responses such that low risk (green) incorporated 36.037.4C, moderate risk (yellow) incorporated 37.537.9C, and high risk (red) incorporated 38.0 to greater than 40.5C. The perceptual hyperthermia index (PHI) may provide a quick and easy momentary assessment of the level of risk for exertional heat stress for reghters engaged in re suppression that may be benecial in high-risk environments that threaten the lives of reghters. Keywords Fireghter Perceptual heat strain Hyperthermia Thermal protective clothing Thermoregulation

Communicated by Narihiko Kondo. This work has been presented at the American College of Sports Medicine Annual Meeting (2010). M. Gallagher Jr (&) Department of Kinesiology and Physical Education, University of Central Arkansas, 201 Donaghey Ave, Conway, AR 72035, USA e-mail: GallagherM@uca.edu R. J. Robertson F. L. Goss E. F. Nagle-Stilley Department of Health and Physical Activity, University of Pittsburgh, Pittsburgh, PA, USA M. A. Schafer Department of Kinesiology, Recreation and Sport, Western Kentucky University, Bowling Green, KY, USA J. Suyama D. Hostler Department of Emergency Medicine, Emergency Responder Human Performance Lab, University of Pittsburgh, Pittsburgh, PA, USA

Introduction Fire suppression and rescue is physiologically demanding on reghters due to the extreme hot environmental conditions as well as the physical and thermal burden of the reghting gear. The hot environment imposes a challenge on the homeostasis of body temperature inducing hyperthermia (Bilzon et al. 2001; Cheung and McLellan 1998; Cheung 2007, Hostler et al. 2010a, b; McLellan and Cheung 2000; Northington et al. 2007). Additionally, the increased metabolic demand associated with physical movement while wearing reghting thermal protective clothing (TPC) further increases body core temperature.

123

2026

Eur J Appl Physiol (2012) 112:20252034

In the absence of direct thermometry, the assessment of hyperthermia is complex, involving both physiological and perceptual strain indicators. Heat stress indices assess the combined contributions of metabolic cost, environmental factors, and clothing requirements and the physiological response resulting from exertional heat stress. A number of indices have been constructed to measure the degree of heat stress imposed on an individual. These heat stress indices may be categorized into three groups: rational, empirical, and direct (Epstein and Moran 2006; McPherson 1992; NIOSH 1986; Parsons 2006). Rational indices are based on a heat balance equation that includes heat storage rate, metabolic rate, external work rate, radiant heat exchange, convective heat exchange, respiratory heat exchange, and evaporative heat loss. Although rational indices are considered the most comprehensive of three categories of indices, a lack of technology prohibits the assessment of the integration of all the environmental and behavioral variables (Epstein and Moran 2006). The heat stress index (HSI; Belding and Hatch 1955) and the physiological heat strain prediction model by McPherson (1992) are examples of rational indices. Direct indices, such as the wet-bulb globe temperature (WGBT; Yaglou and Minard 1957) and the discomfort index (DI; Tennebaum et al. 1961) are based on actual measures of environmental variables to quantify the degree of heat stress imposed on an individual in a given set of environmental conditions. Many health and safety standards employ this direct approach but require meteorological instruments and only simulates the heat strain (Epstein and Moran 2006). Last, empirical indices are based on objective and subjective strain. One such example of an empirical index based on objective measurements is the physiological strain index (PhSI), developed by Moran et al. (1998, 2000). This index was intended to address measurement limitations associated with direct environmental indices. The empirical index incorporates heart rate and body core temperature response to a thermal load as surrogate measures of heat stress imposed on the individual but requires direct contact with the individual or sensitive equipment not suitable for extreme environments. As a possible solution to the disadvantages of the current indices, Cheung (2007) has proposed that occupational exposure standards should not only use physiological outcomes, but also thermal perceptions reecting the behavioral and exercise response to heat stress. The majority of these safety standards and indices of thermal strain do not incorporate perceptual responses to heat stress and hyperthermia. One possible method for monitoring the risk of heat-related injury involves the use of ratings of perceived exertion (RPE) and ratings of thermal sensation (RTS) derived from category metrics incorporated into a perceptual strain index (PeSI; Tikuisis et al. 2002). The

application of a perceptual heat strain index may provide a measure of relative risk for individuals performing exertion while wearing TPC and may provide a link in bridging the gap between the ease and applicability of direct indices and the comprehensiveness of empirical indices as suggested by Epstein and Moran (2006). Structured rest periods, often referred to as emergency incident rehabilitation or reground rehab, are provided during prolonged incidents to provide an opportunity for partial recovery before returning to re suppression. During these periods, heart rate and blood pressure are commonly monitored but direct measurement of core body temperature is difcult (Pryor et al. 2011). A perceptual heat strain index could be applied immediately following reground activities and during rest periods to estimate recovery time and gauge physiological readiness to return to re suppression. Therefore, the next logical step in hyperthermia risk assessment consequent to reghting while wearing TPC was to develop an exertional heat strain metric. This novel metric incorporated components of the PeSI (Tikuisis et al. 2002) into a graphical presentation providing a fast and practical observation tool to assess exertional heat strain during actual re scenarios. The PHI metric was based on a previous investigation examining the effects of hyperhydration on emergency responders performing treadmill exercise while wearing chemical resistant personal protective equipment (PPE). The ndings demonstrated a moderately strong correlation between PeSI and body core temperature under varying environmental heat loads (Hostler et al. 2009).

Methods The PHI was developed using RPE and RTS data derived from the following four interrelated investigations involving exercise-induced thermal stress: (a) FIRE I (a laboratory study on rehydration; Hostler et al. 2010a), (b) FIRE II (a laboratory study of cooling; Hostler et al. 2010b), (c) COG I (a laboratory study of cognitive function), and (d) COG II (a laboratory study of cognitive recovery). Each of the four related investigations employed partially overlapping participant cohorts performing a treadmill exercise forcing function under hot ambient conditions. During exercise, participants wore the same type of TPC. Study design and population A total of 65 female (n = 11) and male (n = 54) adults aged 1944 years who comprised the experimental cohorts for the four investigations were participants in this investigation (Table 1).

123

Eur J Appl Physiol (2012) 112:20252034 Table 1 Descriptive characteristics of the combined sample and the four interrelated investigations Combined (n = 65) Age (years) Height (cm) Weight (kg) VO2max/peak (ml kg BMI (kg m-2)
-1

2027

FIRE I (n = 18) 29.9 8.7 173.1 9.6 89.7 15.8 37.8 3.9 20.4 5.2 30.1 5.4

FIRE II (n = 18) 29.8 7.4 172.1 9.3 75.4 14.4 38.1 6.9 15.1 25.1 25.1 4.0

COG I (n = 10) 28.1 5.3 175.1 6.4 73.2 8.0 54.5 6.7 24.0 3.4

COG II (n = 19) 27.5 4.6 175.1 9.5 73.1 16.2 49.3 7.9 23.1 3.7

28.9 6.8 173.7 9.0 78.3 16.0 min )

-1

43.8 9.3 25.7 5.1

Body Composition (%Fat)

Individuals provided informed consent and completed a survey of demographic information, medical history, and physical activity level. Potential participants underwent a physician administered physical examination including a resting 12-lead ECG and a 12-lead ECG recorded during a graded exercise test. Both ECGs were interpreted by a cardiologist. Individuals with orthopedic, cardiovascular, and/or metabolic contraindications to exercise participation (i.e. coronary artery disease, prior myocardial infarction, peripheral vascular disease, hypertension, chronic obstructive pulmonary disease, and diabetes mellitus) or who were pregnant were excluded from participation. Additionally, individuals taking medications that were expected to alter physiological response to exercise (i.e. beta blockers) were excluded. The University of Pittsburgh Institutional Review Board (Pittsburgh, PA, USA) approved all procedures associated with the investigation. Baseline session Descriptive characteristics of the participants were recorded including body height (cm), body mass (kg), and body fat (%). Body height was determined using a Healthometer Scale with an attached stadiometer (Sunbeam Products, Inc., Boca Raton, FL, USA). Fat-free mass was measured using the Jackson and Pollack 3 skinfold site equation (Jackson and Pollack 1985). Skinfold measurements were obtained using Lange skinfold calipers. Aerobic tness was assessed using a graded exercise test on either a cycle ergometer (FIRE I, FIRE II) or a treadmill (COG I, COG II). Oxygen consumption (VO2; liters per minute; STPD) was measured using an open-circuit respiratory-metabolic system (True Max 2400, Parvo Medics, Salt Lake City, UT, USA) during each minute of the exercise protocol. Heart rate (HR; Polar Electro, Finland) was measured from 45 to 60 s of each minute of the exercise protocol. VO2max/peak was established as the highest measured value that occurred when the participant was no longer able to perform the exercise test due to fatigue. Secondary criteria for peak/maximal measurement included a VO2 plateau (i.e. a change \3.5 ml kg-1 min-1 between contiguous stages at maximal intensity), a RER C1.2, and a peak

exercise heart rate 5 beats min-1 of age-predicted maximal heart rate. Exercise session The evening prior to each exercise testing session, participants were instructed to refrain from alcohol, caffeine, and nicotine use for 12 h prior to the exercise session, to drink approximately 600 mL of water the night before the scheduled exercise session, and to ingest the ingestible thermistor (CorTemp; HQ, Inc., Palmetto, FL, USA) 8 h before their scheduled exercise session. The thermistor transmitted body core temperature to a hand-held monitor during the experimental condition. Participants in FIRE I and FIRE II consumed meal replacement bars and sport drink to standardize morning caloric intake equal to 40% of the basal metabolic rate for a moderately active adult. Upon arriving at the laboratory, participants provided a urine sample for a urine specic gravity (USG) measurement to ensure euhydration (USG \ 1.025). In female participants, this sample was also used for a urine pregnancy test. Participants were given an opportunity to void and then weighed in short pants (shorts and sport bra for females). A heart rate monitor strap was placed around the participants chest. Participants then put on a standardized station uniform of cotton-poly pants and a cotton t-shirt. Standardized instructions that included denition, scale anchors, frequency of administration, and answering any questions for the perceptual measures of RPE and RTS were read to each participant immediately prior to beginning the 50-min exercise session. Next, participants donned reghter TPC (heavy pants, heavy coat, nomex hood, boots, helmet, and gloves). A self-contained breathing apparatus (SCBA) was positioned on their back. After the participants fully donned the TPC, they were instructed to stand on the treadmill. Baseline measures of heart rate, body core temperature, and perceptual measures were assessed at this time. Participants in FIRE I and FIRE II performed an interval treadmill protocol in a heated room (33.040.0C; relative humidity 14.030.0%) to simulate the cyclical nature of re suppression in the presence of uncompensated heat

123

2028

Eur J Appl Physiol (2012) 112:20252034

stress (Hostler et al. 2010a, b). This interval protocol consisted of (1) walking for 20 min at 4.5 km h-1 (2.8 min h-1) at an incline of 2.5%, (2) walking for 3 min at 2.6 km h-1 (1.6 min h-1), 3) standing at rest for 4 min, (4) walking for 3 min at 2.6 km h-1, and (5) walking for 20 min at 4.5 km h-1 at an incline of 2.5%. With the exception of carrying an 8.2 kg bodybar in FIRE I, this 50-min interval protocol was the same for the FIRE I, FIRE II, and COG II investigations. Participants in FIRE I performed the interval protocol on three separate occasions and participants in FIRE II performed the interval protocol on seven separate occasions. Participants in COG I performed a 50-min continuous treadmill protocol at 4.5 km h-1 in a heated room (33.040.0C; relative humidity 14.030.0%) (Hostler et al. 2010a, b; Colburn et al. 2011). Termination criteria for the exercise session were (a) at the participants request (i.e. fatigue), (b) a body core temperature [39.5C, (c) HR [ 10 beats min-1 of ageadjusted HRmax, or (d) at the investigators discretion for the participants safety (i.e. unsteady gait). Heart rate and core body temperature were measured every 2 min of the exercise protocol. RPE and RTS measured at the conclusion of the initial 20-min exercise and again at test termination were used to develop the PHI. Physiological measures Heart rate was measured using a Polar heart rate monitor (Polar Electro, Finland). Body core temperature (Tc) was measured using an indigestible thermistor and a hand-held radio receiver (CorTemp; HQ, Inc., Palmetto, FL, USA). This device provides a core temperature measurement that is intermediate between rectal and esophageal temperature (OBrien et al. 1998). Physiological strain was determined by: PhSI = 5 * [(Tct - Tc0) / (39.5 - Tc0)] ? 5 * [(HRt - HR0)/(HRmax - HR0)]; where t = time of measurement (min), 0 = baseline measurement, Tc = body core temperature (C), and HR = heart rate (beats min-1) (Moran et al. 1998). Perceptual measures Rating of perceived exertion was obtained with the Adult OMNI Walk/Run Perceived Exertion scale. This category metric contains a rating range from 0, extremely easy, to 10 extremely hard (Utter et al. 2004). Ratings of thermal sensation was obtained with the OMNI Thermal Sensation Scale which contains verbal descriptors from the Gagge thermal sensation scale (Gagge et al. 1967) and numerical categories ranging from 1, comfortable, to 5 very hot. Perceptual strain was determined by the equation: PeSI = 5 * [(TSt - 1)/4] ? 5 * (PEt/10); where TS =

thermal sensation, and PE = perceived exertion (Tikuisis et al. 2002). The perceptual strain index used in the investigation modied the original equation by Tikuisis et al. (2002) incorporating the different perceptual metrics used where PE is perceived exertion as measured by the OMNI RPE scale and TS is thermal sensation as measured by the OMNI RTS scale. Data analysis Descriptive values of the combined data for anthropometric, physiological, and perceptual variables were calculated as mean standard deviation (SD) and presented for the separate investigations and the combined data set. Data from FIRE I and FIRE II were the mean response of each participants trials for a given protocol. Separate analyses were conducted for each physiological and perceptual response. Statistical signicance was accepted at the p \ 0.05 level. Regression analyses examined the relation between body core temperature and both RPE and RTS for the combined data and were used to construct the PHI. Statistical analyses were conducted using SPSS 17.0 for Windows (IBM SPSS, Inc., Chicago, IL, USA).

Results The means (SD) of the physiological and perceptual responses for each measurement point are presented separately for the four interrelated investigations and for the combined data set (Table 2). Relation between physiological and perceptual Moderate-to-strong correlations were found between body core temperature and perceived exertion (r = 0.742 to 0.856). Similarly, moderate-to-strong correlations were found between body core temperature and thermal sensation (r = 0.679 to 0.826). Strong correlations were found between perceived exertion and thermal sensation (r = 0.822 to 0.936) and between perceived exertion and heart rate (r = 0.862 to r = 0.916). These ndings were consistent across the four interrelated investigations and the combined data set. Physiological measures The calculated PhSI rating increased from 0.00 (0.00) at baseline to 4.71 (1.16) after 20 mins, and reached 8.00 (1.23) at termination of exercise. The PhSI rating was signicantly correlated (p \ 0.01) with body core temperature (r = 0.877) and heart rate (r = 0.876). Mean (SD)

123

Eur J Appl Physiol (2012) 112:20252034 Table 2 Physiological and perceptual responses during treadmill exercise in the heat Combined Tc (C) HR (b min-1) RPE (010) RTS (15) Baseline 20-min Termination Baseline 20-min Termination Baseline 20-min Termination Baseline 20-min Termination 37.0 0.4 37.6 0.3 38.5 0.6 90 17 147 22 170 17 0.24 0.42 4.88 1.63 7.43 1.86 1.78 0.77 3.65 0.67 4.50 0.68 FIRE I 37.1 0.5 37.8 0.3 38.2 0.6a 99 12 170 12c 179 8 0.15 0.31 6.52 1.55e 7.54 1.92 1.65 0.68 4.03 0.52 4.28 0.74 FIRE II 37.0 0.2 37.5 0.2 38.2 0.3a 95 11 149 19d 167 16 0.15 0.20 4.75 1.50 6.96 2.07 1.46 0.59 3.68 0.64 4.58 0.68 COG I 36.9 0.3 37.5 0.4 39.0 0.5 63 11 130 19 167 20 0.30 0.48 3.70 0.95 7.30 1.83 1.60 0.70 3.10 0.57 4.50 0.71
g b

2029

COG II 37.1 0.3 37.6 0.3 38.9 0.5 99 14 142 19 167 19 0.37 0.60 4.68 1.42 7.84 1.64 2.32 0.82f 3.68 0.67 4.63 0.60

Mean SD Tc body core temperature, HR heart rate, RPE rating of perceived exertion, RTS thermal sensation, FIRE I a laboratory study on rehydration, FIRE II a laboratory study of cooling, COG I a laboratory study of cognitive function, COG II a laboratory study of cognitive recovery
a b c d e f g

Indicates termination Tc was signicantly lower than in COG I and COG II Indicates baseline HR was signicantly lower than in FIRE I, FIRE II, and COG II Indicates 20-min HR was signicantly higher than in FIRE II, COG I, and COG II Indicates 20-min HR was signicantly higher than in COG I Indicates 20-min RPE was signicantly higher than in FIRE II, COG I, and COG II Indicates baseline RTS was signicantly higher than in FIRE I, FIRE II, and COG I Indicates 20-min RTS was signicantly lower than in FIRE I, FIRE II, and COG II

body core temperatures associated with each PhSI rating are presented in Table 3.

Perceptual measures The calculated PeSI rating increased from 1.10 (1.01) at baseline to 5.75 (1.51) after 20-mins, and reached 8.09 (1.59) at termination of exercise. The PeSI was signicantly correlated (p \ 0.01) with RPE (r = 0.969) and RTS (r = 0.970). Mean (SD) of body core temperatures associated with each PeSI rating are presented in Table 4. Development of the PHI The development of the PHI consisted of calculating PeSI values for all RPE-RTS combinations. Using a modication of the concept proposed by Moran et al. (1998) a graphical presentation consisting of RPE on the horizontal axis and RTS on the vertical axis was constructed (Fig. 1). Next, the mean Tc coincident with each calculated PeSI value was determined. These Tc values subsequently replaced the PeSI values on the constructed gure therefore linking the perceptual variables of RPE and RTS with the physiological criterion of Tc. As employed by Moran et al. (1998) zones representing the different levels of exertional heat strain were determined within the ranges of RPE, RTS, and Tc recorded across the four investigations examined presently. Each colored zone of the PHI encompassed several levels of exertional heat strain similar to that presented by

Table 3 Comparison of body core temperature between those reported by Moran et al. (1998) and those measured in the present investigation PhSI 0 1 2 3 4 5 6 7 8 9 10 Mean SD Tre rectal temperature, Tc core temperature using ingestible thermometric pill, Response Frequency number of occurrences for a given PhSI
a

Moran Tre (C) 37.12 0.03 37.15 0.04 37.35 0.03 37.61 0.03 37.77 0.04 37.99 0.05 38.27 0.07 38.60 0.04 38.70a No data available

Gallagher Tc (C) 37.0 0.4 37.3 0.3 37.5 0.3 37.5 0.2 37.8 0.3 37.8 0.3 38.2 0.2 38.8 0.2 39.4 0.2

Response frequency 65 0 0 5 13 17 21 14 21 20 12

indicates only one participant achieved the given PhSI in the study by Moran et al.

123

2030 Table 4 Body core temperatures (Tc) for each calculated perceptual strain rating PeSI 0 1 2 3 4 5 6 7 8 9 10 Mean SD Tc core temperature using ingestible thermometric pill, Response Frequency number of occurrences for a given PeSI Tc (C) 37.1 0.4 37.0 0.5 37.1 0.3 37.2 0.3 37.4 0.3 37.7 0.6 38.0 0.5 37.9 0.5 38.0 0.6 38.6 0.6 38.5 0.6 Frequency 13 19 22 10 11 21 8 22 19 25 18

Eur J Appl Physiol (2012) 112:20252034 Table 5 Mean body core temperature ranges for the perceptual hyperthermia index (PHI) PHI ratings 04 5 8 910 PHI zone Green Yellow Red Range of Tc (C) [36.00 to 37.49 37.50 to 37.99 38.00 to [40.50

Fig. 1 The perceived hyperthermia index (PHI) using ratings of perceived exertion and ratings of thermal sensation

Moran et al. (1998) and provided a specic range of Tc. The clinical denition of heat exhaustion states that it occurs at a body core temperature between 38.5 and 40.0C and that heat stroke occurs at a body core temperature greater than 40.5C. Therefore, it was determined that the upper PHI zone should be colored red as the body core temperature corresponding to the respective RPE-RTS coordinates are consistent with clinical denitions of major heat illness. The lower PHI zone, colored green, includes the normal body core temperature of approximately 37C (range of 36.2 to 37.7C). The PHI zone consisting of body core temperatures between the lower and upper PHI zones is colored yellow indicating cautionary physiologic thermal conditions (Table 5).

Discussion The present investigation developed the PHI to evaluate exertional heat strain in young adults performing treadmill exercise in TPC. The development of the PHI potentially

provides an estimation of body core temperature response to treadmill exercise with participants wearing thermal protective clothing and provides another step in the assessment of exercise-induced thermal stress using perceptual responses. A comparison of the physiological and perceptual responses during treadmill exercise between the four interrelated investigations indicated that they were not homogenous. This was possibly due to the heterogeneous samples across investigations and small methodological differences between the four experimental paradigms. However, the differences in body core temperature, HR, RPE, and RTS between the four interrelated investigations may be benecial and provide a more generalizable scenario in which the newly developed PHI could be used. The present ndings suggest that the PHI could be employed during either continuous or intermittent horizontal exercise (treadmill) with the individual holding equipment similar to a forcible entry tool or ladder. The PHI also takes into consideration a range of participant characteristics that may encompass varying levels of reghting experience from new recruit to novice to experienced reghter. The colored zones of the PHI, using perceptual assessments or RPE and RTS are anchored by body core temperature measurements associated with each level of strain represented by a combination of RPE and RTS. These zones were designed to encompass levels of physiological strain as presented by Moran et al. (1998) that identied ve zones: No/Little Strain (PhSI 02; 37.1 to 37.4C), Low Strain (PeSI 34; 37.6 to 37.8C), Moderate Strain (PeSI 56; 37.9 to 38.3C), High Strain (PeSI 78; 38.6 to 38.7C), and Very High Strain (PeSI 910; [38.7C). Therefore, the green zone of the PHI (36.2 to 37.7C) was selected as that range of body core temperatures are generally accepted as normal and incorporated No/Little Strain (Moran et al. 1998). The yellow zone of the PHI (37.8 to 38.4C) was selected as a precautionary zone immediate to the red zone precipitated by feelings of fatigue and muscle cramps (Casa and Roberts 2003; Latzka et al. 1998). This zone incorporated Little Strain and Moderate Strain (Moran et al. 1998). The red zone of the PHI started at 38.5C as research indicated this temperature as when heat illness may occur and may be precipitated by feelings of fatigue and muscle cramps, but more

123

Eur J Appl Physiol (2012) 112:20252034

2031

importantly syncope and ataxia, and dyspnea. This zone incorporated Moderate Strain, High Strain, and Very High Strain (Moran et al. 1998). Physiological measures Support for the use of the PHI as an analog for physiological feedback to heat stress comes from the physiological and perceptual responses found within the present investigation as well as previous research. The physiological responses recorded for the four interrelated investigations were similar to those published previously for reghters performing re-task simulation (Holmer 2006; von Heimburg et al. 2006), during treadmill exercise in personal protective equipment (Hostler et al. (2009)), and in the development of heat strain indices (Moran et al. 1998; Tikuisis et al. 2002). von Heimburg et al. (2006) observed heart rate responses of 167 b min-1 in rescuers upon reaching the top of a stairwell during high intensity drill. The same individuals achieved a heart rate of 182 b min-1 at the end of the task during simulated rescue of hospital patients which included climbing six oors and rescuing six manikin patients. Similarly, Holmer and Gavhed (2007) showed that heart rates averaged 168 b min-1 for a 22-min simulated re ghting activity and an average heart rate of 179 b min-1 for the heaviest of work tasks during the simulation. The mean heart rate achieved at exercise termination for the present investigation was 170 b min-1. However, the heart rate was 178 b min-1 for the participants in FIRE I who periodically carried a bodybar during the treadmill exercise. This value closely resembled the heart rate observed at the end of the simulated re task reported by von Heimburg et al. (2006) and the heaviest of works tasks used by Holmer and Gavhed (2007). More recently, Colburn et al. (2011) observed HR increase from 108 b min-1 at baseline to 177 b min-1 while assessing cooling techniques during a live-re evolution closely resemble the average peak HR (170 b min-1) achieved during the four laboratory settings examined presently as well as those during the FIRE I trial (Hostler et al. 2010a). Body core temperatures from the present investigation ranged from 37.0 to 38.5C and are similar to those reported by Moran et al. (1998) when developing the physiological heat strain index (37.138.7C), by Tikuisis et al. (2002) when developing the perceptual heat strain index (37.138.3C), and by Colburn et al. (2011) during a live-re evolution (37.438.3C). The four interrelated experimental paradigms employed presently resulted in a range of body core temperatures sufcient for the development of the perceptual hyperthermia index (PHI). The similarities in physiological responses between laboratory studies and live-re evolutions provide face validity for the physiological foundation of the PHI.

Perceptual measures Similar to the physiological responses, the perceptual responses recorded during the four related investigations were comparable with those reported previously under thermally stressful conditions during cycling and upperbody exercise. Galloway and Maughan (1997) recorded Borg Scale RPE during exhaustive cycling exercise at 70% of VO2peak in a warm (31C, 70% relative humidity) environment. Borg-RPE was 13 after 20 mins of exercise and 18 after 50 mins of exercise. These Borg-RPEs generally convert to OMNI-RPEs of 5 and 9, respectively (Robertson 2004). It was expected that uncompensable heat stress created by treadmill exercise performed while wearing thermal protective clothing would produce a greater thermal strain on the individuals in the present investigation than those performing cycling exercise as reported previously. In the present investigation, OMNIRPE was 5 after 20 mins and 7 at exercise termination. Using the inter-scale conversion, the OMNI-RPE from the present investigation at termination was lower than the Borg-RPE reported by Galloway and Maughan (1997) at 50 mins of exercise. Crewe et al. (2008) examined the rate of increase in RPE as a prediction of cycle exercise duration to fatigue under different environmental conditions. Three of the ve exercise conditions were performed in a hot (35C, 50% relative humidity, wind speed at 10 km h-1) environment at intensities of 55, 60, and 65% of peak power output. Borg-RPE after 20 mins of exercise was found to be 14 in the 55% trial, 16 in the 60% trial, and 18 in the 65% trial. These RPEs (converted to OMNI: 6, 7, 9) were higher than the rating of 5 after 20 mins of exercise in the present investigation. Borg-RPE at termination for all three hot trials was 19 (OMNI 10) which well exceeds the seven found in the present investigation. Differences between Crewe et al. (2008) and the present investigation could be attributed to the mode and intensity of exercise. The perception of effort during cycle exercise is localized to the active muscle, in this case, the legs. Given the smaller muscle mass being activated compared with treadmill exercise, the RPE presented by Galloway and Maughan (1997) and Crewe et al. (2008) should be higher and they indeed were higher than the present investigation. Additionally, Price and Campbell (2002) examined thermoregulatory responses during prolonged upper-body exercise in cool and warm conditions. Borg-RPE was approximately 15 after 30 mins during exercise in warm ambient conditions (31.5C, 48.9% relative humidity). A Borg-RPE of 15 generally equates to an OMNI-RPE of 7 (Robertson 2004). The OMNI-RPE of 5 recorded in the present investigation after 20 mins of exercise is lower than the level of perceived exertion for prolonged

123

2032

Eur J Appl Physiol (2012) 112:20252034

upper-body exercise as reported by Price and Campbell (2002). Although the perceptual results differ between investigations, it is plausible to consider that the additional weight of the thermal protective clothing and self-contained breathing apparatus would induce greater upperbody exercise. Price and Campbell (2002) used 60 mins of arm crank exercise at 60% VO2peak, whereas the present investigation employed treadmill mode, i.e., weight bearing aerobic exercise. Arm crank exercise targets a comparatively smaller muscle mass and may produce higher ratings of perceived exertion than treadmill exercise where comparatively large volume of muscle mass is activated. Physiological and perceptual heat strain The relation between the physiological and perceptual responses to the same thermal stimuli was the basis for the development of the PHI. Cheung (2007) has proposed that occupational exposure standards should not only use physiological outcomes, but also thermal perceptions reecting the behavioral and exercise response to heat stress. Previous research has produced conicting results as to the level and the interpretation of the relation between physiological responses and their perceptual analog during exposure to thermally stressful environments. Hostler et al. (2009) reported PeSI was signicantly greater than PhSI after 8, 23 min, and at termination of treadmill exercise for participants wearing chemical protective clothing. Tikuisis et al. (2002) reported that PeSI was lower than PhSI in a trained cohort of highly t (VO2max * 59 ml kg-1 min-1) individuals performing treadmill exercise in semipermeable protective clothing. However, the PeSI was greater than PhSI at the midexercise time points in the untrained cohort which consisted of moderately t (VO2max * 44 ml kg-1 min-1) individuals. The present ndings are consistent with those reported by Tikuisis et al. (2002) for an untrained cohort. The ndings from both Hostler et al. (2009) and Tikuisis et al. (2002) support the use of PHI in the present population of individuals and may represent a conservative approach to monitoring perceptual (and physiological) strain levels as there may be a learned or trained response altering perceptions of thermal strain. Although these results differ slightly from Petruzzello et al. (2009) who found that PeSI was less than PhSI, the differences between the present investigation and that of Petruzzello et al. (2009) are primarily methodological. Petruzzello et al. (2009) utilized short-duration activity in both the laboratory and eld settings as opposed to the longer duration ([20 mins) employed in the present investigation. Petruzzello et al. (2009) also modied the perceptual heat strain indices for both the laboratory and eld studies by altering the baseline to a relative value. In

this alteration, the initial rating of the individual represented the baseline value. Conversely, Tikuisis et al. (2002) and the present investigation used the minimum category scale rating as the absolute baseline value. It is possible that the different perceptual metrics may also have inuenced the calculations where the present investigation is the rst to use OMNI RPE and OMNI RTS as compared with previous investigations that employed the Borg-RPE and Gagge RTS. More research is needed in the area identifying the relation between perceptual and physiological strain and the modiers that could adjust those ratings. Limitations and future directions The PHI is preliminary and was developed using data from four interrelated investigations examining exercise performance in a heated environment but was not the primary outcome for these investigation. The PHI needs prospective validation and renement but the ndings suggest that the use of perceptual measures in place of physiological measures may be useful given the limitations of eld thermometry in the assessment of hyperthermia for individuals performing work under thermal stress, such as reghting (Pryor et al. 2011). The present investigation observed the anticipated doseresponse relation between OMNI-RPE and RTS but could not observe responses on the extremes of the scale (low RPE/high RTS, high RPE/ low RTS). Future investigations should examine scenarios such as high heat stress with low exertion (e.g. soldiers standing post) and high exertion with low heat stress to determine how broadly the PHI can be applied. Methodological differences among the four investigations may have impacted the ndings for the development of the PHI. Some of these differences include use of both individuals with and without reghting experience or experience to the thermal protective clothing, use of an additional weight to incorporate upper-body exercise such as carrying equipment similar to a forcible entry tool or ladder, and use of a continuous or intermittent treadmill protocol. However, the robust ndings in spite of this variation potentially speak to the broad application of the PHI. Future investigations should examine the inuence of reghting experience, upper-body exercise, and other exercise modes (locomotion speeds and exercise modes unique to reghting such as climbing, carrying, and crawling) on the relation between the physiological responses with the perceptual responses as it applies to properly identifying the relative risk of hyperthermia. Additionally, these four interrelated investigations were laboratory-based treadmill exercise in a simulated environment. While this eliminates much of the variation common to eld studies, the next logical step would be to use PHI as an assessment tool during live-re simulations

123

Eur J Appl Physiol (2012) 112:20252034

2033 Bilzon JL, Scarpello EG, Smith CV, Ravehill NA, Rayson MP (2001) Characterization of the metabolic demands of simulated shipboard royal navy re-ghting tasks. Ergonomics 44:766780 Casa DJ, Roberts WO (2003) Considerations for the medical staff: preventing, identifying, and treating exertional heat illnesses. In: Armstrong LE (ed) Exertional heat illnesses. Human Kinetics, Champaign, pp 177181 Cheung SS, McLellan TM (1998) Inuence of hydration status and uid replacement on heat tolerance while wearing NBC protective clothing. Eur J Appl Physiol Occup Physiol 7:139148 Cheung SS (2007). Neuropsychologial determinants of exercise tolerance in the heat. In: Sharma HS (ed) Progress in brain research, Vol. 162. Elsevier B. V., pp 4560 Colburn D, Suyama J, Reis SE, Morley JL, Goss FL, Chen Y, Moore C, Hostler D (2011) A comparison of cooling techniques in reghters after a live-burn evolution. Prehosp Emerg Care 15(2):226232 Crewe H, Tucker R, Noakes TD (2008) The Rate of increase in rating of perceived exertion predicts the duration of exercise to fatigue at a xed power output in different environmental conditions. Eur J Appl Physiol 103:569577 Epstein Y, Moran DS (2006) Thermal comfort and the heat stress indices. Ind Health 44:388398 Gagge AP, Stolwijk JA, Hardy JD (1967) Comfort and thermal sensations and associated physiological responses at various ambient temperatures. Environ Res 1:120 Galloway SD, Maughan RJ (1997) Effects of ambient temperature on the capacity to perform prolonged cycle exercise in man. Med Sci Sports Exerc 29:12401249 Holmer I (2006) Protective clothing in hot environments. Ind Health 44:404413 Holmer I, Gavhed D (2007) Classication of metabolic and respiratory demands in re ghting activity with extreme workloads. Appl Ergonomics 38:4552 Hostler D, Gallagher M Jr, Goss FL, Seitz JR, Reis SE, Robertson RJ, Northington WE, Suyama J (2009) The effect of hyperhydration on physiological and perceived strain during treadmill exercise in personal protective equipment. Eur J Appl Physiol 105(4):607613 Hostler D, Bednez JC, Kerin S, Reis SE, Kong PW, Morley JL, Gallagher M Jr, Suyama J (2010a) Comparison of rehydration regimens for rehabilitation of reghters performing heavy exercise in thermal protective clothing: a report from the reground rehab evaluation (FIRE) trial. Prehosp Emerg Care 14(2):194201 Hostler D, Reis S, Bednez JC, Kerin S, Suyama J (2010b) Comparison of active cooling devices with passive cooling for rehabilitation of reghters performing exercise in thermal protective clothing: a report from the reground rehab evaluation (FIRE) trial. Prehosp Emerg Care 14(3):300309 Jackson AS, Pollack ML (1985) Practical assessment of body composition. Physician Sportsmed 13(5):7690 Latzka WA, Sawka MN, Montain SJ, Skrinar GS, Fielding RA, Motatt RP, Pandolf KB (1998) Hyperhydration: tolerance and cardiovascular effects during uncompensable exercise-heat stress. J Appl Physiol 84(6):18581864 McLellan TM, Cheung SS (2000) Impact of uid replacement on heat storage while wearing protective clothing. Ergonomics 43:20202030 McPherson MJ (1992) The generalization of ACP. In: Proceedings of the 5th International Mine Vent Cong, Johannesburg, pp 2735 Moran DS, Shitzer A, Pandolf KB (1998) A physiological strain index to evaluate heat stress. Am J Physiol (Regul Integr Comp Physiol) 275:R129R134 Moran DS (2000) Stress evaluation by the physiological strain index (PSI). J Basic Clin Physiol Pharmacol 11:403423

and reghting training drills to assess the validity and reliability of the PHI in these practical settings and crossvalidation with currently employed tools that estimate environmental heat stress such as WBGT.

Conclusions In the absence of direct core temperature thermometry, assessment of heat stress under hot, ambient conditions is difcult outside of a laboratory setting (Pryor et al. 2011). The current knowledge base regarding temperature regulation under heat stress indices indicates the need to develop user-friendly methods for quick and accurate assessments in thermally stressful eld conditions. The majority of currently employed methods require expensive equipment, planning, and/or physical interaction between the investigator and the participant. One possible means to assess homeostatic disruption of temperature regulation in reghters is the PHI. The PHI was developed using RPE and RTS data obtained from four interrelated investigations that examined physiological, perceptual, and cognitive responses during treadmill exercise for participants wearing thermal protective clothing. Moderate-to-strong correlations were observed between the physiological measures (body core temperature, heart rate) and the perceptual measures (perceived exertion, thermal sensation). This relation between physiological response to heat stress and the accompanying perceptual identication of the physiological stress supports the use of perceptual measures to evaluate exertional heat strain. The PHI, although preliminary, provides a quick, easy, and unobtrusive method to assess exertional heat strain in the eld immediately following reground activities and may provide insight into recovery time and the reghters physiological readiness to return to the incident.
Acknowledgments The authors would like to acknowledge the Federal Emergency Management Agency Assistance to Fireghters Grant Program (EMW-2006-FP-02245) for funding the FIRE I and FIRE II studies. This study was also supported in part by Grant Number UL1 RR024153 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the ofcial view of FEMA, NCRR, or NIH. We would also like to thank our participants and the assistance of Jamey Bednez, Sarah Kerin, Julia Morley and Gillian Beauchamp in data collection. Conict of interest None.

References
Belding HS, Hatch TF (1955) Index for evaluating heat stress in terms of resulting physiological strain. Heat Pip Air Cond 27:129136

123

2034 NIOSH (1986) Criteria for a recommended standard: occupational exposure to hot environments. DHHS (NIOSH) Publication 86113. National Institute for Occupational Safety and Health, Washington, D.C., pp 101114 Northington WE, Suyama J, Goss FL, Randall C, Gallagher M Jr, Hostler D (2007) Physiological responses during graded treadmill exercise in chemical-resistant personal protective equipment. Prehosp Emerg Care 11:394398 OBrien C, Hoyt RW, Buller MJ, Castellani JW, Young AJ (1998) Telemetry pill measurement of core temperature in humans during active heating and cooling. Med Sci Sports Exerc 30:468472 Parsons K (2006) Heat stress standard ISO 7243 and its global application. Ind Health 44:368379 Petruzzello SJ, Gapin JI, Snook E, Smith DL (2009) Perceptual and physiological heat strain: examination in reghters in laboratory- and eld-based studies. Ergonomics 52(6):747754 Price MJ, Campbell IG (2002) Thermoregulatory responses during prolonged upper-body exercise in cool and warm conditions. J Sports Sci 20:519527 Pryor RR, Seitz JR, Morley J, Suyama J, Guyette FX, Reis SE, Hostler D (2011) Estimating core temperature with external

Eur J Appl Physiol (2012) 112:20252034 devices after exertional heat stress in thermal protective clothing. Prehosp Emerg Care (in press) Robertson RJ (2004) Perceived exertion for practitioners: rating effort with the OMNI picture system. Human Kinetics, Champaign Tennebaum J, Sohar E, Adar R, Gilat T, Yaski D (1961) The physiological signicance of the cumulative discomfort index (Cum DI). Harefuah 60:315319 Tikuisis P, McLellan TM, Selkirk G (2002) Perceptual versus physiological heat strain during exercise-heat stress. Med Sci Sports Exer 34(9):14541461 Utter AC, Robertson RJ, Green JM, Suminski RR, McAnulty SR, Nieman DC (2004) Validation of the adult OMNI scale of perceived exertion for walking/running exercise. Med Sci Sports Exerc 36(10):17761780 von Heimburg ED, Rasmussen AK, Medbo JI (2006) Physiological responses of reghters and performance predictors during a simulated rescue of hospital patients. Ergonomics 49(2):111126 Yaglou CP, Minard D (1957) Control of heat casualties at military training centers. Am Med Ass Arch Ind Health 16:302316

123