- Email: [email protected]
Comparative effectiveness of hypothermia rewarming techniques: radio frequency energy vs. warm water
RESUSCYTATION
Resuscitation 29 (1995) 203-214
Comparative effectiveness of hypothermia rewarming techniques: radio frequency energy vs. warm water J.W. Kaufman*, R. Hamilton, K.Y. Dejneka, G.K. Askew Naval
Air
Warfare
Center,
Aircrafr
Division,
Environmental
Effects
Branch,
Code 6023,
Warminster,
PA 38974-5MNI.
1JSA
Received 16 August 1994, accepted 5 December 1994
Abstract
The purpose of this study was to compare the rewarming effectivenessof a radio frequency coil (13.56 MHz) at a specific absorption rate (SAR) of 2.5 W/kg (RF) with warm water immersion (40°C) (WW) and an insulated mummytype insulating sack (IS) under simulated field conditions. Four male subjects, ages 24-35, were immersed in 10°C water for up to 90 min or until their rectal temperatures (T,) decreased to 35°C. Each subject had 3 trials in which they wereimmersed. After each immersion, rewarming was accomplished with either RF, WW, or IS, so that each subject wasrewarmed once with each method. Comparisons of the 3 rewarming methods werebased on the rate of increase of,T, during rewarmirig (T,$), T, 60 min after the start of rewarming (T,,.&, the time-interval measured from extraction from the water to the end of afterdrop (t& and the magnitude of any observed T, afterdrop (T&. WW had signi@antly greater TJt and TM than either RF or IS (P < 0.03) and a smaller fadthan IS (P < 0.05). IS had sign&a&y greater Tul than either WW or RF (P < 0.05). No significant differences in TJt, T,, or tnd were observed betweenIS and RF. The results of this study indicate that for mildly hypothermic individuals, active rewarming with RF at a SAR of 2.5 W/kg is less effective than WW and roughly equivalent to passive rewarming with IS. Keyworrls: Cold; Thermal; Cold water; Human exposure; Survival; Electromagnetic radiation
1. Introduction
Present resuscitation methods for victims of accidental hypothermia are unsatisfactory for a number of reasons. Active internal rewarming measures such as cardiopulmonary bypass, Abbreviations: IS, insulating sack; RF, radio frequency coil; SA, surface area; SAR, specific absorption rate; WW, warm water immersion. l
Correspondingauthor, Tel.: +I 215441 2565;Fax: +I 215
441 3739.
mediastinal lavage, and hemodialysis, while effective, are highly invasive, require extensive equipment, and are not readily performed in a field or emergency room setting. The effectiveness of these methods results from their ability to warm internal organs with minimal skin surface heating [ 1,2]. This minimizes the possibility of core temperature afterdrop, i.e. the continued drop in a hypothermia victim’s core temperature after removal from a cold environment and initiation of rewarming. Afterdrop is minimized by eliminating the temper-
0300-9572/95/$09.05 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0300-9572(94)00845-K
204
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
ature gradient between the core and skin surface during the early stages of rewarming [3,4]. Active external methods such as warm bath immersion are also effective but limit access for patient care, including cardiac monitoring, and place the patient at risk for core temperature afterdrop during the rewarming process. For patients with cardiovascular collapse and hypothermia, active external rewarming is contra-indicated and may be fatal, because core temperature afterdrop in these patients may result in ventricular fibrilation [l]. Passive rewarming while useful in the field, is a poor method for victims whose endogenous heat production is depressed due to lowered core temperature. A major goal in hypothermia research has been the development of a field-usable rewarming method capable of non-invasive heating of deep body tissues. Inhaling warm humidified air or oxygen is good at minimizing respiratory heat loss, but its advantage over shivering in warming the body core is; questionable [1,5]. Since inspiring humidified air or oxygen provides roughly 24 k&h, approximately 2.5 h are required to deliver the 58 kcal needed to raise the core temperature of a 70 kg person by 1°C [6]. Radio frequency induction coils are an active non-invasive core rewarming technique, based on selective energy absorption by muscle and bone tissue. By minimizing energy absorption at the skin, radio frequency coils hold the promise of minimizing afterdrop. As sudden deaths among ostensibly recovering hypothermia victims can often be attributed to core temperature afterdrop [l], this would be a signifi-
cant contribution to field treatment. Rewarming of mammals has been reported with radio frequency coils [7,8], but development has been hampered by a proclivity for skin burns. Development by Olsen et al. [9] of a tunable radio-frequency induction coil has led to the successful rewarming of hypothermic rhesus monkeys. As part of a multilaboratory study [lo], we report the testing of this device by rewarming mildly hypothermic human subjects under simulated field conditions. These results were compared with rewarming by 2 other methods: warm water immersion and a thermal rewarming sack. 2. Materials and methods 2.1. Subjects Four males (Table l), ages 24-35 years, volunteered to participate as subjects, after being fully informed of the details of the experimental protocol and associated risks. These procedures were approved by the Naval Air Development Center’s Advisory Committee for the Protection of Human Subjects and the Food and Drug Administration’s Section for Investigational Medical Devices. Weight was recorded prior to each test run and the mean for each subject calculated. Body surface area (SA) was calculated [ll] from the mean weight and height of each subject. Percent body fat was determined from estimates of body density [12], which were computed from skinfold measurements obtained with Lange Skinfold Calipers (Cambridge Scientific Inc., Cambridge, MD) and the equation of Lohman [13].
Table 1 Physical characteristics of subjects Subject
Gender
Age (yea@
Height (4
Weight 0%)
% Body fat
Surface area (m*)
A B C D
M M M M
35 26 30 24
1.78 1.85 1.75 1.79
79 91.4 71.9 67.9
13 17 17 10
1.97 2.16 1.87 1.85
29 3
1.79 0.02
77.6 6.8
14 2
1.96 0.05
Mean S.E.M.
J. W. Kaufinan et al. /Resuscitation 29 (1995) 203-214
Personal
Flotation
205
Device
Fig. 1. Approximate body position during cold water immersions. The flotation device used in the study is de&@d keep the UPPer body at approximately 45O relative to the water surface so that the distance from the water to the mouth is maximized.
2.2. Experimental procedures
Subjects generally reported to the laboratory at the same time of day for each trial (either morning or afternoon) and were given physical examinations by the attending flight surgeon. Subjects were asked to follow their normal diets during the study and abstain from alcohol for at least 24 h prior to a trial. All subjects were non-smokers. Urine samples were collected and analyzed as part of the flight surgeon’s examination of the subject. Subjects wore a bathing suit, cotton tee-shirt, neoprene wet suit booties (thickness, 4.8 mm), anti-exposure mittens, a personal flotation device, and were instrumented with rectal and 6 surface skin temperature sensors. Booties and mittens were worn in an attempt to preclude trial terminations due to low extremity temperatures. Subjects were immersed to the neck in a stirred pool (1.5 m deep x 2.4 m diameter) of 10°C water (T,,,,,) using a personal flotation device for buoyancy (Fig. 1). This placed subjects at roughly a 45“ angle in the water, allowing their upper chest to rise out of the water, and represents the position individuals assume in the water during many survival situations. Subjects also kept their hands out of the water by resting them on the flotation bladders. Immersions were terminated after 90 min had elapsed, rectal temperature (T,) = 35”C, hand or foot temperature (Thand) = lO”C, or when the
exposure exceeded the subject’s tolerance. Following removal from the water, the flotation device, mittens, and booties were removed from the subject, who was then quickly dried with towels rtnd rewarmed. The time from exiting the water to start of rewarming was 2.9 * 1.0 min. by either the radio-frequency water immersion (WW), or a the chamber to the rewarmi
area in a prone
Ambient air temperature 25 f 4.O”C. The ex balanced the order of subject acted as their own control. Triels for individual subjects were separated by at least 48 h. 2.3. Rewarming methods
Rewarming by each of the methods was continued until T, WZIS37.0°C, unless 60 trGn elapsed without a marked rise in T,. When a increase in T, was observed after 60 min, rewarming using either RF or IS was terminated and the subject was rewarmed by WW. 2.4. Radio frequency rewarming coil
The RF rewarming coil consisted of a helical coil of steel cable within aglastic sheath connected
206
J. W. Kaujkan et al. /Resuscitation 29 (1995) 203-214
Radio Frequency
Coil
00 Nonmetallic
RF
Table
POWER 1 SOURCE
1
Fig. 2. Schematic drawing of the RF coil as used in this study.
to a crystal oscillator (Electronic Navigational Instruments, Rochester, NY, model ACG-5) emitting radio frequency energy at 13.56 * 0.01 MI-Ix (Fig. 2). The coil was kept away from the skin surface by 4 pads of Nautilus foam located about the torso. Specific absorption rates (SAR) were adjusted to 2.5 W/kg body weight, by changing the output power of the coil for each subject (based on measured weight). While SARs of 4.0-5.0 Wikg body weight are suggested by animal studies to maximize rewarming rates [7], a lower SAR was used because higher SARs increase the risk of localized hotspots [14]. The RF system was tuned for each subject during rewarming to minimize the amount of reflected power. A RF field density of 2.6 mW/cm2 was measured at a distance of 1.0 m from the RF coil during operation. During this study, the RF coil covered the subject’s torso from just below the clavicle to approximately 10 cm below the level of the umbilicus. All metal was kept clear of the RF field during rewarming to prevent skin burns. Hot spots generated by variations in coil Iit were identified by the subject and the coil adjusted accordingly. 2.5. Mummy-type insulating sack The insulating sack (Domtex International, Yorkshire, UK, model Decupad Thermal Recovery Capsule) was a mummy-type bag with a high pile polypropylene lining designed for recovery
and field treatment of hypothermia victims. The lining is intended to provide thermal insulation and wick moisture from a hypothermia victim’s clothing and skin. The bag completely covered an individual except for the face. Multiple zippers allowed access to all body areas during use. 2.6. Warm water immersion While not normally a practical field rewarming technique, WW was included as the study control, because in most laboratory hypothermia studies it is the chosen rewarming method. Subjects were immersed in stirred 40°C water with their head and extremities (distal to mid-thigh and mid-upper arm) out. Water temperature was continuously monitored with a 26 AWG type T thermocouple. Adding 41°C water to the tub whenever water temperature fell below 39.5”C maintained a roughly constant temperature (40 f l°C). When a positive trend was observed for the change in T, lasting more than 10 min, subjects were permitted to place their extremities in the water. This conforms to a standard laboratory method of warm water rewarming [1,5,15-181. 2.7. Physiological measurements Use of RF energy precluded the use of thermocouple or thermistor temperature sensors during rewarming, because of their metallic composition. To measure temperatures accurately during rewarming, a tibre optic temperature measurement system designed for use in RF fields was used’(Luxtron Corp., Mountain View, CA, model 3000). Dual sensors were inserted 10 cm past the anal sphincter for measuring T, while surface skin temperatures were measured at 6 sites (i.e., forehead, arm (biceps), lateral surface of torso (approx. 4th intercostal space), upper chest, anterior thigh, and lower back. Esophageal temperatures were not obtained because of reluctance on the part of subjects. Thermistors were used for monitoring hand and foot temperatures during prerewarming immersions - as they could not be employed during RF rewarming, they were removed prior to initiating rewarming in all trials. Electrocardiograph (ECG) signals were monitored during pre-rewarming immersions with ECG electrodes (3M, Minneapolis, MN, Red Dot)
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
amplified with isolated ECG amplifiers (Gould, Cleveland, OH, model 4600 series amplifiers), The ECG electrodes were also removed prior to the start of rewarming procedures. Subjects’ heart rates during rewarming were monitored by means of a stethoscope. During RF rewarming, a plastic stethoscope was used. 2.8. Subjective responses Subjective feelings of fatigue, shivering, temperature, and comfort were evaluated every 15 min throughout the exposure period by means of category scales (Table 3). Subjects were instructed to indicate their subjective sensation for each criterion on a l-7 scale where 1 indicated the most pleasant situation and 7 the most unpleasant.
cal-‘), respectively. It was assumed for these calculations that there was no net metabolic contribution to P,, i.e. metabolic heat production was balanced by heat losses to the environment. The power output of the RF coil was calculated from: PRF = @AR)@&,) (watts)
0.8W+%WWt.~l PRW
=
W
(watts)
(1)
t
where Paw = power needed to rewarm subject, Mb = body mass (kg), AT, = T, increase during rewarming (“C), t = rewarming time in hours, 0.83 = body specific heat (cal g-‘“C-‘), 1009 and 0.001164 = conversion factors (g kg-‘) and (W h
(2)
It was assumed for this calculation that there were negligible losses from reflection. As the coil was tuned for individual subjects during rewarming, actual losses due to reflection were generally < lO%, based on measurements of generated and reflected power. Efficiency was then determined by: %Efficiency = Paw/Par X 100
2.9. Calculated values Comparisons of the 3 rewarming methods were based on the T, rewarming rate (AT,&), afterdrop duration (At&, magnitude of T, afterdrop (AT&, and the actual or estimated T, after 60 min of rewarming (T,& [l&19,20]. The AT& was calculated from the slope of the T, first order regression line divided by the duration of rewarming [ 161. The slope was calculated from the time a positive trend was observed in T, following the start of rewarming procedures. The At4 was calculated from the end of immersion to the beginning of a monotonic increase in T,, i.e. the ‘afterdrop phase’ [ 161. The ATd was the difference between T, at the start of rewarming and the minimum T, during rewarming [la]. Values of Treso were linearly extrapolated if rewarming did not last 60 min. The energy transfer efficiency of the RF was determined by estimating the power required to raise T, by the amount observed in each of the RF runs. This was given by:
207
(3)
2. IO. Statistical analysis Physiological data from this study was analyzed using the non-parametric Fri When Variance (ANOVA). detected, the Mann-Whitney U test was used to identify where the differences existed between rewarming methods. The non-pammet& KruskalWallis ANOVA was used to detect dif&rences in subjective responses (as a function of time) between rewarming methods. Regression analysis was used to determine the slopes of the rewarming T, first order regression lines. Missing initial values were estimated by the technique of Winer [21]. Differences were considered significant at the level of P < 0.05. 3. Results 3.1. Induction of mild hypothermia The immersions prior to rewarming had a mean duration of 13.5 f 19.7 min. During the cooling period, a mean T, drop = -1.7 i 0.7”C was observed, with a mean T, at the end of cooling of 35.8 f 0.8”C. The mean T, drop during the cooling period for each rewarming method is given in Fig. 3. While the final T,‘s were higher than desired (i.e. T, were 35.0%), they were found to be significantly lower than the initial T, (P < 0.001). Individual skin site temperatures varied
J. W. kiwfmnn et al. /Resuscitation 29 (1995) 203-214 I
I
3
I
I
I
I
I
I
I
‘..‘....” insulating sack -. warm water radio frequency device
2
1
Y -
0
2 a -1
-2
4
-3 1
I
-0.9
-0.8
I
I
-0.7
I
-0.6
-0.5
I
I
-0.4
-0.3
-0.2
-0.1
*exl&tal Fig. 3. Mean change in rectal temperature (AT, at time i - starting TJ as a function of non-dimensional immersion time during the cooling phase. T’ii has been normalized by dividing time from the start of rewarming (t& by the total duration of a given trial’s cooling phase (&J). Each curve represents mean values for the cold water immersion periods preceding each rewarming method f S.E.M., n = 4. 1.0 0.8 --- .-. .-. -
0.6
insul:tling wck rack warn waler radio freclucncy
device
0.4 0.2 Y u" & -3
__-.--. -*~g?.
0.0
t
I -,..
-0.2
-.
_ ,,Ai_-i
., .-IL..
-. r --
-0.4
-----1
‘1‘-.. l‘-‘-I---r --A_
----
-0.6 r -0.8 -1.0 0
5
10 El:lpScd
H~Will7llillg
15 The,
20 tre,y:,r,,,,
25
30
minutes
Fig. 4. Mean change in rectal temperature (T, at time i - T, immediately prior to rewarming) during the first 30 min of rewarming. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
J. W. Kaufinan
et al. /Resuscitation
29 (1995)
203-214
Table 2 Rewarming rate and estimated or actual Tr, after 60 mm of rewarming for each rewarming method Rewarming technique
Rewarming rate (“cm
Estimated T,, 60 min v-2
Radio frequency
Warm water
Insulating sack
0.54 (* 0.2) 36.1 (* 0.2)
1.82 (* 0.5) 31.5 (a 0.3)
0.85 (* 0.3) 36.4 (* 0.3) __l_._ll___.
Values are means f S.E.M.
according to distance from the water. The immersed skin sites had temperatures approaching Atwater at the end of cold water immersion (Tback = 11.0 f 0*8”C, Tsi& = 13.8 f 2*5”C, Tthigh = 10.2 f O.YC). The other skin sites were considerably warmer (T’r,,=k-,-~ = 32.8 f l.O”C, r,,,, = 25.6 f 6S’C, T,, = 17.0 f 6.O”C). The relatively high Tchestwas due to the position assumed by a subject’s body while floating in water supported by a personal flotation device. Mean heart rate (HR) at the end of the cooling phase = 100 f 14 beats/mm (bpm) with no ECG abnormalities observed during these immersions. No significant differences between rewarming methods were observed for T,, skin temperatures, or HR during induction of mild hypothermia. 3.2. Rewarming Fig. 4 shows the mean AT, for each rewarming method over the first 30 min. At minute 30, one of the WW subjects had completed the rewarming process, so this represents the minimum time re-
quired for rewarming all subjects. The mean duration of rewarming was 46.0 * 7.1 min for WW and 76.8 f 9.1 min for RF. In 2 cases, IS rewarming proved to be too slow and had to be augmented with WW after 60 min. No duration is reported for IS for this reason. 3.3. Duration and magnitude of rectal temperature afterdrop A significantly shorter Atad was observed during the WW trials compared with the IS trials (P C 0.05). No other significant di%?rences in At,, were noted between the rewarming methods (Fig. 4). ATad was significantly larger in the IS trials compared with WW (P < 0.05) and RF (P < 0.03) trials. The differences in AT,, between RF and WW trials were not observed to be significant (Fig. 4). 3.4. Rate of increase of T,, during rewarming Warm water rewarming (WW) resulted in a significantly larger AT,& than either the RF coil
Table 3 Category scales used to evaluate subjective responses to environmental conditions Scale
1 2 3 4
Category Comfort
Fatigue
Shivering
Temperature
Extremely comfortable
Extremely energetic
No shivering
Very hot
Neutral
Neutral
Neutral
Extremeiy uncomfortable
Extremely exhausted
Severe shivering
5 6 7
Values are means f S.E.M.
Very cold
J. W. Kaufma et al. /Resuscitation 29 (1995) 203-214
210
radio frequency device r:lclio
--
.9 E
100
60
50 50
I 0
I 10 10
5
I
I
15
20
1 25
30
Fig. 5. Mean heart rate during the first 30 min of rewarming. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
, “i’,A--fr
25
--!2----p-~--q--~-~
b:lck, WW
0 0
5
W)scd
I
1
I
I
10
15
20
25
retvamirlg
tirlle, trC.,,,,,l.,,,,lllinules
Fig. 6. Mean changes in back and chest skin temperatures during the first 30 min of rewarming. Abbreviations for the rewarming methods are: IS - insulating sack, RF - radio frequency device; and WW - warm water. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
J. W. Kaufm
et al. /Resuscitation 29 (1995) 203-214
(RF) (P < 0.03) or the thermal sack (IS) (P < 0.03). No significant difference was observed between the RF and IS (Table 2). A significantly greater Tti was observed for WW than either RF or IS (P c 0 0.01) (Table 2). 3.5. Heart rate No significant differences between methods were observed for heart rate during the rewarming trials (Fig. 5). The initially elevated HR observed for IS and WW correspond to bouts of shivering. During RF trials, shivering was probably sup pressed initially by the body movements required to don the RF coil. These findings indicate that there was no significant difference in physiological stress during rewarming with either RF, WW, or IS. 3.4. Skin temperatures WW rewarming resulted in significantly higher skin temperatures throughout most of rewarming (P < 0.01) compared with RF and IS. Fig. 6 show back and chest temperatures, which together represent the part of the body (i.e. torso) where the RF deposited most of its energy. No significant differences in skin temperatures were noted between RF and IS. 3.7. Subjective responses No sign&ant differences were observed between rewarming methods for any of the subjective measures recorded in this study. Differences in subjective responses between subjects were also not significant during rewarming. 3.8. RF efficiency The mean efficiency of the RF coil calculated from equation [3] was 32 i 11%. An important factor in the observed variability in efficiency was differences in measured reflected power between subjects, which was probably due to the quantity and distribution of body fat. In addition, while the RF coil had a focused field, there were stray losses to the surroundings which contribute to lowered efficiency but should be similar between runs. 4. Discrasioo From a theoretical standpoint,
RF should pro-
211
vide a superior means of increasing T, while minimizing afterdrop compared to either WW or IS [2,8-lo] based on the RF’s m&h& of operation. This was not observed in this study the AT,&, At,,,+ and AT4 results. It is that the factors contributing to these results include: (a) a lower SAR than required to quickly rewarm mildly hypothermic humans; (b) i&e&d power and energy absorption by the ski% subcutaneous tissues, and surface moisture; and (c) lack of insulation between the skin surface and the ambient environment, allowing heat losses across the skin surface. The T,& for RF of O.lPC/h observed in #he present study, at a mean rewarming starting T, of 35.9”C, compares unfavorably with ear&~ work with RF coils. Rhesus monkeys, when initially Tn/t of cooled to T, of 28.3”C, d -ted 5.6”Clh at a SAR of 5.5 Wikg body weight [22], while hypothermic dogs cooled to a perature (T,) of 25’C, had a T@ of 5 SARs of 4-6 W/kg body weight @j. Thi that an increased SAR would probabty merease T,.&. The increased risk of skin burna, however, becomes a particular concern in a f&d situation where measuring skin temperatures are diEcult.
a linear relationship between SAR and rewarming rate appears ill-advised. The relatively low efticiency in tranr&rring RF energy to the body can be attributed to a number of factors. The most obvious is losses due to reflected power, although careful tuning limited losses to less than 10%. Surf&e absorption is known to occur with RF energy [7,14] with upwards of 25% of the energy being deposited there [R. Olsen, personal communication]. Water on the skin surface was also recognized as a pot&al energy sink. In an attempt to address this problem, subjects were dried off with a towel immediately prior to rewarming. While this was generally successful in removing excess water, there is no doubt that some water was retained which would have reduced the efficiency of the RF. Water retained in clothing could pose a significant problem when using RF for field rewarming. The lack of insulation between the skin surface
212
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
and the ambient environment during RF rewarming resulted in continual heat losses. Increasing skin temperatures increased the temperature gradient between the skin and environment during RF. Rising skin temperatures, while precluding vasoconstriction during either RF or WW, had the effect of eliminating the only means available to minimize transcutaneous heat losses during RF under the study conditions. Transcutaneous heat losses were not a problem with either IS or WW because WW utilizes the thermal gradient between skin and water to transfer heat to the body, while IS uses insulation to minimize heat transfer between the skin and ambient environment. Future work with the RF coil should include clothing, which will increase skin surface insulation so that this source of heat loss is mitigated and provide for a more realistic test of RF rewarming in a field scenario, though exacerbating the problem of energy absorption by water at the skin surface. A previous study by Hesslink et al. [lo] reported that RF was superior to WW. This finding was based on the more rapid responses of T,, to RF than either WW or IS, though they also found that T, responded more quickly to RF [lo]. The T, rewarming rate for WW of 1PC/h observed in the present study is similar to those found in other WW studies by Romet [19] of 1.6Wh and Hoskin et al. [16] of 2.3”C/h, but considerably larger than 0.3Wh found by Hesslink et al. [lo]. While Hoskin et al. [ 161 began rewarming after a AT, of 1.6”C during the cooling period, similar to the present study’s pre-rewarming AT, of 1.9”C, Romet [19] had a pre-rewarming AT= of OS’C. Curiously, Hesslink et al. [lo] also had a AT, of 0.5”C prior to rewarming, despite a considerably smaller overall WW T, rewarming rate. This anomalous result may be due to the position in the tub during rewarming, i.e. Hesslink et al. [lo] kept the upper chest and neck out of the water in contrast to the other studies which immersed people up to the neck. Differences in the extent of body cooling may have also contributed to the conflicting observations between this work and that of Hesslink et al. [lo]. In the present study, immersed skin temperatures asymptotically approached Twater toward the end of the cooling phase. During the
time between cold water extraction and initiation of rewarming, Tbck increased only 3.7”C. Back temperatures reported in the Hesslink study prior to rewarming were at least 16°C warmer than water temperature, which suggests only slight body cooling [lo]. The quasi-steady state temperature losses observed in longer duration cold exposures [17,18,24] had apparently not been achieved. Not only would less energy be required to raise body temperatures, but transcutaneous heat losses would be minimized because of the small temperature gradient between the skin and ambient air. The rapid response of T, compared with T, observed in the Hesslink et al. study [lo] may represent the fact that the mass of the body was not deeply cooled. Mittleman and Mekjavic [3] demonstrated that T,, is influenced by peripheral blood temperatures while T, is not. Thus the relatively warm back and triceps temperatures observed by Hesslink et al. [lo] suggest that the peripheral blood supply enhanced the observed T, rebound. As T, is dependent upon convective heat exchange from the central blood supply and conductive heat exchange with surrounding tissues, the relatively warm peripheral tissues at the start of rewarming would selectively aid an increase in T,, without directly influencing T,. A number of practical considerations must also be addressed if RF is to become a useful field rewarming method. Radio frequency interference with electronic devices precluded continuous metabolic measurements and ECG monitoring during the present study. This would be unacceptable when dealing with cases of severe hypothermia, where cardiac monitoring is essential, and during aeromedical transport, when navigational device would likely be affected. There is also a possible danger to crew members because they would be exposed to unshielded RF waves. Identification of local hot spots and adjustments to the coil to avoid skin bums was dependent upon an alert and cooperative subject. For field use, a more dependable method of skin temperature monitoring would be required. In addition, adjustments to the coil could be difficult or hazardous with an unconscious or injured hypothermia victim. These deficiencies can possibly be corrected with further development,
J. W. Koufin
et al. / Resuscitution
but they must be borne in mind when discussing an RF rewarming coil as a potential field device. This study demonstrated the effectiveness of WW rewarming as a means of quickly raising T, in mildly hypothermic individuals. Afterdrop during WW rewarming was not as dramatic as suggested in earlier literature [10,25-271. Its simplicity and effectiveness support the continued use of WW as a method of choice for rapid rewarming of hypothermic individuals under the test conditions. RF and IS proved to be roughly equivalent in the ability to rapidly increase T,, though use of IS was observed to produce a greater afterdrop. Both RF or IS were observed to be significantly less successful in rapidly increasing T, during rewarming than WW. In conclusion, WW rewarming was observed to be the most effective means of rewarming mildly hypothermic individuals when compared with RF or IS. While little difference was found between RF and IS in the present study, these results cannot be extrapolated to cases of severe hypothermia. Since metabolic energy production is diminished in hypothermic individuals, active rewarming by RF has a theoretical advantage over IS, which depends on the passive retention of metabolically generated heat. Given these considerations, a study in which individuals with differing metabolic rates are rewarmed could be used to differentiate between the effectiveness of RF and IS. Considerable work has yet to be done to conclusively demonstrate that RF can rewarm at least as effectively as WW when SARs are increased.
Acknowledgements The authors wish to thank the subjects of this study, whose perseverence despite considerable personal discomfort made this work possible. We would also like to thank Walter Soroka, LCDR Angus Rupert, and the Biomedical Support Group (Code 6025) for their technical support, without whom completion of this study would not have been possible. The views expresses herein are solely those of the authors and should not be construed as the
29 (19951
203-214
official policy or position of the Department Defense or Department of the Navy.
213
of
References 111Hamett RM, Pmitt JR, Sias FR. A review of the literature concerning resuscitation from hypothermia: Part I - The problem and general approaches. Aviat Space Environ Med 1983; 54: 425-434. PI Hamett RM, Pruitt JR, Sias FR. A review of the literature concerning resuscitation from hypothermia: Part II - Selected rewarming protocols. Aviat Space En&on Med 1983; 54: 487-495. 131 Mittleman KD, Mekjavic IB. Effect of o&&d venous return on core temperature during cold water immersion. J Appl Physiol 1988; 65: 2709-2713. [41 Webb P. Afterdrop of body temperature during rewarming: an alternative explanation. J Appl Physiol 1986; 60: 385-390. 151 Marcus P. Laboratory comparison of techniques for rewarming hypothermic casualties. Aviat Space Environ Med 1978; 49: 692-697. I61 Bangs CC. Hypothermia and Frostbite. Emerg Med Clin N Am 1984; 2: 475-487. I71 Olsen RG, Ballinger MB. David TD, Lotz WG. Rewanning of the hypothermic rhesus monkey with electromagnetic radiation. Bioelectromagnetics 1987; 8:!83-193. 181 White JD, Butterfield AB, Greer KA, Schoem S, Johnson C, Holloway RR. Comparison of rewarming by radiowave regional hyperthermia and warm humidified inhalation. Aviat Space Environ Med 1984 [91 Olsen RG. Reduced temperature afterdrop in rhesus monkeys with radio frequency rewarming. Aviat Space Environ Med 1988; 59: 78-80. 1101Hesslink RL Jr, Pepper S, OIsen RG, Lewis SB, Homer LD. Radio frequency (13.56 MHz) energy enhances recovery from mild hypothermia. J Appl Physiol 1989; 67: 1208-1212. 1111 Dubois EF, Dubois D. Measurement of surface area of man. Arch Int Med 1915; 15: 868. 1121Brozek J, Grande F, Anderson JT, Keys A. Densiometric analyses of body composition: revision of some quantitative assumptions. Ann New York .Acad Sci 1963; I 10: 113-140. 1131Sinning WE, Dolny DG, Little KD, Cunningham LN, Racanielli A, Siconolfi SF, Sholes JL. Validity of “generalized“ equations for body composition analysis in male athletes. Med Sci Sports Exert 1985; I?: !24- 130. [I41 D’Andrea JA, Emmerson RY, Dewitt JR, Gandhi OP. Absorption of microwave radiation by the anesthetized rat: Electromagnetic and thermal hotspots in body and tail. Bioelectromagnetics 1987; 8: 385-396. 1151 Daanen HAM, Van De Linde FJG. Comparison of four noninvasive rewarming methods for miId hypothermia. Aviat Space Environ Med 1992; 63: !070- IO76
214
J. W. KaufmMet al. /Resuscitation 29 (1995) 203-214
[16] Hoskin RW, Melinyshyn MJ, Romet ‘l-f, Goode RC. Bath rewarming from immersion hypothermia. J Appl Physiol 1986; 61: 1518-1522; [17] Kaufman JW, Dejneka KY, Bagian JP, Morrissey SJ, Bittner Jr, AC. Cold water evaluation of NASA Launch Entry Suit (LES). Naval Air Development Center, Warminster. PA, Tech. Rept. NAJXXO-017, July, [18] Kaufman JW, Dejneka KY. Cold water evaluation of constant-wear anti-exposure suit systems. Naval Air Development Center, Warminster, PA, Tech. Rept. NADC-8509240, June, 1985. [19] Romet TT. Mechanism of afterdrop after cold water immersion. J Appl Physiol 1988; 65: 1535-1538. [20] Romet Tf, Hoskin RW. Temperature and metabolic responses to inhalation and bath rewarming protocols. Aviat Space Environ Med 1988; 59: 630-634. [21] Wirier BJ. Statistical Principles in Experimental Design. New York: McGraw-Hill, 1962. [22] Olsen RG, David TD. Hypothermia and electromagnetic
[23]
[24]
[25] [26] [27]
rewarming in the rhesus monkey. Aviat Space Environ Med 1984; 55: 1111-1117. Morrison JB, Conn ML, Hayward JS. Accidental hypothermia: The effect of initial body temperatures and physique on the rate of rewarming. Aviat Space Environ Mcd 1980; 51: 1095-1099. Steimnan AM, Kubihs PS. Survival at sea: The effects of protective clothing and survivor location on core and skin temperatures. U.S. Coast Guard Rept. No. CG-D 26-86. Nat. Tech. Info. Serv. Springfield, VA, 1986. Collis ML, Steimnan AM, Chaney RD. Accidental hypothermia: An experimental study of practical rewarming methods. Aviat Space Environ Med 1977; 48: 625-632. Hayward JS, St&mum AM. Accidental hypothermia: An experimental study of inhalation rewarming. Aviat Space Environ Med 1975; 46: 1236-1240. Hayward JS, Bckerson JD, Kemna D. Thermal and cardiovascular changes during three methods of resuscitation from mild hypothermia. Resuscitation 1984, 11: 21-33.
Resuscitation 29 (1995) 203-214
Comparative effectiveness of hypothermia rewarming techniques: radio frequency energy vs. warm water J.W. Kaufman*, R. Hamilton, K.Y. Dejneka, G.K. Askew Naval
Air
Warfare
Center,
Aircrafr
Division,
Environmental
Effects
Branch,
Code 6023,
Warminster,
PA 38974-5MNI.
1JSA
Received 16 August 1994, accepted 5 December 1994
Abstract
The purpose of this study was to compare the rewarming effectivenessof a radio frequency coil (13.56 MHz) at a specific absorption rate (SAR) of 2.5 W/kg (RF) with warm water immersion (40°C) (WW) and an insulated mummytype insulating sack (IS) under simulated field conditions. Four male subjects, ages 24-35, were immersed in 10°C water for up to 90 min or until their rectal temperatures (T,) decreased to 35°C. Each subject had 3 trials in which they wereimmersed. After each immersion, rewarming was accomplished with either RF, WW, or IS, so that each subject wasrewarmed once with each method. Comparisons of the 3 rewarming methods werebased on the rate of increase of,T, during rewarmirig (T,$), T, 60 min after the start of rewarming (T,,.&, the time-interval measured from extraction from the water to the end of afterdrop (t& and the magnitude of any observed T, afterdrop (T&. WW had signi@antly greater TJt and TM than either RF or IS (P < 0.03) and a smaller fadthan IS (P < 0.05). IS had sign&a&y greater Tul than either WW or RF (P < 0.05). No significant differences in TJt, T,, or tnd were observed betweenIS and RF. The results of this study indicate that for mildly hypothermic individuals, active rewarming with RF at a SAR of 2.5 W/kg is less effective than WW and roughly equivalent to passive rewarming with IS. Keyworrls: Cold; Thermal; Cold water; Human exposure; Survival; Electromagnetic radiation
1. Introduction
Present resuscitation methods for victims of accidental hypothermia are unsatisfactory for a number of reasons. Active internal rewarming measures such as cardiopulmonary bypass, Abbreviations: IS, insulating sack; RF, radio frequency coil; SA, surface area; SAR, specific absorption rate; WW, warm water immersion. l
Correspondingauthor, Tel.: +I 215441 2565;Fax: +I 215
441 3739.
mediastinal lavage, and hemodialysis, while effective, are highly invasive, require extensive equipment, and are not readily performed in a field or emergency room setting. The effectiveness of these methods results from their ability to warm internal organs with minimal skin surface heating [ 1,2]. This minimizes the possibility of core temperature afterdrop, i.e. the continued drop in a hypothermia victim’s core temperature after removal from a cold environment and initiation of rewarming. Afterdrop is minimized by eliminating the temper-
0300-9572/95/$09.05 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0300-9572(94)00845-K
204
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
ature gradient between the core and skin surface during the early stages of rewarming [3,4]. Active external methods such as warm bath immersion are also effective but limit access for patient care, including cardiac monitoring, and place the patient at risk for core temperature afterdrop during the rewarming process. For patients with cardiovascular collapse and hypothermia, active external rewarming is contra-indicated and may be fatal, because core temperature afterdrop in these patients may result in ventricular fibrilation [l]. Passive rewarming while useful in the field, is a poor method for victims whose endogenous heat production is depressed due to lowered core temperature. A major goal in hypothermia research has been the development of a field-usable rewarming method capable of non-invasive heating of deep body tissues. Inhaling warm humidified air or oxygen is good at minimizing respiratory heat loss, but its advantage over shivering in warming the body core is; questionable [1,5]. Since inspiring humidified air or oxygen provides roughly 24 k&h, approximately 2.5 h are required to deliver the 58 kcal needed to raise the core temperature of a 70 kg person by 1°C [6]. Radio frequency induction coils are an active non-invasive core rewarming technique, based on selective energy absorption by muscle and bone tissue. By minimizing energy absorption at the skin, radio frequency coils hold the promise of minimizing afterdrop. As sudden deaths among ostensibly recovering hypothermia victims can often be attributed to core temperature afterdrop [l], this would be a signifi-
cant contribution to field treatment. Rewarming of mammals has been reported with radio frequency coils [7,8], but development has been hampered by a proclivity for skin burns. Development by Olsen et al. [9] of a tunable radio-frequency induction coil has led to the successful rewarming of hypothermic rhesus monkeys. As part of a multilaboratory study [lo], we report the testing of this device by rewarming mildly hypothermic human subjects under simulated field conditions. These results were compared with rewarming by 2 other methods: warm water immersion and a thermal rewarming sack. 2. Materials and methods 2.1. Subjects Four males (Table l), ages 24-35 years, volunteered to participate as subjects, after being fully informed of the details of the experimental protocol and associated risks. These procedures were approved by the Naval Air Development Center’s Advisory Committee for the Protection of Human Subjects and the Food and Drug Administration’s Section for Investigational Medical Devices. Weight was recorded prior to each test run and the mean for each subject calculated. Body surface area (SA) was calculated [ll] from the mean weight and height of each subject. Percent body fat was determined from estimates of body density [12], which were computed from skinfold measurements obtained with Lange Skinfold Calipers (Cambridge Scientific Inc., Cambridge, MD) and the equation of Lohman [13].
Table 1 Physical characteristics of subjects Subject
Gender
Age (yea@
Height (4
Weight 0%)
% Body fat
Surface area (m*)
A B C D
M M M M
35 26 30 24
1.78 1.85 1.75 1.79
79 91.4 71.9 67.9
13 17 17 10
1.97 2.16 1.87 1.85
29 3
1.79 0.02
77.6 6.8
14 2
1.96 0.05
Mean S.E.M.
J. W. Kaufinan et al. /Resuscitation 29 (1995) 203-214
Personal
Flotation
205
Device
Fig. 1. Approximate body position during cold water immersions. The flotation device used in the study is de&@d keep the UPPer body at approximately 45O relative to the water surface so that the distance from the water to the mouth is maximized.
2.2. Experimental procedures
Subjects generally reported to the laboratory at the same time of day for each trial (either morning or afternoon) and were given physical examinations by the attending flight surgeon. Subjects were asked to follow their normal diets during the study and abstain from alcohol for at least 24 h prior to a trial. All subjects were non-smokers. Urine samples were collected and analyzed as part of the flight surgeon’s examination of the subject. Subjects wore a bathing suit, cotton tee-shirt, neoprene wet suit booties (thickness, 4.8 mm), anti-exposure mittens, a personal flotation device, and were instrumented with rectal and 6 surface skin temperature sensors. Booties and mittens were worn in an attempt to preclude trial terminations due to low extremity temperatures. Subjects were immersed to the neck in a stirred pool (1.5 m deep x 2.4 m diameter) of 10°C water (T,,,,,) using a personal flotation device for buoyancy (Fig. 1). This placed subjects at roughly a 45“ angle in the water, allowing their upper chest to rise out of the water, and represents the position individuals assume in the water during many survival situations. Subjects also kept their hands out of the water by resting them on the flotation bladders. Immersions were terminated after 90 min had elapsed, rectal temperature (T,) = 35”C, hand or foot temperature (Thand) = lO”C, or when the
exposure exceeded the subject’s tolerance. Following removal from the water, the flotation device, mittens, and booties were removed from the subject, who was then quickly dried with towels rtnd rewarmed. The time from exiting the water to start of rewarming was 2.9 * 1.0 min. by either the radio-frequency water immersion (WW), or a the chamber to the rewarmi
area in a prone
Ambient air temperature 25 f 4.O”C. The ex balanced the order of subject acted as their own control. Triels for individual subjects were separated by at least 48 h. 2.3. Rewarming methods
Rewarming by each of the methods was continued until T, WZIS37.0°C, unless 60 trGn elapsed without a marked rise in T,. When a increase in T, was observed after 60 min, rewarming using either RF or IS was terminated and the subject was rewarmed by WW. 2.4. Radio frequency rewarming coil
The RF rewarming coil consisted of a helical coil of steel cable within aglastic sheath connected
206
J. W. Kaujkan et al. /Resuscitation 29 (1995) 203-214
Radio Frequency
Coil
00 Nonmetallic
RF
Table
POWER 1 SOURCE
1
Fig. 2. Schematic drawing of the RF coil as used in this study.
to a crystal oscillator (Electronic Navigational Instruments, Rochester, NY, model ACG-5) emitting radio frequency energy at 13.56 * 0.01 MI-Ix (Fig. 2). The coil was kept away from the skin surface by 4 pads of Nautilus foam located about the torso. Specific absorption rates (SAR) were adjusted to 2.5 W/kg body weight, by changing the output power of the coil for each subject (based on measured weight). While SARs of 4.0-5.0 Wikg body weight are suggested by animal studies to maximize rewarming rates [7], a lower SAR was used because higher SARs increase the risk of localized hotspots [14]. The RF system was tuned for each subject during rewarming to minimize the amount of reflected power. A RF field density of 2.6 mW/cm2 was measured at a distance of 1.0 m from the RF coil during operation. During this study, the RF coil covered the subject’s torso from just below the clavicle to approximately 10 cm below the level of the umbilicus. All metal was kept clear of the RF field during rewarming to prevent skin burns. Hot spots generated by variations in coil Iit were identified by the subject and the coil adjusted accordingly. 2.5. Mummy-type insulating sack The insulating sack (Domtex International, Yorkshire, UK, model Decupad Thermal Recovery Capsule) was a mummy-type bag with a high pile polypropylene lining designed for recovery
and field treatment of hypothermia victims. The lining is intended to provide thermal insulation and wick moisture from a hypothermia victim’s clothing and skin. The bag completely covered an individual except for the face. Multiple zippers allowed access to all body areas during use. 2.6. Warm water immersion While not normally a practical field rewarming technique, WW was included as the study control, because in most laboratory hypothermia studies it is the chosen rewarming method. Subjects were immersed in stirred 40°C water with their head and extremities (distal to mid-thigh and mid-upper arm) out. Water temperature was continuously monitored with a 26 AWG type T thermocouple. Adding 41°C water to the tub whenever water temperature fell below 39.5”C maintained a roughly constant temperature (40 f l°C). When a positive trend was observed for the change in T, lasting more than 10 min, subjects were permitted to place their extremities in the water. This conforms to a standard laboratory method of warm water rewarming [1,5,15-181. 2.7. Physiological measurements Use of RF energy precluded the use of thermocouple or thermistor temperature sensors during rewarming, because of their metallic composition. To measure temperatures accurately during rewarming, a tibre optic temperature measurement system designed for use in RF fields was used’(Luxtron Corp., Mountain View, CA, model 3000). Dual sensors were inserted 10 cm past the anal sphincter for measuring T, while surface skin temperatures were measured at 6 sites (i.e., forehead, arm (biceps), lateral surface of torso (approx. 4th intercostal space), upper chest, anterior thigh, and lower back. Esophageal temperatures were not obtained because of reluctance on the part of subjects. Thermistors were used for monitoring hand and foot temperatures during prerewarming immersions - as they could not be employed during RF rewarming, they were removed prior to initiating rewarming in all trials. Electrocardiograph (ECG) signals were monitored during pre-rewarming immersions with ECG electrodes (3M, Minneapolis, MN, Red Dot)
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
amplified with isolated ECG amplifiers (Gould, Cleveland, OH, model 4600 series amplifiers), The ECG electrodes were also removed prior to the start of rewarming procedures. Subjects’ heart rates during rewarming were monitored by means of a stethoscope. During RF rewarming, a plastic stethoscope was used. 2.8. Subjective responses Subjective feelings of fatigue, shivering, temperature, and comfort were evaluated every 15 min throughout the exposure period by means of category scales (Table 3). Subjects were instructed to indicate their subjective sensation for each criterion on a l-7 scale where 1 indicated the most pleasant situation and 7 the most unpleasant.
cal-‘), respectively. It was assumed for these calculations that there was no net metabolic contribution to P,, i.e. metabolic heat production was balanced by heat losses to the environment. The power output of the RF coil was calculated from: PRF = @AR)@&,) (watts)
0.8W+%WWt.~l PRW
=
W
(watts)
(1)
t
where Paw = power needed to rewarm subject, Mb = body mass (kg), AT, = T, increase during rewarming (“C), t = rewarming time in hours, 0.83 = body specific heat (cal g-‘“C-‘), 1009 and 0.001164 = conversion factors (g kg-‘) and (W h
(2)
It was assumed for this calculation that there were negligible losses from reflection. As the coil was tuned for individual subjects during rewarming, actual losses due to reflection were generally < lO%, based on measurements of generated and reflected power. Efficiency was then determined by: %Efficiency = Paw/Par X 100
2.9. Calculated values Comparisons of the 3 rewarming methods were based on the T, rewarming rate (AT,&), afterdrop duration (At&, magnitude of T, afterdrop (AT&, and the actual or estimated T, after 60 min of rewarming (T,& [l&19,20]. The AT& was calculated from the slope of the T, first order regression line divided by the duration of rewarming [ 161. The slope was calculated from the time a positive trend was observed in T, following the start of rewarming procedures. The At4 was calculated from the end of immersion to the beginning of a monotonic increase in T,, i.e. the ‘afterdrop phase’ [ 161. The ATd was the difference between T, at the start of rewarming and the minimum T, during rewarming [la]. Values of Treso were linearly extrapolated if rewarming did not last 60 min. The energy transfer efficiency of the RF was determined by estimating the power required to raise T, by the amount observed in each of the RF runs. This was given by:
207
(3)
2. IO. Statistical analysis Physiological data from this study was analyzed using the non-parametric Fri When Variance (ANOVA). detected, the Mann-Whitney U test was used to identify where the differences existed between rewarming methods. The non-pammet& KruskalWallis ANOVA was used to detect dif&rences in subjective responses (as a function of time) between rewarming methods. Regression analysis was used to determine the slopes of the rewarming T, first order regression lines. Missing initial values were estimated by the technique of Winer [21]. Differences were considered significant at the level of P < 0.05. 3. Results 3.1. Induction of mild hypothermia The immersions prior to rewarming had a mean duration of 13.5 f 19.7 min. During the cooling period, a mean T, drop = -1.7 i 0.7”C was observed, with a mean T, at the end of cooling of 35.8 f 0.8”C. The mean T, drop during the cooling period for each rewarming method is given in Fig. 3. While the final T,‘s were higher than desired (i.e. T, were 35.0%), they were found to be significantly lower than the initial T, (P < 0.001). Individual skin site temperatures varied
J. W. kiwfmnn et al. /Resuscitation 29 (1995) 203-214 I
I
3
I
I
I
I
I
I
I
‘..‘....” insulating sack -. warm water radio frequency device
2
1
Y -
0
2 a -1
-2
4
-3 1
I
-0.9
-0.8
I
I
-0.7
I
-0.6
-0.5
I
I
-0.4
-0.3
-0.2
-0.1
*exl&tal Fig. 3. Mean change in rectal temperature (AT, at time i - starting TJ as a function of non-dimensional immersion time during the cooling phase. T’ii has been normalized by dividing time from the start of rewarming (t& by the total duration of a given trial’s cooling phase (&J). Each curve represents mean values for the cold water immersion periods preceding each rewarming method f S.E.M., n = 4. 1.0 0.8 --- .-. .-. -
0.6
insul:tling wck rack warn waler radio freclucncy
device
0.4 0.2 Y u" & -3
__-.--. -*~g?.
0.0
t
I -,..
-0.2
-.
_ ,,Ai_-i
., .-IL..
-. r --
-0.4
-----1
‘1‘-.. l‘-‘-I---r --A_
----
-0.6 r -0.8 -1.0 0
5
10 El:lpScd
H~Will7llillg
15 The,
20 tre,y:,r,,,,
25
30
minutes
Fig. 4. Mean change in rectal temperature (T, at time i - T, immediately prior to rewarming) during the first 30 min of rewarming. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
J. W. Kaufinan
et al. /Resuscitation
29 (1995)
203-214
Table 2 Rewarming rate and estimated or actual Tr, after 60 mm of rewarming for each rewarming method Rewarming technique
Rewarming rate (“cm
Estimated T,, 60 min v-2
Radio frequency
Warm water
Insulating sack
0.54 (* 0.2) 36.1 (* 0.2)
1.82 (* 0.5) 31.5 (a 0.3)
0.85 (* 0.3) 36.4 (* 0.3) __l_._ll___.
Values are means f S.E.M.
according to distance from the water. The immersed skin sites had temperatures approaching Atwater at the end of cold water immersion (Tback = 11.0 f 0*8”C, Tsi& = 13.8 f 2*5”C, Tthigh = 10.2 f O.YC). The other skin sites were considerably warmer (T’r,,=k-,-~ = 32.8 f l.O”C, r,,,, = 25.6 f 6S’C, T,, = 17.0 f 6.O”C). The relatively high Tchestwas due to the position assumed by a subject’s body while floating in water supported by a personal flotation device. Mean heart rate (HR) at the end of the cooling phase = 100 f 14 beats/mm (bpm) with no ECG abnormalities observed during these immersions. No significant differences between rewarming methods were observed for T,, skin temperatures, or HR during induction of mild hypothermia. 3.2. Rewarming Fig. 4 shows the mean AT, for each rewarming method over the first 30 min. At minute 30, one of the WW subjects had completed the rewarming process, so this represents the minimum time re-
quired for rewarming all subjects. The mean duration of rewarming was 46.0 * 7.1 min for WW and 76.8 f 9.1 min for RF. In 2 cases, IS rewarming proved to be too slow and had to be augmented with WW after 60 min. No duration is reported for IS for this reason. 3.3. Duration and magnitude of rectal temperature afterdrop A significantly shorter Atad was observed during the WW trials compared with the IS trials (P C 0.05). No other significant di%?rences in At,, were noted between the rewarming methods (Fig. 4). ATad was significantly larger in the IS trials compared with WW (P < 0.05) and RF (P < 0.03) trials. The differences in AT,, between RF and WW trials were not observed to be significant (Fig. 4). 3.4. Rate of increase of T,, during rewarming Warm water rewarming (WW) resulted in a significantly larger AT,& than either the RF coil
Table 3 Category scales used to evaluate subjective responses to environmental conditions Scale
1 2 3 4
Category Comfort
Fatigue
Shivering
Temperature
Extremely comfortable
Extremely energetic
No shivering
Very hot
Neutral
Neutral
Neutral
Extremeiy uncomfortable
Extremely exhausted
Severe shivering
5 6 7
Values are means f S.E.M.
Very cold
J. W. Kaufma et al. /Resuscitation 29 (1995) 203-214
210
radio frequency device r:lclio
--
.9 E
100
60
50 50
I 0
I 10 10
5
I
I
15
20
1 25
30
Fig. 5. Mean heart rate during the first 30 min of rewarming. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
, “i’,A--fr
25
--!2----p-~--q--~-~
b:lck, WW
0 0
5
W)scd
I
1
I
I
10
15
20
25
retvamirlg
tirlle, trC.,,,,,l.,,,,lllinules
Fig. 6. Mean changes in back and chest skin temperatures during the first 30 min of rewarming. Abbreviations for the rewarming methods are: IS - insulating sack, RF - radio frequency device; and WW - warm water. Each curve represents mean values for each rewarming method f S.E.M., n = 4.
J. W. Kaufm
et al. /Resuscitation 29 (1995) 203-214
(RF) (P < 0.03) or the thermal sack (IS) (P < 0.03). No significant difference was observed between the RF and IS (Table 2). A significantly greater Tti was observed for WW than either RF or IS (P c 0 0.01) (Table 2). 3.5. Heart rate No significant differences between methods were observed for heart rate during the rewarming trials (Fig. 5). The initially elevated HR observed for IS and WW correspond to bouts of shivering. During RF trials, shivering was probably sup pressed initially by the body movements required to don the RF coil. These findings indicate that there was no significant difference in physiological stress during rewarming with either RF, WW, or IS. 3.4. Skin temperatures WW rewarming resulted in significantly higher skin temperatures throughout most of rewarming (P < 0.01) compared with RF and IS. Fig. 6 show back and chest temperatures, which together represent the part of the body (i.e. torso) where the RF deposited most of its energy. No significant differences in skin temperatures were noted between RF and IS. 3.7. Subjective responses No sign&ant differences were observed between rewarming methods for any of the subjective measures recorded in this study. Differences in subjective responses between subjects were also not significant during rewarming. 3.8. RF efficiency The mean efficiency of the RF coil calculated from equation [3] was 32 i 11%. An important factor in the observed variability in efficiency was differences in measured reflected power between subjects, which was probably due to the quantity and distribution of body fat. In addition, while the RF coil had a focused field, there were stray losses to the surroundings which contribute to lowered efficiency but should be similar between runs. 4. Discrasioo From a theoretical standpoint,
RF should pro-
211
vide a superior means of increasing T, while minimizing afterdrop compared to either WW or IS [2,8-lo] based on the RF’s m&h& of operation. This was not observed in this study the AT,&, At,,,+ and AT4 results. It is that the factors contributing to these results include: (a) a lower SAR than required to quickly rewarm mildly hypothermic humans; (b) i&e&d power and energy absorption by the ski% subcutaneous tissues, and surface moisture; and (c) lack of insulation between the skin surface and the ambient environment, allowing heat losses across the skin surface. The T,& for RF of O.lPC/h observed in #he present study, at a mean rewarming starting T, of 35.9”C, compares unfavorably with ear&~ work with RF coils. Rhesus monkeys, when initially Tn/t of cooled to T, of 28.3”C, d -ted 5.6”Clh at a SAR of 5.5 Wikg body weight [22], while hypothermic dogs cooled to a perature (T,) of 25’C, had a T@ of 5 SARs of 4-6 W/kg body weight @j. Thi that an increased SAR would probabty merease T,.&. The increased risk of skin burna, however, becomes a particular concern in a f&d situation where measuring skin temperatures are diEcult.
a linear relationship between SAR and rewarming rate appears ill-advised. The relatively low efticiency in tranr&rring RF energy to the body can be attributed to a number of factors. The most obvious is losses due to reflected power, although careful tuning limited losses to less than 10%. Surf&e absorption is known to occur with RF energy [7,14] with upwards of 25% of the energy being deposited there [R. Olsen, personal communication]. Water on the skin surface was also recognized as a pot&al energy sink. In an attempt to address this problem, subjects were dried off with a towel immediately prior to rewarming. While this was generally successful in removing excess water, there is no doubt that some water was retained which would have reduced the efficiency of the RF. Water retained in clothing could pose a significant problem when using RF for field rewarming. The lack of insulation between the skin surface
212
J. W. Kaufman et al. /Resuscitation 29 (1995) 203-214
and the ambient environment during RF rewarming resulted in continual heat losses. Increasing skin temperatures increased the temperature gradient between the skin and environment during RF. Rising skin temperatures, while precluding vasoconstriction during either RF or WW, had the effect of eliminating the only means available to minimize transcutaneous heat losses during RF under the study conditions. Transcutaneous heat losses were not a problem with either IS or WW because WW utilizes the thermal gradient between skin and water to transfer heat to the body, while IS uses insulation to minimize heat transfer between the skin and ambient environment. Future work with the RF coil should include clothing, which will increase skin surface insulation so that this source of heat loss is mitigated and provide for a more realistic test of RF rewarming in a field scenario, though exacerbating the problem of energy absorption by water at the skin surface. A previous study by Hesslink et al. [lo] reported that RF was superior to WW. This finding was based on the more rapid responses of T,, to RF than either WW or IS, though they also found that T, responded more quickly to RF [lo]. The T, rewarming rate for WW of 1PC/h observed in the present study is similar to those found in other WW studies by Romet [19] of 1.6Wh and Hoskin et al. [16] of 2.3”C/h, but considerably larger than 0.3Wh found by Hesslink et al. [lo]. While Hoskin et al. [ 161 began rewarming after a AT, of 1.6”C during the cooling period, similar to the present study’s pre-rewarming AT, of 1.9”C, Romet [19] had a pre-rewarming AT= of OS’C. Curiously, Hesslink et al. [lo] also had a AT, of 0.5”C prior to rewarming, despite a considerably smaller overall WW T, rewarming rate. This anomalous result may be due to the position in the tub during rewarming, i.e. Hesslink et al. [lo] kept the upper chest and neck out of the water in contrast to the other studies which immersed people up to the neck. Differences in the extent of body cooling may have also contributed to the conflicting observations between this work and that of Hesslink et al. [lo]. In the present study, immersed skin temperatures asymptotically approached Twater toward the end of the cooling phase. During the
time between cold water extraction and initiation of rewarming, Tbck increased only 3.7”C. Back temperatures reported in the Hesslink study prior to rewarming were at least 16°C warmer than water temperature, which suggests only slight body cooling [lo]. The quasi-steady state temperature losses observed in longer duration cold exposures [17,18,24] had apparently not been achieved. Not only would less energy be required to raise body temperatures, but transcutaneous heat losses would be minimized because of the small temperature gradient between the skin and ambient air. The rapid response of T, compared with T, observed in the Hesslink et al. study [lo] may represent the fact that the mass of the body was not deeply cooled. Mittleman and Mekjavic [3] demonstrated that T,, is influenced by peripheral blood temperatures while T, is not. Thus the relatively warm back and triceps temperatures observed by Hesslink et al. [lo] suggest that the peripheral blood supply enhanced the observed T, rebound. As T, is dependent upon convective heat exchange from the central blood supply and conductive heat exchange with surrounding tissues, the relatively warm peripheral tissues at the start of rewarming would selectively aid an increase in T,, without directly influencing T,. A number of practical considerations must also be addressed if RF is to become a useful field rewarming method. Radio frequency interference with electronic devices precluded continuous metabolic measurements and ECG monitoring during the present study. This would be unacceptable when dealing with cases of severe hypothermia, where cardiac monitoring is essential, and during aeromedical transport, when navigational device would likely be affected. There is also a possible danger to crew members because they would be exposed to unshielded RF waves. Identification of local hot spots and adjustments to the coil to avoid skin bums was dependent upon an alert and cooperative subject. For field use, a more dependable method of skin temperature monitoring would be required. In addition, adjustments to the coil could be difficult or hazardous with an unconscious or injured hypothermia victim. These deficiencies can possibly be corrected with further development,
J. W. Koufin
et al. / Resuscitution
but they must be borne in mind when discussing an RF rewarming coil as a potential field device. This study demonstrated the effectiveness of WW rewarming as a means of quickly raising T, in mildly hypothermic individuals. Afterdrop during WW rewarming was not as dramatic as suggested in earlier literature [10,25-271. Its simplicity and effectiveness support the continued use of WW as a method of choice for rapid rewarming of hypothermic individuals under the test conditions. RF and IS proved to be roughly equivalent in the ability to rapidly increase T,, though use of IS was observed to produce a greater afterdrop. Both RF or IS were observed to be significantly less successful in rapidly increasing T, during rewarming than WW. In conclusion, WW rewarming was observed to be the most effective means of rewarming mildly hypothermic individuals when compared with RF or IS. While little difference was found between RF and IS in the present study, these results cannot be extrapolated to cases of severe hypothermia. Since metabolic energy production is diminished in hypothermic individuals, active rewarming by RF has a theoretical advantage over IS, which depends on the passive retention of metabolically generated heat. Given these considerations, a study in which individuals with differing metabolic rates are rewarmed could be used to differentiate between the effectiveness of RF and IS. Considerable work has yet to be done to conclusively demonstrate that RF can rewarm at least as effectively as WW when SARs are increased.
Acknowledgements The authors wish to thank the subjects of this study, whose perseverence despite considerable personal discomfort made this work possible. We would also like to thank Walter Soroka, LCDR Angus Rupert, and the Biomedical Support Group (Code 6025) for their technical support, without whom completion of this study would not have been possible. The views expresses herein are solely those of the authors and should not be construed as the
29 (19951
203-214
official policy or position of the Department Defense or Department of the Navy.
213
of
References 111Hamett RM, Pmitt JR, Sias FR. A review of the literature concerning resuscitation from hypothermia: Part I - The problem and general approaches. Aviat Space Environ Med 1983; 54: 425-434. PI Hamett RM, Pruitt JR, Sias FR. A review of the literature concerning resuscitation from hypothermia: Part II - Selected rewarming protocols. Aviat Space En&on Med 1983; 54: 487-495. 131 Mittleman KD, Mekjavic IB. Effect of o&&d venous return on core temperature during cold water immersion. J Appl Physiol 1988; 65: 2709-2713. [41 Webb P. Afterdrop of body temperature during rewarming: an alternative explanation. J Appl Physiol 1986; 60: 385-390. 151 Marcus P. Laboratory comparison of techniques for rewarming hypothermic casualties. Aviat Space Environ Med 1978; 49: 692-697. I61 Bangs CC. Hypothermia and Frostbite. Emerg Med Clin N Am 1984; 2: 475-487. I71 Olsen RG, Ballinger MB. David TD, Lotz WG. Rewanning of the hypothermic rhesus monkey with electromagnetic radiation. Bioelectromagnetics 1987; 8:!83-193. 181 White JD, Butterfield AB, Greer KA, Schoem S, Johnson C, Holloway RR. Comparison of rewarming by radiowave regional hyperthermia and warm humidified inhalation. Aviat Space Environ Med 1984 [91 Olsen RG. Reduced temperature afterdrop in rhesus monkeys with radio frequency rewarming. Aviat Space Environ Med 1988; 59: 78-80. 1101Hesslink RL Jr, Pepper S, OIsen RG, Lewis SB, Homer LD. Radio frequency (13.56 MHz) energy enhances recovery from mild hypothermia. J Appl Physiol 1989; 67: 1208-1212. 1111 Dubois EF, Dubois D. Measurement of surface area of man. Arch Int Med 1915; 15: 868. 1121Brozek J, Grande F, Anderson JT, Keys A. Densiometric analyses of body composition: revision of some quantitative assumptions. Ann New York .Acad Sci 1963; I 10: 113-140. 1131Sinning WE, Dolny DG, Little KD, Cunningham LN, Racanielli A, Siconolfi SF, Sholes JL. Validity of “generalized“ equations for body composition analysis in male athletes. Med Sci Sports Exert 1985; I?: !24- 130. [I41 D’Andrea JA, Emmerson RY, Dewitt JR, Gandhi OP. Absorption of microwave radiation by the anesthetized rat: Electromagnetic and thermal hotspots in body and tail. Bioelectromagnetics 1987; 8: 385-396. 1151 Daanen HAM, Van De Linde FJG. Comparison of four noninvasive rewarming methods for miId hypothermia. Aviat Space Environ Med 1992; 63: !070- IO76
214
J. W. KaufmMet al. /Resuscitation 29 (1995) 203-214
[16] Hoskin RW, Melinyshyn MJ, Romet ‘l-f, Goode RC. Bath rewarming from immersion hypothermia. J Appl Physiol 1986; 61: 1518-1522; [17] Kaufman JW, Dejneka KY, Bagian JP, Morrissey SJ, Bittner Jr, AC. Cold water evaluation of NASA Launch Entry Suit (LES). Naval Air Development Center, Warminster. PA, Tech. Rept. NAJXXO-017, July, [18] Kaufman JW, Dejneka KY. Cold water evaluation of constant-wear anti-exposure suit systems. Naval Air Development Center, Warminster, PA, Tech. Rept. NADC-8509240, June, 1985. [19] Romet TT. Mechanism of afterdrop after cold water immersion. J Appl Physiol 1988; 65: 1535-1538. [20] Romet Tf, Hoskin RW. Temperature and metabolic responses to inhalation and bath rewarming protocols. Aviat Space Environ Med 1988; 59: 630-634. [21] Wirier BJ. Statistical Principles in Experimental Design. New York: McGraw-Hill, 1962. [22] Olsen RG, David TD. Hypothermia and electromagnetic
[23]
[24]
[25] [26] [27]
rewarming in the rhesus monkey. Aviat Space Environ Med 1984; 55: 1111-1117. Morrison JB, Conn ML, Hayward JS. Accidental hypothermia: The effect of initial body temperatures and physique on the rate of rewarming. Aviat Space Environ Mcd 1980; 51: 1095-1099. Steimnan AM, Kubihs PS. Survival at sea: The effects of protective clothing and survivor location on core and skin temperatures. U.S. Coast Guard Rept. No. CG-D 26-86. Nat. Tech. Info. Serv. Springfield, VA, 1986. Collis ML, Steimnan AM, Chaney RD. Accidental hypothermia: An experimental study of practical rewarming methods. Aviat Space Environ Med 1977; 48: 625-632. Hayward JS, St&mum AM. Accidental hypothermia: An experimental study of inhalation rewarming. Aviat Space Environ Med 1975; 46: 1236-1240. Hayward JS, Bckerson JD, Kemna D. Thermal and cardiovascular changes during three methods of resuscitation from mild hypothermia. Resuscitation 1984, 11: 21-33.