|Home | About | Journals | Submit | Contact Us | Français|
Jeremy R. Hawkins, MS, ATC; Kenneth L. Knight, PhD, ATC; and Blaine C. Long, MS, ATC, contributed to conception and design; acquisition and analysis and interpretation of the data; and drafting, critical revision, and final approval of the article.
Context: Therapeutic modality control variables are thought to be thermal neutral, a term sometimes used interchangeably with room temperature. We question this common assumption.
Objective: To determine thermal neutrality of common therapeutic modality control variables.
Design: We performed 5 laboratory experiments, including (1) water temperature over 3 weeks in 3 different containers (glass, plastic, and polystyrene); (2) water temperature and volume of 4 beakers (2 insulated, 2 uninsulated) over 4 weeks, with 1 beaker of each type covered by polyethylene; and skin interface temperature of (3) a dry, nonheated hydrocollator pack held against the chest, (4) kitty litter applied to the knee, and (5) room-temperature ultrasound gel to the forearm.
Setting: Therapeutic modalities laboratory.
Patients or Other Participants: College student volunteers were subjects in experiments 3, 4, and 5.
Main Outcome Measure(s): We measured temperature and volume change. Data were evaluated using descriptive and interferential statistics.
Results: Water temperature plateaued significantly below room temperature. Temperatures significantly increased in all but the open, insulated container. Open containers plateaued at approximately 2°C below room temperature and lost significant amounts of water; closed containers plateaued at room temperature with negligible water loss. In experiments 3 through 5, skin temperatures rose significantly during hydrocollator pack, kitty litter, and ultrasound gel application.
Conclusions: Room-temperature water baths, dry hydrocollator packs, kitty litter, and ultrasound gel were not thermally neutral. Room temperature should not be used synonymously with thermal neutral. Care must be taken to ensure that control variables truly are controlled.
Control variables are essential in scientific research. They indicate how a subject or system reacts under normal conditions (ie, without intervention). In therapeutic modality research, a control variable must be thermal neutral (ie, the same temperature as the body part to which the modality is applied, thus not adding to or taking heat away from the body) in order to ensure that normal conditions are present. One common way of describing thermal neutral is to say the variable is at room temperature. 1–6 In fact, thermal neutral and room temperature often are used interchangeably. Using these terms synonymously leads one to believe that room-temperature substances do not affect the temperature of the body part to which they are applied. However, a substance at room temperature is cooler than the body, so when it is applied to the body, it extracts heat, making it not thermal neutral and not a good control variable.
During recent research, we observed that “room-temperature” water baths were 4°C to 6°C below room temperature after 2 to 4 weeks in a laboratory. 1 During those same weeks, room temperature remained virtually unchanged (between 22°C and 25°C). This raised questions about the temperature of water and other substances (eg, ultrasound gel and kitty litter) that have been used as control variables during therapeutic modality research. Therefore, our purpose was twofold: (1) to determine whether it is appropriate to use the term room temperature in referring to control variables in therapeutic modality research and (2) to determine what effect a room-temperature variable has on interface temperatures (ie, are they thermal neutral?). We conducted 5 experiments to answer the following questions:
Time was the independent variable for all 5 experiments. Additional independent variables included container type for experiment 1 (glass, plastic, or polystyrene), container insulation (with or without polystyrene) and covering (with or without polyethylene) for experiment 2, and presence or absence of a control substance (room-temperature dry, unheated hydrocollator pack, kitty litter, or ultrasound gel) for experiments 3, 4, and 5. Temperature was the dependent variable in each experiment. Water volume was an additional dependent variable in experiment 2.
Temperature (°C) was measured with PT-6 thermocouples (Physitemp Instruments, Inc, Clifton, NJ) interfaced with an Iso-Thermex electrothermometer (−20°C to 80°C; Columbus Instruments, Columbus, OH) for all but experiment 2. During experiment 2, we used a calibrated mercury thermometer (model 15-059-18; Fisher Scientific International Inc, Hampton, NH; National Institute of Standards and Technology [NIST] 9 traceable) graded at 0.1°C. Water volume (mL) in experiment 2 was measured using a 1000-mL graduated cylinder. Ambient temperature and humidity were measured during experiment 2 using a NIST-traceable digital thermometer (model 15-077-22; Fisher Scientific International Inc). A 12 × 12-in (30.48 × 30.48-cm) hydrocollator pack (Dynatronics Corp, Salt Lake City, UT) was used in experiment 3. Kitty litter (Ori-Dri Corp of America, Chicago, IL) was used as the noncooling substance in experiment 4. Blue Aquasonic ultrasound gel (Parker Laboratories, Inc, Fairfield, NJ) was used in experiment 5.
Experiments 3 through 5 involved human subjects. A total of 18 subjects (female: n = 13, age = 20.8 ± 1.5 years, height = 165.1 ± 7.1 cm, mass = 66.7 ± 11.5 kg; male: n = 5, age = 22.8 ± 1.3 years, height = 190.5 ± 3.3 cm, mass = 84.2 ± 4.8 kg) volunteered for experiment 3. Four female subjects (age = 20.0 ± 0.0 years, height = 167.6 ± 7.5 cm, mass = 66.5 ± 8.5 kg) volunteered for experiments 4 and 5. Our institutional review board approved the study, and each subject gave written informed consent before participating.
Data were evaluated for each experiment separately, using descriptive statistics (mean ± SD), either 1-way or repeated-measures analysis of variance, followed by Tukey post hoc analysis or paired t tests where appropriate. Results were considered statistically significant at an α level of P ≤ .05.
For increased comprehension and clarity, experimental procedures and results for each experiment are presented together.
Three containers (glass, plastic, and polystyrene) were filled with 500 mL of 18.9°C tap water. The temperature of each container and ambient room temperature were measured every hour with an Iso-Thermex electrothermometer using 1 PT-6 thermocouple secured to the inside of each container and an additional PT-6 thermocouple approximately 1.22 m (4 ft) away from the containers. The central heating and cooling system within the building regulated the temperature of the room. We conducted 5 trials, each 105 hours long, with the containers sitting on a vinyl-upholstered plinth.
Room temperature was 22.9°C ± 0.1°C during the 5 trials. Water in the 3 containers increased from 1.5°C to 3°C in the first 8 hours and remained virtually unchanged during the remainder of the experiment (glass = 21.4°C ± 0.2°C, plastic = 21.8°C ± 0.2°C, polystyrene = 20.4°C ± 0.4°C; Figure 1). The water temperatures were below room temperature, and the temperature of the water in the polystyrene was below that in the glass and plastic containers (F 9,27 = 4.72, P < .001, Tukey).
Water temperature and volumetric changes were measured in 4 glass beakers, 2 of which were insulated with polystyrene. One of the insulated and 1 of the uninsulated beakers also were covered with polyethylene and secured with Transpore surgical tape (3M, St Paul, MN) to limit water evaporation. All 4 beakers initially contained 750 mL of 17.1°C tap water. The beakers were placed on a polystyrene base to limit heat exchange with the plinth. 10 Maximum and minimum ambient room temperature and relative humidity were measured daily using a digital NIST thermometer. Water temperature was recorded every 7 days as follows: a mercury thermometer was inserted into the beakers and allowed to stabilize (typically occurring in less than 5 minutes). We measured temperature at 2, 3, and 4 minutes. If the temperature change was less than 0.1°C, we proceeded with the experiment. If a change occurred, we continued to measure temperature until we received 3 consecutive readings within 0.1°C of each other, representing stability. Water temperature then was measured 3 times at 1-minute intervals. After recording the temperature, we recorded water volume using a 1000-mL graduated cylinder. The previously covered beakers were covered again using a new piece of polyethylene, secured with tape. We conducted 3 trials, each 4 weeks long.
Room temperature was 22.3°C ± 0.5°C (range, 22.0°C to 25.0°C) with relative humidity of 34.4 ± 7.6 mm Hg (range, 23.0 to 55.0 mm Hg) during the 3 trials. Temperature increased in all but the open, insulated container between the initial reading and week 1 (F 12,36 = 189.2, P < .001, Tukey; Figure 2). The open, uninsulated container plateaued at approximately 2°C below room temperature, whereas the 2 closed containers plateaued at room temperature. Water volume decreased in both open containers by week 3 (F 12,12 = 258.2, P < .001, Tukey; Figure 3). The open, uninsulated container lost approximately 130 mL per week, whereas the open, insulated container lost about 110 mL per week.
In experiment 3, we replicated a cryokinetics protocol in which a patient holds a moist heat pack to the chest while immersing an ankle in an ice bath. The experiment lasted for 45 minutes. Subjects (n = 18) sat in a chair for 5 minutes while a 5-minute baseline temperature was established for the dry, unheated hydrocollator pack. The subjects then held a dry, unheated hydrocollator pack to the chest during a 20-minute ice-bath immersion. For the remaining 20 minutes, the subjects alternated between holding and not holding the hydrocollator pack, 5 minutes off and 5 minutes on, repeated twice. The 5 minutes off was meant to replicate the time taken to perform exercises. During this time, the dry, unheated hydrocollator pack was set on a vinyl-upholstered plinth. While subjects held the hydrocollator pack to the chest, we recorded interface temperature using a PT-6 thermocouple interfaced to an Iso-Thermex electrothermometer. 7 Temperature measures were recorded every 30 seconds.
The mean dry, unheated hydrocollator pack temperature (21.9°C ± 0.1°C) was less than room temperature (22.9°C ± 0.2°C, t 2 = −13.0, P < .01) before application. The mean temperature before application and at 20, 30, and 40 minutes was determined. The hydrocollator pack interface resulted in an increase in temperature between each pair of points (preapplication mean and 20-minute mean, 20-minute mean and 30-minute mean, 30-minute mean and 40-minute mean) analyzed (F 3,4 = 5539.9, P < .001). We observed an initial 0.8°C decrease in interface temperature immediately after application, followed by a steady increase to a final temperature approximately 7.5°C above the baseline readings ( Figure 4).
In experiment 4, we tested the theory of a sham ice-bag treatment. We measured skin interface temperature using a noncooling substance (kitty litter). 8 We made a crushed ice bag, weighed the bag, then filled another bag with kitty litter to an equivalent weight (approximately 700 g). With subjects (n = 4) seated, 2 PT-6 thermocouples were taped over the anterior thigh. A bag of kitty litter was then secured to the anterior thigh using a double-length 15.24-cm (6-in) elastic wrap. An additional PT-6 thermocouple was placed inside the bag of kitty litter. Temperature measurements were recorded for 10 minutes to establish a baseline, after which the kitty litter was applied for 20 minutes. After the 20-minute treatment, the kitty litter was removed, and we continued to record the temperature for 10 minutes to determine the effect of kitty litter on skin temperature and of skin temperature on kitty litter. Temperature was recorded every 30 seconds with an Iso-Thermex electrothermometer throughout the experiment. The average temperatures during the 5 minutes before application and the last 5 minutes of application were used to compare the 2 conditions.
The application of kitty litter resulted in an increase ( t 9 = −22.7, P < .001) in skin interface temperature (30.3°C ± 0.5°C to 30.9°C ± 0.7°C). This increase was not maintained during the 10-minute postapplication period. When the kitty litter was applied to the leg, the temperature within the bag also increased ( t 9 = −35.2, P < .001), from 23.4°C ± 0.2°C to 25.3°C ± 0.4°C. This increase in temperature was maintained throughout the 10-minute postapplication period. Room temperature was 22.9°C ± 0.2°C during the 4 trials ( Figure 5). The initial temperature within the bag of kitty litter was different ( t 9 = −21.5, P < .001) from room temperature.
In experiment 5, we measured skin interface temperature under room-temperature ultrasound gel. 2, 3 Two PT-6 thermocouples were taped to the anterior thigh, with the tips inside a template that was twice the size of the effective radiating area (5-cm 2 transducer head). The template served to standardize the size of the treatment area and the amount of ultrasound gel being applied. Temperature measurements were recorded for 10 minutes to establish a baseline, after which a single layer of ultrasound gel (approximately 0.5-cm thick) was applied and left in place for 20 minutes. After 20 minutes, we wiped the gel off and continued to record temperature measurements for an additional 10 minutes to determine the effect of ultrasound gel on skin temperature. Temperature measurements were recorded every 30 seconds with an Iso-Thermex electrothermometer throughout the experiment. Temperature within the ultrasound gel container also was measured with an Iso-Thermex electrothermometer, every minute for 50 minutes. The average temperatures during the 5 minutes before application and the last 5 minutes of application were used to compare the 2 conditions.
Applying ultrasound gel to the anterior thigh resulted in an approximate 5.0°C decrease in skin interface temperature during the 20 minutes of application (30.1°C ± 0.5°C to 25.5°C ± 0.5°C, t 9 = 119.1, P < .001; Figure 5). During the 10-minute postapplication period, skin interface temperature increased to 26.2°C ± 0.9°C but did not reach preapplication temperature. Room temperature was 22.9°C ± 0.2°C throughout the trials. The temperature within the ultrasound gel container (22.8°C ± 0.1°C) was different from room temperature ( t 10 = −2.67, P = .01).
The need for thermal-neutral control variables in therapeutic modality research is self-evident. In order for a substance to be thermal neutral, it must be the same temperature as the body part to which it is applied. This was not the case in our experiments.
Regardless of container type or insulation, water in an open container will not equilibrate to room temperature because evaporation removes thermal energy from the water. The tendency for water molecules to arrange themselves through hydrogen binding is outweighed by the energetic push for randomness at room temperature. 11 The energy generated through the breaking of the hydrogen bonds—44 kJ/mol 11—results in water evaporation. Preventing evaporation by closing the container, whether or not the container is insulated, will allow it to equilibrate with room temperature. Because water in an open container will not equilibrate with room temperature, it is incorrect to refer to this water as being at room temperature.
Our initial question was why tap water left for up to 30 days in a room did not equilibrate to room temperature. The answer to that led to our questioning the thermal neutrality of other control substances. Researchers trying to determine the effect of a modality on temperature change or afferent sensory stimulus must have a control that will not itself change the temperature of the body. As demonstrated by the hydrocollator pack, kitty litter, and ultrasound gel experiments, this becomes a matter of appropriate terminology: are thermal neutral and room temperature synonymous?
The hydrocollator pack, kitty litter, and ultrasound gel applied at room temperature withdrew heat from the body and, therefore, were not true controls (ie, thermal neutral). This finding is evidenced by the increase in hydrocollator pack and kitty litter interface temperatures and the decrease in ultrasound gel interface temperature. To control for the results observed, these substances must be near the same temperature as the skin to which they are applied. Skin temperature varies by individual and body part, so the temperature of the control variable must vary also. It is appropriate to refer to these control variables as being room temperature; each variable differed minimally from the room temperature. However, the only time a control variable and room temperature should be considered thermal neutral with the skin is if they are the same temperature as the skin, approximately 33°C. 12
Water temperature plateau below room temperature occurs when heat is allowed to dissipate through evaporation. Sealing the container allows the water to reach room temperature. Additionally, it is not appropriate to use the terms thermal neutral and room temperature interchangeably. Control variables at room temperature withdrew heat from the skin and, therefore, were not thermal neutral, as demonstrated by experiments 3, 4, and 5. We propose skin temperature and thermal neutral as synonymous terms, but additional experiments need to be conducted to determine if this is indeed the case. Additional research also is needed to determine if control variables at room and skin temperatures have any effect on the outcomes of a sensory perception study.