Measurement of reactive vasodilation during cold gel pack application to nontraumatized ankles.

The use of cold modalities such as ice packs, gel packs, and ice massage to decrease the effects of the local inflammatory response created by soft tissue trauma is widely accepted. Tissue cooling produces an initial vasoconstriction in the cutaneous blood vessels, which results in a decrease in local blood volume. This reaction is explained by Olson and Stravino as the body seeking to preserve the internal core temperature by alterations in the peripheral or skin temperature through vasomotor regulation of small blood vessels in the skin.(1) (p840)

There is uncertainty, however, regarding the responses that follow this intense vasoconstriction. In 1930, Lewis(2) found that skin temperature initially fell rapidly when a subject's finger was immersed in ice water. This temperature drop was replaced after the first few minutes with a spontaneous rise in skin temperature. The rise in skin temperature occurred when the tissue temperature of the finger reached 15[degrees]C. Cyclic phases of cooling and rewarming followed. Lewis called this phenomenon a "hunting response." First identified in the fingers, Lewis subsequently identified this response in the toes, nose, ears, and chin.

In a later experiment, Clarke and Hellon(3) reported a marked vasodilation in local vasculature upon submersion of the forearm in a water bath at a temperature of 1[degrees]C. Fox and Wyatt(4) reported that cold-induced vasodilation occurred over the patella; in the buttocks, the perianal region, and the chest, close to the nipples; and in all areas of the head and neck except the vertex of the skull. Neither Clarke and Hellon nor Fox and Wyatt reported Lewis's (2) hunting response. Instead, both studies reported initial

C Taber, PT, is Associate Professor, Department of Physical Therapy, Northern Arizona University, PO Box 15105. Flagstaff, AZ 86011 (USA). He is also Director of the University Physical Therapy Clinic. Address all correspondence to Mr Taber.

K Contryman is a student in the Master's Program in Physical Therapy at Northern Arizona University.

J Fahrenbruch is a student in the Master's Program in Physical Therapy at Northern Arizona University.

K LaCount is a student in the Master's Program in Physical Therapy at Northern Arizona University.

MW Cornwall, PhD, PT, is Assistant Professor, Department of Physical Therapy, Northern Arizona University.

This study was approved by the Northern Arizona University Human Subjects Review Committee. vasostriction in response to cold application, followed by slow, steady vasodilation. Termed reactive vasodilation, this localized response to cold application was thought to occur as a protective measure to prevent soft tissue death attributable to freezing.

Whether reactive vasodilation or a more cyclic hunting response occurs, a vasodilation response has important implications for the application of a cold modality to control local edema following soft tissue trauma. The vasodilation would result in an increase in the local blood volume. This increase in blood volume, superimposed upon the existent increase in local capillary permeability as part of the inflammatory response to tissue trauma, would significantly reduce a cold modality's ability to control local tissue swelling.

Two recent studies, however, do not support the existence of reactive vasodilation in response to local cold application. Using a strain-gauge plethysmograph to measure blood flow change, Knight and Londeree(5) reported that vasodilation did not occur in the ankle when the local tissue temperature was decreased to 7[degrees]C for 25 minutes. In a subsequent study involving the fingers, Knight and colleagues(6) again concluded that reactive vasodilation does not occur.

It is evident that the effect of cold on peripheral vascular dynamics is controversial. In a 1983 review of the literature on the physiological effects of cold, Kowal(7) concluded that reported findings are inconsistent regarding cold's effect on swelling and peripheral vascular dynamics. Meeusen and Leivens, in their 1986 review of the literature on the use of cold for athletic injuries, stated that at present the literature provides us with indications for the duration and temperature of the application methods which are purely speculative and chosen on an arbitrary basis or empirically supported.(8)(p411)

Research regarding the physiological effects of local cold application has also been criticized based on the instrumentation used to measure these effects. Early reports used measurement of the subject's skin temperature to indirectly indicate the vasoconstriction-vasodilation process. It was assumed that when skin temperature increased, vasodilation was the cause. Measurement of temperature change with such devices as thermistors, however, is not a direct indicator of blood volume change and thus would not directly measure vasoconstriction and vasodilation changes.

Plethysmographic measurement of limb blood volume offers a more direct method by which to measure the vasoconstriction-vasodilation process. There is extensive research reported in the literature regarding plethysmography, with strain-gauge and impedance plethysmography the most frequently described methods. Mercury strain-gauge plethysmography, as first described by Whitney,(9) has been further described in the literature.(3,10-12) Mercury strain-gauge plethysmography requires meticulous attention to detail and measures the volume change in only a small portion of the limb. As a result, mercury strain-gauge plethysmography is time-consuming and results are subject to a variety of errors.(3,9,10) Impedance plethysmography is safe, easily applied, and, as noted by Anderson,(13) Golden and Miles,(14) and Nyboer,(15) can accurately measure blood volume change in an extremity.

The purpose of this study was to determine the occurrence of reactive vasodilation during application of a cold gel pack over a time period common to clinical practice. Our null hypothesis was that a cold gel pack would not decrease local blood volume significantly more than application of a room-temperature pack of equal weight.

Method

Subjects

A pilot study was initially conducted using 7 volunteers (2 male, 5 female). The pilot study was used primarily to establish the testing protocol. In addition, the number of subjects needed for the study was estimated from the pilot study, using a power analysis.(16) The results of this analysis indicated that 13 subjects would be needed to have a statistical power level of .80. Based on this finding, 13 nondisabled volunteers (9 male, 4 female) between the ages of 18 and 30 years ([- over X ]=22;SD=3.9) were recruited for inclusion in the principal study.

Prior to participation in the study, each subject completed a medical questionnaire to rule out any type of hypersensitivity to cold and to ensure there was no history of ankle or foot pathology. A verbal explanation of the procedures was given to all subjects, and informed consent was obtained.

Modality Selection

In order to reflect current clinical practice, we sent a survey questionnaire to 60 randomly selected orthopedic physical therapists throughout Arizona to determine which cold modality is most frequently used for local inflammation control and the most common length of time for cold modality application. The 49 returned survey questionnaires indicated that the cold gel pack was the most commonly used modality and that the most common duration of application varied from 10 to 20 minutes. Based on the survey results and current literature,(8) we used 20-minute applications of a cold gel pack in this study.

Instrumentation

Data were collected using a Model 720 Bilateral Impedance Plethysmograph. Impedance plethysmography measures fluctuation in the fluid volume of a segment by recording its bioelectrical impedance, which is the segment's opposition to the flow of electrical current through its tissues. As fluid volume increases in a segment, its resistance to AC current (impedance) decreases. Conversely, as the fluid volume is decreased, impedance increases. The impedance plethysmograph used in this study measures impedance by applying a constant sinusoidal current of 4 mA at a frequency of 100 kHz across the segment. This current, because of its high frequency and low amplitude, cannot be perceived by the subject and will not stimulate nerve or muscle tissue. The change in voltage measured across the limb is directly proportional to the impedance (Ohm's law: Z=V/I).

Procedure

On each day of testing, the subjects were asked to refrain from eating and participating in any type of physical activity for 2 hours prior to data collection. This was done to help ensure a stable extremity blood flow. The experimental session began with each subject sitting for approximately 10 minutes. Blood pressure, resting heart rate, and room temperature were measured. The subject then assumed a prone position on a treatment table, with the knees flexed at 15 to 20 degrees and the feet relaxed over the edge of the table. The right ankle and foot were cleaned with isopropyl alcohol.

Each subject lay prone for 10 minutes before data collection took place. Four 2.5-x3.0-cm electrodes were placed on the lateral aspect of each subject's lower leg, two above the lateral malleolus and two below the lateral malleolus. The two outer electrodes sent and received the current, and the two inner electrodes measured the voltage change of the segment between the current electrodes (Fig. 1). In order to promote representative tissue measurement through the depth of the sampling field of each subject's ankle, the voltage electrodes were placed apart at approximately one third of the ankle circumference measurement and the current electrodes were placed apart at approximately one half of this same measurement, as recommended by

Wheeler and Penney.(17) Final placement of the electrodes was determined by the manufacturer's specifications that a beginning resistance range of 46 to 54[Omega] be obtained. This resistance range limited the variance of electrode placement to approximately a 1-cm difference among subjects.

The resting condition served as a control for all subjects. This condition consisted of lying prone with the knees flexed at 15 to 20 degrees for 20 minutes. The two other conditions were the room-temperature gel pack (RTGP) condition and the cold gel pack (CGP) condition. These conditions were also 20 minutes in duration. The RTGP condition consisted of placing a room-temperature gel pack wrapped in a dry cotton pillowcase on the subject's right ankle. This was done because we believed that any externally applied weight could decrease local limb volume. For the CGP condition, a cold gel pack was taken from a refrigerated container, which was maintained at a temperature of 10[degrees] to 20[degrees]F. The cold gel pack was wrapped in a dry cotton pillowcase and placed on the subject's right ankle for the entire trial. Both the room-temperature and cold gel packs were of equal size and weight.

Data were obtained on each subject on two separate days. On the first day, each subject underwent the control condition followed by a randomly selected RTGP or CGP condition. On the second day, each subject again underwent the control condition followed by the remaining experimental condition. Data were collected between the hours of 4 and 8 PM.

The between-trial reliability of the plethysmographic measurements were assessed for the resting condition using an intraclass correlation coefficient (ICC[K,1]). The resulting ICC was found to be .971. Because of this high trial-to-trial reliability and the fact that no significant differences (P>.05) were found between the baseline values obtained on both test days, a mean of the two resting condition baseline values was used for comparison with measurements obtained for the experimental conditions.

For each condition, plethysmographic readings were taken at 3-minute intervals, at which venous return in the right leg was occluded by inflating a blood pressure cuff placed proximal to the knee for 30 seconds. Cuff pressure was maintained at each subject's resting diastolic blood pressure. Cuff inflation was accomplished using a foot pump connected to the cuff. A valve was closed to allow air into the cuff and sustain the pressure for 30 seconds and was then opened to release the air. The foot pump was used to decrease the amount of time it took for the cuff pressure to reach the subject's resting diastolic blood pressure. The subject's heart rate and skin temperature and the room temperature were also recorded at each 3-minute interval.

Upon completion of the session, the electrodes were removed and the subject was allowed to sit up. Blood pressure, heart rate, skin temperature, and room temperature were again measured. Once skin sensation returned to the subject's foot and vital signs were normal, the subject was allowed to leave.

Data Analysis

When the thigh cuff was inflated, venous outflow from the lower leg was occluded, producing an increase in volume in the lower leg distal to the cuff over the 30 seconds of cuff inflation. As venous outflow was occluded, a volume increase in the lower leg was produced for each of the six measurement points during all three conditions. The impedance plethysmograph measured this volume increase between the electrodes applied at the ankle. A typical volume increase following cuff inflation, as recorded by the plethysmograph, is shown in Figure 2. The plethysmograph recorded volume increase as an increase in percentage of impedance. There is a direct linear relationship between percentage of impedance change and percentage of blood volume change when an impedance plethysmograph uses a 100-kHz current frequency.13 A percentage of blood volume change measurement was therefore obtained at six intervals over each 20-minute condition for each subject. The means for percentage of blood volume change were compared over time for each of the three conditions using a two-way (condition, time) repeated-measures analysis of variance (ANOVA). An alpha level of .05 was used for all tests of statistical significance.

Results

A post hoc power analysis of the data from the 13 subjects was performed using the formula recommended by Keppel.(16) The results of this analysis indicated that our study had a power level of .95. As power is an indication of a statistical test's sensitivity with regard to rejection of the null hypothesis,16 we felt confident that the sample size of 13 subjects was adequate. If the power level had been low, a larger sample size would have been considered to increase our chances of detecting possible differences among the various experimental factors. Subject means and standard errors of the mean for percentage of blood volume change at each of the six time intervals for the three conditions are given in Table 1. As an example of data interpretation from this table, the 0.96 value appearing at the 3-minute time interval in the control condition column indicates that, by the end of the 30-second thigh cuff inflation, there was a mean blood volume increase of 0.96 at the ankle. The 0.46 value appearing at the 3-minute time interval in the CGP condition indicates that, by the end of the 30-second cuff inflation, there was a mean blood volume increase of 0.46% at the ankle. Comparison of mean percentage of blood volume increases over time for all three conditions is shown in Figure 3.

The results of the ANOVA showed statistical significance among the three conditions P<.05). A Tukey's Honestly Significant Difference post hoc comparison showed significant differences (P<.05) between the resting and CGP conditions. Although blood volume increases measured during the RTGP condition were less than those measured during the control condition, mean differences between the control and RTGP conditions and the RTGP and CGP conditions were not significant (Tab. 2). In addition, no significant difference was found between the six time intervals and no significant interaction was found between the three conditions and the six time intervals.

The results of a second two-way (sex, condition) repeated-measures ANOVA were not significant (P>.05) for blood volume changes between male and female subjects for any of the three conditions. Mean heart rate and blood pressure data are presented in Table 3. The results of a third two-way (heart rate, condition) ANOVA were not significant (P>.05) for heart rate, condition, and heart rate x condition interaction. A t test comparing mean pretreatment and posttreatment blood pressures for each of the three conditions was also performed and revealed no statistical significance.

After 20 minutes of cold gel pack application, mean subject skin temperature measured at the calcaneus was 8.5[degrees]C. Mean reduction in skin temperature was 17.7[degrees]C. These two values are comparable to skin temperature values obtained during cold modality application reported elsewhere in the literature.(8)

Discussion

The study produced two important findings. First, application of a cold gel pack to the ankle for a therapeutically accepted time period of 20 minutes and at a temperature of 10[degrees] to 20[degrees]F produced an immediate and sustained reduction in local blood volume increase during venous occlusion. The data in Table 1 indicate that absolute blood volume increases were small over the 30 seconds of venous occlusion. The blood volume increase during cold gel pack application, however, was 60% less than the blood volume increase during the resting condition. Because the room-temperature gel pack decreased local blood volume compared with the resting level, it appears that local blood volume reduction noted during cold gel pack application may be attributed to the combined effect of the temperature of the pack and the pressure from the weight of the pack. As local tissue was not traumatized and no significant change in subject heart rate or blood pressure was observed, it may be concluded that the reduced rise in blood volume during cold gel pack application is directly attributable to constriction of the local vasculature.

The second important finding was that reactive vasodilation was not observed during the 20-minute period of cold gel pack application. This finding supports the earlier work by Knight and colleagues.(5,6) It is interesting to note that, at 13.5 minutes of cold gel pack application, mean blood volume increase reached its lowest point Fig. 3). At 17 minutes, a slightly greater mean blood volume increase occurred. The mean volume increase then fell again at 20 minutes. Although these changes in blood volume could represent reactive vasodilation, this pattern was also observed beginning 13.5 minutes after the application of the weight at room temperature. The cold stimulus, therefore, would not appear to be the cause. The rise-and-fall pattern of blood volume increase beginning at 13.5 minutes with both the weight and cold pack application is not easily explained. Further study with cold pack and weight applications for longer than a 20-minute time period is indicated. The 10*F temperature variation of the cold pack container is of note when drawing conclusions from this study. The variation may be explained by the fact that the cold pack container was also being used clinically during this study. The cold gel pack used in the study, however, was set aside within the container and not used for patient care. The variation in container temperature could have been more closely controlled; however, we allowed the variation to occur in order to more closely approximate clinical conditions.

The question of the effect of any condensation that might have formed on the skin during cold gel pack application must be considered when drawing conclusions from the results of our study. Preliminary pilot work in our laboratory showed no change in impedance when the electrodes were sprayed with tap water. In addition, if condensation had occurred, the plethysmograph would have recorded a reduction in impedance. Instead, when the cold gel pack was applied, impedance increased. We therefore do not believe that condensation, if present, would have had a significant impact on impedance.

Summary

A cold gel pack applied to the ankle for 20 minutes at a temperature of 10 to 20[degrees]F significantly reduced local blood volume increase during venous occlusion in nontraumatized subjects and sustained this reduction. This impact on blood volume appears to be caused by the combined effects of pressure from the weight of the cold pack and the cold stimulus. Reactive vasodilation was not observed within the time and temperature parameters used in this study. The consistent and sustained reduction in local blood volume increase, combined with no apparent reactive vasodilation, suggests the value of cold gel pack application for control of local circulation immediately following soft tissue trauma. Further study of cold gel pack application to human subjects with acute local soft tissue trauma is indicated.

References

1 Olson JE, Stravino VD. A review of cryotherapy. Phys Ther. 1972;52:840-853.

2 Lewis T. Observations upon the reactions of the vessels of the human skin to cold. Heart. 1930; 15:177-208.

3 Clarke R, Hellon R. Vascular reactions of the human forearm to cold. Clin Sci. 1958;17: 165-179.

4 Fox R, Wyatt H. Cold induced vasodilation in various areas of the body surface in man. J Physiol (Lond). 1962; 162:289-297.

5 Knight KL, Londeree B. Comparison of blood flow in the ankle of uninjured subjects during therapeutic application of heat, cold, and exercise. Med Sci Sports Exerc. 1980; 12: 76-80.

6 Knight KL, Aquino J, Johannes SM, Urban CT. A reexamination of Lewis' cold induced vasodilation: in the finger and the ankle. Athletic Training. 1980;15:248-250.

7 Kowal MA. Review of physiological effects of cryotherapy. Journal of Orthopaedic and Sports Physical Therapy. 1983;5:66-73.

8 Meeusen R, Leivens P. The use of cryotherapy in sports injuries. Sports Med. 1986;3: 398-414.

9 Whitney R. The measurement of volume changes in human limbs. J Physiol (Lond) 1953;121:1-27.

10 Dahn I, Hallbook T. Simultaneous blood flow measurements by water and strain gauge plethysmography. Scand J Clin Lab Invest. 1970;25:419-428.

11 Needham T. The measurement of blood flow: strain gauge plethysmography. Biomedical Engineering. july 1972:266-269.

12 Sakaguchi S, Ishitobi K, Kameda T. Functional segmental plethysmography with mercury strain gauge. Angiology. 1972;23:127-135.

13 Anderson F. Impedance plethysmography in the diagnosis of arterial and venous disease. Ann Biomed Eng. 1984;12:79-102.

14 Golden JC, Miles DS. Assessment of peripheral hemodynamics using impedance plethysmography. Phys Ther. 1986;66:1544-1547.

15 Nyboer J. Electrical Impedance Plethysmography. 2nd ed. Springfield, Ill: Charles C Thomas, Publisher; 1970.

16 Keppel G. Design and Analysis: A Researcher's Handbook. Englewood Cliffs, NJ: Prentice-Hall Inc; 1973.

17 Wheeler HB, Penney BC. impedance plethysmography: theoretic, experimental, and clinical considerations. In: Bernstein E, ed. Noninvasive Diagnostic Techniques in Vascular Disease. St Louis, Mo: CV Mosby Co; 1985: chap 14.