Localized cold or cryotherapy is frequently applied after acute soft tissue injury in an attempt to minimize the inflammatory response, local edema, swelling, and pain and therefore enhance recovery from soft tissue injury (2). Because inflammation is also integral in the development of exercise-induced muscle damage (27), whole-limb cooling via cold-water immersion is now increasingly applied after exercise to alleviate some of the physiologic and functional deficits associated with exercise-induced muscle damage (2,31).
The physiological effects of cryotherapy are thought to be partly underpinned by reductions in microvascular blood flow to the injured muscle (22,28), which subsequently reduces edema and the induction of inflammatory events (3,8). Whole-limb cold-water immersion is therefore likely to be effective by virtue of its effect on deep muscle blood flow (13,22). We have recently assessed skin and femoral artery blood flow using laser Doppler flowmetry and high-resolution duplex ultrasound, respectively, to provide an indirect estimate of muscle blood flow in the lower limbs in response to cold (8°C) and cool (22°C) water immersion at rest (13). Immersion in both cold and cool water promoted similar reductions in femoral artery blood flow but increased blood flow to the skin in the cold (8°C) water. This suggests that colder water temperatures may induce greater reductions in muscle blood flow at rest and may therefore be more effective in the treatment of exercise-induced muscle damage and injury rehabilitation.
It is well documented that the vascular response to sympathetic stimulation is blunted during exercise and whole-body heat stress compared with rest (11,30,35). Reduced vasoconstrictor responsiveness in the skin when cold (8°C) water immersion is applied after exercise compared with rest may therefore lead to similar changes in skin blood flow to those associated with cool (22°C) water. Such information has important implications for treatments guidelines because the application of cold-water immersion frequently occurs immediately after exercise, when core body and local limb temperatures are elevated. To date, only two studies have investigated the blood flow response to postexercise cold-water immersion (17,32). However, no attempt has been made to compare the effects of different degrees of cooling on these responses. Previous observations are either limited by their use of venous occlusion plethysmography (32), which is highly sensitive to motion artifact and is limited to the assessment of whole-limb blood flow (26). Alternatively, observations (17) are limited by the use of near infrared spectroscopy (NIRS), which provides indirect estimates of relative changes in blood volume within the microcirculation. In addition, the NIRS signal is confounded under conditions in which marked changes in skin blood flow arise (e.g., exercise, heating, and cooling) (6).
The purpose of the present study was to examine the effects of cold (8°C) and cool (22°C) water immersion on cutaneous and femoral artery blood flow responses after exercise using laser Doppler flowmetry and high-resolution duplex ultrasound. We hypothesized that whole-limb immersion in 8°C and 22°C would decrease femoral artery and skin blood flow to a similar extent after exercise, but colder water would reduce muscle temperature to a greater extent.
MATERIALS AND METHODS
Twelve recreationally active men were studied (mean ± SD: age, 25.5 ± 4.7 yr; height, 1.8 ± 0.1 m; mass, 78.4 ± 7.7 kg; V˙O2peak, 47 ± 8 mL·kg−1·min−1). Participants were familiarized with the experimental procedure and associated risks and gave their written informed consent to participate. The study conformed to the Declaration of Helsinki and was approved by the Institutional Ethics Committee. It was estimated that 12 participants would enable the detection of a 25% reduction in femoral artery blood flow after 10 min of cold-water immersion, assuming a test–retest coefficient of variation of 20% (see below) and a statistical power of 80%.
Before any experimental trials, each participant completed a maximal incremental cycling protocol on a cycle ergometer (Daum Ergo Bike, Premium 8i, Germany) while simultaneous breath-by-breath (V˙O2) measurements were recorded (Oxycon Pro, Jaeger, Germany). The cycling protocol commenced at 75 W and was increased 25 W every 2 min until volitional exhaustion was reached. Peak power output was derived as the highest power output attained at this point. Peak oxygen uptake (V˙O2peak) (mL·kg−1·min−1) was recorded as the highest 30-s average recorded before volitional exhaustion.
On the day of the experimental trials, participants arrived at the laboratory at least 3 h postprandial having refrained from exercise, alcohol, tobacco, and caffeine during the previous 24 h. All participants recorded nutritional and fluid intake before the first exercise trial. This record was photocopied and returned to permit them to repeat their preparation for the remaining trials. They also consumed 5 mL·kg−1 of water 2 h before arriving at the laboratory. All trials were conducted under an ambient temperature of 22°C–24°C to control variability in cutaneous blood flow (5) and at the same time of day to avoid the circadian variation in internal body temperature.
Each participant was required to complete a submaximal cycle ergometer protocol, followed by a 10-min period of immersion in 8°C and 22°C water and seated rest (control). The water temperature and immersion protocol was based on data frequently reported in the literature (34). The different temperature trials were conducted in a counterbalanced order, 1 wk apart. On arrival at the laboratory, the participant’s nude body mass (kg) was obtained using digital scales (Seca, Hamburg, Germany). A rectal probe was self-inserted, and a heart rate monitor was positioned across the chest. Participants were then laid in a supine position for 30 min on a bed (next to the water tank) for the attachment of instrumentation and to stabilize physiological status, wearing training shorts and a tracksuit top. After baseline measurements, participants cycled at 70% V˙O2peak until a core temperature of 38°C was attained. Participants then returned to a supine position for 10 min to enable preimmersion measurements to be taken. Participants were subsequently raised from the bed in a semirecline position using an electronic hoist (Bianca, Arjo Ltd., Gloucester, UK) and either lowered into the water tank (ECB, Gloucester, UK) until the thighs were fully submerged for a duration of 10 min or remained suspended above the bed (control). At the end of immersion, participants were returned to the bed using the electronic hoist and remained in a supine position for a period of 30 min in the laboratory under an ambient temperature of 22°C–24°C. The use of the hoist to raise and lower the participants was important to avoid the effect of muscle activation on blood flow measures. Rectal temperature, thigh and calf skin temperature, heart rate, and thigh and calf cutaneous blood flow were continuously monitored at baseline, before immersion, throughout water immersion, and during the 30-min postimmersion period. Muscle temperature was recorded at baseline, immediately before and after immersion, and 30 min after immersion. Superficial femoral artery blood flow and mean arterial blood pressure was measured at baseline, before immersion, immediately after immersion, and continuously during the 30-min postimmersion period. Perceived thermal comfort (40) and shivering (33) were also recorded at the same time points and during immersion. All pre- and postimmersion measurements were made in a supine position. A schematic illustration of the experimental design is shown in Figure 1.
Rectal, thigh, skin, and muscle temperature
A rectal probe (A/S Krondalve 9 DK-2610; ELAB, Rodovre, Denmark) was inserted 15 cm beyond the external anal sphincter for the assessment of rectal temperature. A skin thermistor was attached to the upper thigh for the assessment of skin temperature. Muscle temperature was assessed using a needle thermistor inserted into the vastus lateralis (13050; ELAB). Thigh skinfold thickness was measured using Harpenden skinfold calipers (HSK BI; Baty International, West Sussex, UK) and divided by 2 to determine the thickness of the thigh subcutaneous fat layer over each participant’s vastus lateralis (9). The needle thermistor was then placed at a depth of 3 cm plus one-half the skinfold measurement for the determination of deep muscle temperature (3 cm). The thermistor was then withdrawn at 1-cm increments for the determination of muscle temperature at 2 cm and 1 cm below the subcutaneous layer. Rectal temperature, skin temperature, and muscle temperature were recorded using an electronic measuring system (CTF 9004; ELAB).
Heart rate and arterial blood pressure
Heart rate was continuously measured during the experimental protocol using a three-lead wire electrocardiogram connected to a data software acquisition system (PL3516 PowerLab 16/35; AD Instruments Ltd., Oxford, UK). Arterial blood pressure was measured noninvasively via automated brachial auscultation (Dinamap; GE Pro 300V2, Tampa, FL), and mean arterial blood pressure was calculated as [diastolic + (0.333 × (systolic − diastolic))].
Femoral artery blood flow
A 10-MHz multifrequency linear array transducer attached to a high-resolution ultrasound machine (Acuson P50; Siemens, Munich, Germany) was used to measure femoral artery diameter and velocity. The images were taken at the superficial femoral artery in the proximal third of the left leg approximately 3 cm distal to the bifurcation. This position was marked on the skin for ultrasound head repositioning during the remaining measures. Ultrasound parameters were set to optimize longitudinal B-mode images of the lumen/arterial wall interface. Continuous and synchronized pulsed wave Doppler velocities were also obtained using the ultrasound machine. Data were collected using an insonation angle of 60°, and each measurement was recorded for 2 min. Analysis of diameter was performed using custom designed edge-detection and wall-tracking software as described previously (38), which provides simultaneous and continuous measurements of arterial diameter and blood flow velocity. The assessment of blood flow velocity uses the edge detection algorithm to assess the peak velocity envelope from the Doppler gate, which is placed in the middle of the artery. From these data, the software calculates blood flow (the product of cross-sectional area and blood flow velocity) at 30 Hz. In this experiment, femoral artery blood flow was shown to have a coefficient of variation of 20%. The edge-detection and wall-tracking software is semiautomated and provides diameter measurements, which are considerably more repeatable (coefficient of variation = 6.7%) than manual methods and are associated with less observer error (12,38). This method of blood flow assessment is closely correlated with actual flow through a phantom arterial flow system (38). All data were written to file and retrieved for analysis in the custom-designed analysis package. Resting diameter, blood velocity, and blood flow were calculated as the mean of the data collected for a 20-s period of each 2-min scan for statistical analysis. Using the mean blood flow and the mean arterial pressure data, femoral vascular conductance was calculated as the ratio of blood flow/mean arterial pressure.
Cutaneous blood flow
Red blood cell flux was used as an index of skin blood flow via laser Doppler flowmetry (Periflux System 5001; Perimed Instruments, Jarfalla, Sweden). A laser Doppler probe (PROBE 455; Perimed, Suffolk, UK) was attached to the midanterior thigh, midline, halfway between the inguinal line and the patella, and on the calf, left of the midline, in the region of the largest circumference. Once affixed, the probes were not removed until the completion of each trial. The laser Doppler flowmetry data were converted from perfusion units to cutaneous vascular conductance units (AU) by the ratio of laser Doppler flux to mean arterial blood pressure (cutaneous vascular conductance = laser Doppler flux/mean arterial blood pressure × 100) and expressed as a percentage change from preimmersion values. When expressed as a percentage change from baseline to maximum, cutaneous blood flow has a coefficient variation of 4% in our laboratory with a coefficient of variation of 10% observed for resting cutaneous blood flow.
All data are presented as mean ± SD. A two-factor (condition × time) within-participants general linear model was used to evaluate treatment differences between the 8°C, the 22°C, and the control conditions. A three-way ANOVA (condition × depth × time) was used to analyze muscle temperature (Statistical Package for the Social Sciences version 18.0; SPSS Inc., Chicago, IL). The assumption of sphericity (homogeneity of variance) was assessed and corrected for using the Greenhouse–Geisser [Latin Small Letter Open E] (1). The main effect of condition was followed up using multiple comparisons, as there were three levels for condition. A significant effect of time was followed up with planned multiple contrasts in line with the a priori hypotheses. Consequently, data at the specific time points were compared with the baseline (first) time point using Newman–Keuls multiple contrasts (Statistica, version 10; StatSoft Ltd., Milton Keynes, UK). As thigh and calf skin conductance was expressed as percentage change from preimmersion (zero), a one-sample t-test was used in follow-up analyzes. Therefore, at each measured time point, the main effect of time for thigh and calf skin conductance was compared with the preimmersion (first) time point. Where a significant interaction between condition and time was observed, differences between conditions were examined at each time point using Newman–Keuls multiple contrasts. A one-way repeated-measures ANOVA was used to determine any differences between conditions at baseline for femoral artery conductance and at preimmersion for thigh and calf cutaneous vascular conductance. The α level for evaluation of statistical significance was set at P < 0.05.
Thermoregulatory responses during immersion and after immersion
Rectal temperature was similar between conditions at baseline, before immersion and throughout the immersion, and postimmersion period (P = 0.38). Rectal temperature was reduced over time (P < 0.01) and to the greatest extent 30 min after immersion (8°C, 0.7°C ± 0.3°C; 22°C, 0.6°C ± 0.2°C; control, 0.7°C ± 0.2°C; P = 0.21) but still remained above baseline (P < 0.001; Fig. 2).
Thigh skin temperature was similar at baseline and before immersion (P > 0.05) but was lower throughout the postimmersion period in the 8°C and 22°C conditions compared with the control condition (P < 0.001) (Fig. 3). The colder water temperature also reduced thigh skin temperature to a greater extent compared with 22°C immediately after immersion (P < 0.001). The rate of decrease in skin temperature was different between all conditions (P < 0.001), with the largest difference occurring immediately after immersion. Thigh skin temperature increased during the 30-min recovery period in both cooling conditions, whereas values remained relatively stable in the control condition. Thigh skin temperature remained below baseline at the end of the recovery period in the 8°C and 22°C conditions (P < 0.001) and slightly above baseline in the control condition (P = 0.03).
Thermal comfort decreased after cooling (8°C, 2 ± 2.0 rating; 22°C, 4 ± 2.0 rating; control, 6 ± 1.0 rating) relative to baseline (8°C, 5 ± 1.0 rating; 22°C, 5 ± 1.0 rating; control, 5 ± 1.0 rating). Thermal comfort returned to baseline values at the end of the 30-min recovery period. Shivering during immersion in the 8°C condition was slight to moderate, compared with a no or slight shivering during the 22°C condition. A similar shivering response in each cooling condition was observed during the initial 10 min of the recovery period, after which no shivering occurred. No shivering was observed throughout the control condition.
At baseline, muscle temperature was similar between conditions at each depth (P > 0.05). Muscle temperature was also similar before immersion at a depth of 3 cm (8°C, 37.9°C ± 0.3°C; 22°C, 38.0°C ± 0.3°C; control, 38.0°C ± 0.4°C; P > 0.05), 2 cm (8°C, 37.7°C ± 0.3°C; 22°C, 37.7°C ± 0.3°C; control, 37.6°C ± 0.6°C; P > 0.05), and 1 cm (8°C, 37.1°C ± 0.5°C; 22°C, 37.2°C ± 0.4°C; control, 37.1°C ± 0.7°C; P > 0.05). Muscle temperature was reduced over time (P < 0.01) (Fig. 4), generally decreasing immediately after immersion (P < 0.01) and continuing to decrease up to 30 min after immersion (P < 0.01). These reductions depended on probe depth (P < 0.01). At the deeper depths, the greatest declines in muscle temperature occurred 30 min after immersion (P < 0.01). At the 1-cm probe depth, the greatest decline in muscle temperature was observed immediately after immersion in the cooling conditions, before gradually increasing toward baseline values at the end of the recovery period (Fig. 4).
The reductions in muscle temperature over time were also dependent on condition (P < 0.001). Muscle temperature was generally lower in the cooling conditions both immediately (8°C, 34.1°C ± 1.0°C; 22°C, 35.6°C ± 0.8°C; control, 36.8°C ± 0.5°C; P < 0.001) and 30 min after immersion (8°C, 34.3°C ± 0.8°C; 22°C, 35.2°C ± 0.7°C; control, 36.5°C ± 0.4°C; P < 0.001). At both time points, there was also a difference between cooling conditions (P < 0.001), with a greater decrease in muscle temperature observed in colder water (P < 0.001).
The differences in muscle temperature between condition and time points were in turn dependent on the probe depth (P < 0.001). Compared with the control condition, both cooling conditions induced marked reductions in muscle temperature immediately after immersion at 1- and 2-cm probe depths (P < 0.001). Muscle temperature was also reduced at a 3-cm probe depth in the 8°C condition (P < 0.001), but not 22°C (P = 0.16) (Fig. 4). At 30 min after immersion, muscle temperature was also lower in both cooling conditions compared with the control condition across all probe depths (P < 0.01). Muscle temperature was also reduced in 8°C cooling compared with 22°C immediately and 30 min after immersion at 1-cm (P < 0.001) and 2-cm (P < 0.001) probe depths. Muscle temperature was similar at the 3-cm probe depth immediately after immersion (P = 0.16); however, it showed a tendency to be lower in 8°C at 30 min after immersion (P = 0.06) (Fig. 4).
Heart rate and mean arterial pressure during immersion and after immersion
Mean heart rate across the immersion and postimmersion period was similar between conditions (P = 0.17); however, the change in heart rate was different (P < 0.001). Heart rate was similar at baseline (P > 0.05) and immediately before immersion (8°C, 88 ± 7 beats per minute; 22°C, 86 ± 6 beats per minute; control, 87 ± 9 beats per minute; P > 0.05) but increased slightly during the 10-min immersion in 8°C water (91 ± 5 beats per minute) compared with 22°C (78 ± 8 beats per minute; P < 0.001) and control (80 ± 6 beats per minute; P < 0.001). Heart rate remained similar between all conditions until the end of the postimmersion recovery period (P > 0.05) and remained above baseline at the end of the 30-min recovery period in each condition (P < 0.001).
Mean arterial pressure was generally higher in the 8°C condition (83 ± 5 mm Hg) compared with the 22°C (79 ± 6 mm Hg) and control (79 ± 6 mm Hg) conditions (P < 0.01). The change in mean arterial pressure over time was also different between conditions (P = 0.01). Mean arterial pressure was similar at baseline and immediately before immersion (P > 0.05); however, a higher mean arterial pressure was observed during the 10-min immersion and the initial 10-min postimmersion period in 8°C water (86 ± 10 mm Hg) compared with 22°C (79 ± 7 mm Hg) and control (78 ± 7 mm Hg) conditions (P < 0.05). Mean arterial pressure was similar between all conditions throughout the remaining period of the postimmersion phase (P > 0.05) and remained above baseline at the end of the 30-min recovery period in each condition (P = 0.02).
Femoral artery blood flow responses during immersion and after immersion
Femoral artery blood flow (P = 0.66) and femoral artery conductance (P = 0.37) were similar between cooling conditions but lower compared with the control condition (P < 0.01). The rate of decrease in femoral artery blood flow (P = 0.02) and conductance (P = 0.04) was greater in both cooling conditions in contrast to the control condition. Arterial blood flow and conductance was similar at baseline and immediately before immersion (P > 0.05); however, a lower femoral artery blood flow and conductance (∼40%) was observed from 10 min after immersion until the end of the 30-min recovery period (P < 0.05; Fig. 5). At the end of the 30-min recovery period, femoral artery blood flow and femoral artery conductance in the cooling conditions (∼55%) were below the control values (P < 0.01). Femoral artery blood flow and conductance were reduced in the cooling conditions by ∼30% and ∼75% relative to baseline and preimmersion values, respectively.
Cutaneous blood flow responses during immersion and after immersion
Preimmersion thigh (8°C, 50.2 ± 28.8 AU; 22°C, 46.2 ± 17.9 AU; control, 51.4 ± 21.0 AU; P = 0.74) and calf (8°C, 30.2 ± 9.2 AU; 22°C, 32.0 ± 14.6 AU; control, 28.8 ± 13.8 AU; P = 0.64) cutaneous vascular conductance was similar between conditions. When the data were expressed as a percentage change before immersion, there was greater skin vasoconstriction observed in both cooling conditions at the thigh (P < 0.01) and calf (P < 0.05) relative to the control condition during immersion and throughout the postimmersion recovery period (Fig. 6). No difference was observed between cooling conditions (P > 0.05). Significant vasoconstriction relative to preimmersion was observed throughout the postimmersion period under both cooling conditions (∼60%–70%).
The major finding of this study is that 10 min of immersion in both 8°C and 22°C water after moderately intense endurance exercise reduces femoral artery blood flow (∼55%) and cutaneous blood flow compared with postexercise rest. Nevertheless, despite greater reductions in muscle and thigh skin temperatures in colder water, no differences were observed between cooling conditions in terms of the magnitude of effect on femoral artery and cutaneous blood flow. Collectively, these findings suggest that the application of 8°C and 22°C water after exercise will influence any potential changes in muscle blood flow to a similar extent compared with postexercise rest. Furthermore, any additional treatment benefits arising from the use of colder temperatures are likely to be mediated through the effects of reduced tissue temperature rather than any further reductions in muscle blood flow. These findings provide important insights into the mechanisms, which underpin the use of cold-water immersion after exercise as well as a basis for improving treatment outcomes in clinical and sporting environments.
This is the first study, to our knowledge, that has addressed the influence of different degrees of cooling via whole-limb cold-water immersion on changes in limb artery and cutaneous blood flows after exercise. Previous work has evaluated the influence of cold-water immersion on whole-limb blood flow after exercise using venous occlusion plethysmography (32). Although plethysmography can be reliably used to measure relative changes in whole-limb blood flow at rest (4), it does not distinguish between limb artery and skin blood flows and cannot be used during water immersion. Attempts to determine the effects of more localized cooling strategies such as ice packs on postexercise muscle blood flow have previously been undertaken using radioactive tracers (e.g., 133Xe) (29). However, the estimation of muscle blood flow using radioactive tracers is associated with several recognized limitations (26) and ice packs are not ideal for cooling large muscle groups. Recent work (17) has examined the effects of cold-water immersion on relative changes in postexercise muscle perfusion using NIRS. However, NIRS only provides indirect estimates of relative changes in blood volume within the microcirculation. Furthermore, the NIRS signal is confounded under conditions were marked changes in skin blood flow arise (e.g., exercise, heating, and cooling) (6). In the present study, we continuously measured changes in lower limb cutaneous blood flow using laser Doppler flowmetry while measuring femoral artery blood flow via conduit artery high-resolution duplex ultrasound (13). The latter has the advantage of providing absolute blood flow measures, and it has an improved temporal resolution compared with plethysmography (26). Femoral artery blood flow represents the sum of flow to muscle, skin, subcutaneous tissue, and bone, with most flow deemed to perfuse muscle and skin (7). Absolute perfusion of the entire cutaneous vascular bed cannot be assessed in humans (5); however, laser Doppler flowmetry can accurately assess relative changes in cutaneous blood flow (5).
In the present investigation, 10 min of lower limb immersion in either 8°C or 22°C water reduced femoral artery conductance by ∼30% and ∼75% relative to preexercise baseline and preimmersion values, respectively. This led to a ∼55% reduction in femoral artery conductance compared with the control condition at the end of the 30 min postexercise recovery period. The magnitude of the reduction in femoral artery conductance with cooling relative to the control condition confirms observations in previous studies, which have used isotope clearance (29) and venous occlusion plethysmography (32) measurement techniques to assess the effects of localized cooling (ice packs) (29) and whole-limb cooling (cold-water immersion) (32) on limb blood flow after exercise. In our study, it is likely that the activation of thermoreceptors in response to 8°C cooling led to a reflex increase in sympathetic nerve activity and the reduction in femoral artery blood flow (20,21). Indeed, this reduction in blood flow and the accompanying increases in heart rate and mean arterial blood pressure in colder water are consistent with the cardiovascular cold pressor response (23,25,39). Despite a similar reduction in femoral artery blood flow in 22°C water, no marked changes in heart rate or mean arterial blood pressure were observed. We have previously reported similar findings (13) at comparable water temperatures, with reductions in blood flow in temperate water attributed to the activation of nonnoxious thermoreceptors (21,36) known to be operable within the skin temperatures seen in this study (15).
In line with the changes in femoral artery blood flow, we also currently observed a similar reduction in cutaneous blood flow under both cooling conditions compared with the control condition. Taken together, the similar change in femoral artery and skin blood flow suggest that both cooling strategies will be equally effective in promoting any potential reductions in muscle blood flow that may occur when applied immediately after exercise. The present changes in cutaneous blood flow contrast with previous work under resting conditions, where cold-induced vasodilation led to higher cutaneous blood flows in 8°C water compared with 22°C (13). Under resting conditions, increases in cutaneous blood flow during colder water immersion may redistribute blood from the underlying muscle; consequently, colder water may be more effective in reducing muscle blood flow and inflammation at rest by virtue of its associated increase in cutaneous blood flow (13). Differences in the cutaneous responses to marked cooling (8°C) after exercise compared with rest may reflect the exercise-induced increase in body and local limb temperatures, which preceded immersion. In nonacral (hairy) skin, the precise mechanisms mediating cold-induced vasodilation have yet to be elucidated; however, marked local cooling may directly inhibit the normal vasoconstrictor response leading to vasodilation (10). Increases in skin temperature and/or core temperature associated with heat stress attenuate cutaneous vasoconstrictor responsiveness (35). Consequently, the increased body temperature (and attendant skin blood flow) before immersion in the present investigation may have reduced the degree of vasoconstriction and prevented skin blood flow reaching low levels and the associated onset of cold-induced vasodilation, leading to sustained vasoconstriction in both 8°C and 22°C conditions. This sympatholytic effect may also account for the similar cutaneous blood flow observed in the two cooling conditions after immersion despite a lower skin temperature in the 8°C condition. Indeed, a local cooling stimulus of similar magnitude to the difference in skin temperature between conditions (∼6°C) during the initial 10 min after immersion has been shown to induce greater reductions in skin blood flow when occurring at rest compared with those presently observed (16).
In previous work, the effects of different degrees of cooling on both femoral and cutaneous blood flows within the limb were used to provide valid, qualitative estimates of the effects of cooling on muscle perfusion at rest (13). Under these conditions, differences in the direction of change in blood flow in the skin and femoral artery enabled inference with respect to muscle blood flow perfusion. In contrast, in the present investigation, both femoral artery and cutaneous blood flow declined with cooling. Because skin blood flow increased before immersion, decreases in femoral artery blood flow with cooling may reflect decreasing skin blood flow, as opposed to any alterations in muscle perfusion per se and the interpretation of these distinct compartmental changes becomes more problematic. The qualitative nature of cutaneous blood flow (5) therefore does not enable direct determination of the true extent to which changes in cutaneous blood flow may have influenced muscle perfusion. Future work using methods such as positron emission tomography (14) to apportion changes in specific domains may permit quantification of postexercise muscle perfusion in response to different degrees of cooling.
The physiological effects of cryotherapy are thought to be mediated through temperature-induced reductions in microvascular blood flow around the injured area, which in turn reduces edema and the induction of inflammatory events (22,28). Whole-limb cold-water immersion may therefore be an effective treatment modality by virtue of its effect on muscle temperature. Our findings indicate that both 8°C and 22°C water decreased deep and superficial muscle temperature relative to the control condition, with greater reductions in colder water. Furthermore, greater reductions in superficial temperature were initially observed after immersion with deep muscle temperature declining to a greater extent 30 min after immersion. The observation of a sustained reduction in deep tissue temperature and blood flow to the deeper muscle tissue confirm previous reports at rest (13) and provide support for the application of cold-water immersion soon after cessation of the activity that has led to exercise-induced muscle damage and injury. Similarly, the transition from superficial to deep tissue cooling confirms findings from previous studies undertaken at rest, which have used localized cooling (9) and whole-limb immersion cooling (13,18). The increase in skin and superficial (1-cm depth) muscle temperature during the 30 min after water immersion partly reflects removal of the cooling source per se and subsequent exposure of the limbs to room air temperature. In addition, the hemodynamic exchange between the cooler surface and the warmer deeper tissue is likely to have promoted a net flow of heat leading to a gradual increase in superficial tissue temperature and a corresponding decrease in deep tissue temperature over time (9).
Recent work demonstrated that marked increases (∼4°C) in deep (∼2 cm) muscle temperature associated with local heating induced increases in skeletal muscle blood flow (14). As noted previously, femoral artery blood flow in the present investigation was reduced with cooling relative to the control condition; however, similar reductions in blood flow were observed in the cooling conditions despite differences in muscle temperature. In contrast to skin, where an elevated body temperature impairs vasoconstrictor responsiveness to cooling, such effects do not arise in the whole limb (19) and therefore do not explain the failure to observe greater reductions in femoral artery blood flow with greater decreases in muscle temperature in 8°C water relative to 22°C. We have previously reported similar reductions in femoral artery blood flow in response to 8°C and 22°C cooling at rest (13). It is therefore possible that the magnitude of the difference in deep muscle temperature observed between cooling conditions (∼0.5°C) in the present investigation may not have been sufficient to directly modify femoral artery blood flow.
Our findings have shown that colder water temperatures cause greater reductions in muscle tissue temperature, supporting the view that better treatment outcomes after muscle injury may arise from the selection of cooling modalities, which promote greater tissue cooling (24). Similar to cold-water immersion under resting conditions, a decreased pain tolerance (13,37) was observed in several participants during the 8°C noxious cooling, especially in the extremities. Although muscle temperature was not reduced to the same extent as with 8°C cooling, deep muscle temperature was still significantly reduced after 22°C water immersion compared with the control condition. Less noxious cooling above temperatures that cause cold pain (>18°C) (37) may therefore provide a suitable alternative to individuals unable to tolerate more noxious degrees of cooling (13). However, it remains to be ascertained whether colder water temperatures provide any added benefit to treatment outcomes than less noxious cooling in athletic settings and should provide a focus for future work.
In summary, we assessed postexercise femoral artery and cutaneous blood flows during and postwater immersion at 8°C or 22°C compared with rest. The present findings suggest that being immersed in cool and cold-water temperatures causes similar reductions in blood flow. However, greater reductions in muscle temperature arise with colder water temperatures, suggesting that such modalities may be more effective in the treatment of exercise-induced muscle damage and injury. Our findings provide important insights into the possible mechanisms responsible for cold-water immersion in alleviating inflammation in sport and athletic contexts.
The authors express their gratitude to all who participated in this study. They also thank ECB Cold Spa for providing the water tanks and UK Sport for funding the present investigation.
Warren Gregson has received funding from ECB Cold Spas Ltd. for the cold-water immersion facility and from UK Sport for part funding of a Ph.D. degree program. Chris Mawhinney, Helen Jones, Chang Hwa Joo, David A. Low, and Daniel J. Green have no conflicts of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Atkinson G, Nevill AM. Selected issues in the design and analysis of sport performance research. J Sports Sci
. 2001; 19 (10): 811–27.
2. Bailey DM, Erith SJ, Griffin PJ, et al. Influence of cold-water immersion on indices of muscle damage following prolonged intermittent shuttle running. J Sports Sci
. 2007; 25 (11): 1163–70.
3. Bleakley C, McDonough S, MacAuley D. The use of ice in the treatment of acute soft-tissue injury: a systematic review of randomized controlled trials. Am J Sports Med
. 2004; 32 (1): 251–61.
4. Casey DP, Curry TB, Joyner MJ. Measuring muscle blood flow: a key link between systemic and regional metabolism. Curr Opin Nutr Metab Care
. 2008; 11 (5): 580–6.
5. Cracowski JL, Minson CT, Salvat-Melis M, Halliwill JR. Methodological issues in the assessment of skin microvascular endothelial function in humans. Trends Pharmacol Sci
. 2006; 27 (9): 503–8.
6. Davis SL, Fadel PJ, Cui J, Thomas GD, Crandall CG. Skin blood flow influences near-infrared spectroscopy-derived measurements of tissue oxygenation during heat stress. J Appl Physiol
. 2006; 100 (1): 221–4.
7. Dinenno FA, Jones PP, Seals DR, Tanaka H. Limb blood flow and vascular conductance are reduced with age in healthy humans: relation to elevations in sympathetic nerve activity and declines in oxygen demand. Circulation
. 1999; 100 (2): 164–70.
8. Dolan MG, Thornton RM, Fish DR, Mendel FC. Effects of cold water immersion on edema formation after blunt injury to the hind limb of rats. J Athl Train
. 1997; 32 (3): 233–7.
9. Enwemeka CS, Allen C, Avila P, Bina J, Konrade J, Munns S. Soft tissue thermodynamics before, during, and after cold pack therapy. Med Sci Sports Exerc
. 2002; 34 (1): 45–50.
10. Faber JE. Effect of local tissue cooling on microvascular smooth muscle and postjunctional a2-adrenoceptors. Am J Physiol Heart Circ Physiol
. 1998; 255 (1 Pt 2): H121–30.
11. Flouris AD, Westwood DA, Mekjavic IB, Cheung SS. Effect of body temperature on cold induced vasodilation. Eur J Appl Physiol
. 2008; 104 (4): 491–9.
12. Green D, Cheetham C, Reed C, Dembo L, O’Driscoll G. Assessment of brachial artery blood flow across the cardiac cycle: retrograde flows during cycle ergometry. J Appl Physiol
. 2002; 93 (1): 361–8.
13. Gregson W, Black MA, Jones H, et al. Influence of cold water immersion on limb and cutaneous blood flow at rest. Am J Sports Med
. 2011; 39 (6): 1316–23.
14. Heinonen I, Brothers RM, Kemppainen J, Knuuti J, Kalliokoksi KK, Crandall CG. Local heating, but not indirect whole body heating, increases human skeletal muscle blood flow. J Appl Physiol
. 2011; 111: 818–24.
15. Hensel H, Boman KKA. Afferent impulse in cutaneous sensory nerves in human subjects. J Neurophysiol
. 1960; 23: 564–78.
16. Hodges GJ, Zhao K, Kosiba WA, Johnson JM. The involvement of nitric oxide in the cutaneous vasoconstrictor response to local cooling in humans. J Physiol
. 2006; 574 (Pt 3): 849–57.
17. Ihsan M, Watson G, Lipski M, Abbiss CR. Influence of postexercise cooling on muscle oxygenation and blood volume changes. Med Sci Sports Exerc
. 2013; 45 (5): 876–82.
18. Johnson DJ, Moore S, Moore J, Oliver RA. Effect of cold submersion on intramuscular temperature of the gastrocnemius muscle. Phys Ther
. 1979; 59 (10): 1238–42.
19. Keller DM, Sander M, Stallknecht B, Crandall CG. α-Adrenergic responsiveness is preserved in the heated human leg. J Physiol
. 2010; 588 (Pt 19): 3799–808.
20. Kellogg DL Jr. In vivo mechanisms of cutaneous vasodilation and vasoconstriction in humans during thermoregulatory challenges. J Appl Physiol
. 2005; 100 (5): 1709–18.
21. Kregal KC, Seals DR, Callister R. Sympathetic nervous system activity during skin cooling in humans: relationship to stimulus intensity and pain sensation. J Physiol
. 1992; 454: 359–71.
22. Lee H, Natsui H, Akimoto T, Yanagi K, Ohshima N. Effects of cryotherapy after contusion using real-time intravital microscopy. Med Sci Sports Exerc
. 2005; 37 (7): 1093–8.
23. Lovallo W. The cold pressor test and autonomic function: a review and integration. Psychophysiology
. 1975; 12 (3): 268–82.
24. Merrick MA, Jutte LS, Smith ME. Cold modalities with different thermodynamic properties produce different surface and intramuscular temperature. J Athl Train
. 2003; 38 (1): 28–33.
25. Raven PB, Wilkerson JE, Horvath SM, Bolduan NW. Thermal, metabolic and cardiovascular responses to varying degrees of cold stress. Can J Physiol Pharmacol
. 1975; 53 (2): 293–8.
26. Rådegran G. Limb and skeletal muscle blood flow measurements at rest and during exercise in human subjects. Proc Nutr Soc
. 1999; 58 (4): 887–98.
27. Smith LL. Acute inflammation: the underlying mechanism in delayed onset muscle soreness? Med Sci Sports Exerc
. 1991; 23 (5): 542–51.
28. Thorlacius H, Vollmar B, Westermann S, Torkvist L, Menger MD. Effects of local cooling on microvascular hemodynamics and leukocyte adhesion in the striated muscle of hamsters. J Trauma
. 1998; 45 (4): 715–9.
29. Thorsson O, Lilja B, Ahigren L, Hemdal B, Westlin N. The effect of local cold application on intramuscular blood flow at rest and after running. Med Sci Sports Exerc
. 1985; 17 (6): 710–3.
30. Tschakovsky ME, Sujirattanawimol K, Ruble SB, Vailic Z, Joyner MJ. Is sympathetic neural vasoconstriction blunted in the vascular bed of exercising human muscle? J Physiol
. 2002; 541 (Pt 2): 623–35.
31. Vaile J, Halson S, Gill N, Dawson B. Effect of hydrotherapy on the signs and symptoms of delayed onset muscle soreness. Eur J Appl Physiol
. 2008; 102 (4): 447–55.
32. Vaile J, O’Hagan C, Stefanovic B, Walker M, Gill N, Askew CD. Effect of cold water immersion on repeated cycling performance and limb blood flow. Br J Sports Med
. 2011; 45 (10): 825–9.
33. Wakabayashi H, Hanai A, Yokoyama S, Nomura T. Thermal insulation and body temperature wearing a thermal swimsuit during water immersion. J Physiol Anthropol
. 2006; 25 (5): 331–8.
34. Wilcock IM, Cronin JB, Hing WA. Physiological responses to water immersion. A method for sport recovery? Sports Med
. 2006; 36 (9): 747–65.
35. Wilson TE, Cui J, Crandall CG. Effect of whole-body and local heating on cutaneous vasoconstrictor responses in humans. Auton Neurosci
. 2002; 97 (2): 122–8.
36. Wilson TE, Sauder CL, Kearney ML, et al. Skin-surface cooling elicits peripheral and visceral vasoconstriction in humans. J Appl Physiol
. 2007; 103 (4): 1257–62.
37. Wolf S, Hardy JD. Studies on pain. Observations on pain due to local cooling and on factors involved in the “cold pressor” effect. J Clin Invest
. 1941; 20 (5): 521–33.
38. Woodman RJ, Playford DA, Watts GF, et al. Improved analysis of brachial artery ultrasound using novel edge-detection software system. J Appl Physiol
. 2001; 91 (2): 929–37.
39. Wray DW, Donato AJ, Nishiyama SK, Richardson RS. Acute sympathetic vasoconstriction at rest and during dynamic exercise in cyclists and sedentary human. J Appl Physiol
. 2007; 102 (2): 704–12.
40. Young AJ, Sawka MN, Epstein Y, Decristofano B, Pandolf KB. Cooling different body surfaces during upper and lower body exercise. J Appl Physiol
. 1987; 63 (3): 1218–23.