THE EFFECT OF EXERCISE INTENSITY ON CALF VOLUME AND THERMOREGULATORY RESPONSES DURING UPPER BODY EXERCISE

During upper body exercise the vascular adaptations of the leg have been reported to play an important thermoregulatory role. This study examined the effect of exercise intensity on thermoregulation during upper body exercise. Nine healthy male participants undertook an incremental exercise test on an arm crank ergometer to determine peak power (Wpeak). The participants performed four experimental trials involving 5 minutes of arm exercise at either 45, 60, 75, or 90% Wpeak (70 rev.min -1) followed by 30 minutes of passive recovery. Aural and skin temperatures, upper arm and calf heat fl ow were recorded. Calf volume was measured during exercise using plethysmography. During exercise at 45, 60, 75 and 90% Wpeak calf volume decreased (P<0.05) by -0.7±0.8, -1.4±0.9, -1.2±0.6 and -1.6±0.7% respectively. Differences were observed between 45 and 60% Wpeak, and 45 and 90% Wpeak (P<0.05). The results of this study suggest a redistribution of blood from the relatively inactive lower body during arm exercise of intensities up to 60%Wpeak after which point calf volume does not signifi cantly decrease further. Therefore, the redistribution of blood from the inactive lower body does not produce a similar intensity dependent response to visceral blood fl ow during lower body exercise.


INTRODUCTION
During lower body exercise of moderate intensity in cool conditions skin blood fl ow increases (Johnson and Rowell, 1975), this in turn increases the thermal gradient between the skin and the atmosphere.An increased gradient provides heat for convection, radiation and evaporation of sweat enabling an improved heat exchange (Rowell, 1974).In contrast during upper body exercise in cool conditions calf volume has been demonstrated to decrease (Hopman, Verheifen & Binkhorst, 1993) suggesting arterial blood fl ow to the calf decreases.This is supported by a decrease in calf skin temperature during prolonged upper body exercise (Price and Camp-During lower body exercise forearm skin blood fl ow, which is often used to represent whole body cutaneous blood fl ow (Johnson & Rowell, 1975), has been observed to increase linearly with oesophageal temperature (Smolander et al., 1987;Nadel et al.,1979) up to an exercise intensity of 80% VO 2max .After this point the relationship is attenuated with no further increases in skin blood fl ow (Smolander et al., 1987).The same relationship has not been found for upper body exercise and lower body responses.
In particular research has found a decrease in calf volume (representing skin, muscle and bone blood fl ow) during 10 min of upper body exercise at 50% W peak suggesting a decrease in calf blood fl ow (Hopman, Verheifen & Binkhorst, 1993).However, little is known regarding changes in calf volume at a range of exercise intensities and potential skin blood fl ow responses during upper body exercise.Blood pooling in the lower body could potentially cause thermoregulatory and cardiovascular stability problems during upper body exercise.As exercise intensity increases and the demand for increased blood fl ow at the active muscle and skin occurs, a greater redistribution of blood from the inactive muscles will be required.If blood pooling occurs in the lower body this could potentially cause an increased thermal strain on the body.Consequently, the vascular adaptations of the leg during upper body exercise have been suggested to play an important physiological and, more specifi cally, thermoregulatory role during upper body exercise, but has yet to be fully examined (Price & Campbell, 1997).
During lower body exercise, changes in visceral blood fl ow are intensity dependent (Rowell, 1974) and it is possible that the reported decreases in calf volume may also exhibit the same intensity dependent response.Establishing the thermoregulatory responses in the lower body of able bodied individuals during upper body exercise at different intensities of exercise will help establish a thermoregulatory model that could assist in future exercise recommendations for individuals with for example spinal cord injuries or Multiple Sclerosis.Therefore, the principal aim of this study was to examine the effect of exercise intensity on calf volume changes during upper body exercise in order to examine the redistribution of blood during upper body exercise.It was hypothesised that as exercise intensity increased calf volume would decrease.

Participants
Nine male participants (Mean±s age 22.9±3.6 yrs, weight 78.4±13.7kg)generally but not specifically upper body trained, volunteered to participate in the study.University Ethics Committee approval for the study's experimental procedures was obtained along with written informed consent and followed the principles outlined in the Declaration of Helsinki.

Preliminary Tests
Participants attended the laboratory on fi ve separate occasions.On the fi rst visit, participants performed an incremental exercise test using an electronically braked arm crank ergometer (Lode, Angio, Groningen, the Netherlands), to determine peak oxygen uptake ( V  O 2peak ) and peak power output (W peak ).
The latter was used to determine the exercise intensity for each subsequent experimental trial.Following a ten minute cool down, participants were familiarised with the four different exercise intensities, which were to be undertaken in the following four visits.Each subject's V  O 2peak was determined using a discontinuous incremental exercise protocol (Smith, Price & Doherty, 2001).Participants sat with their shoulder joint in line with the crank shaft of the ergometer.Their feet were positioned fl at on the fl oor in metal cups to ensure the knee joint was at 90°.The protocol consisted of four, four-minute submaximal exercise stages (30, 50, 70, and 90W respectively) with two minutes rest between stages.Subjects were required to maintain a cadence of 70 rev.min - throughout exercise.After the fi nal four minute stage and rest period, the protocol increased by 20W every two minutes from 110W until volitional fatigue, or when participants were unable to maintain 70 rev.min -1 .Expired gas (Metamax 3B, Leipzig, Germany) and heart rate (HR; Polar Accurex Plus, Kempele, Finland) were continually monitored.

Submaximal Trials
All exercise trials were performed at an ambient temperature of 20.51.4C and 63.8±5.8%humidity.Participants arrived at the laboratory after fasting for two hours and abstaining from strenuous exercise during the previous 24 hours.The study was Bellevue, USA) consisted of two mercury fi lled tubes of silicon rubber, positioned around the thickest part of the left calf and taped into position.A contoured blood pressure cuff was placed on the left thigh, just above the knee, which was connected to a rapid cuff infl ator (Hokanson E20, Bellevue, USA) set to infl ate to 50mmHg for fi ve seconds and defl ate for eight seconds.The plethysmograph was connected to a personal computer via Powerlab (AD Instruments, Chalgrove, UK) and online recordings displayed through Chart4windows.An average of three infl ations was used to determine blood fl ow at each time point.
Participants performed arm crank exercise for 5 minutes at 70 rev.min - at the randomly assigned exercise intensity.Heat fl ow and skin temperatures were recorded every minute during and on the cessation of exercise.Changes in calf volume from rest were recorded throughout exercise.RPE central and RPE local were recorded 30s before the cessation of exercise.Oxygen uptake ( V  O 2 ), minute ventila- tion (VE) and respiratory exchange ratio (RER) were averaged over the last 30s of each minute during exercise and at fi ve minute intervals during 30 minutes of passive recovery.Blood fl ow was recorded at rest and at the cessation of exercise.Immediately post exercise a further capillary blood sample was taken from the earlobe for the measurement of blood lactate.Participants then remained seated for 30 minutes of recovery.Heat fl ow, blood fl ow, aural and skin temperatures were recorded every 5 minutes of passive recovery.Body mass was recorded pre and post exercise to calculate sweat losses.

Statistical Analysis
All data were analysed via Minitab version 14.0.Data were analysed by Two-Way Analysis of Variance (time x intensity) with repeated measures.Where signifi cance was obtained, Tukey post hoc was performed.Where appropriate Pearson's correlation was performed to investigate non-causal relationships between variables e.g.blood lactate verse RPE.Data are represented as mean ± SEM in fi gures, and mean±s in tables.Signifi cance was taken as P<0.05.conducted as a cross-over design.Each subject was tested on four separate occasions with at least three days separating each trial.The trials consisted of 5 minutes of arm exercise at either 45, 60, 75 or 90% of W peak , followed by 30 minutes of passive recovery.
On arrival at the laboratory, body mass was recorded using a balance beam scale (Seca, Hamburg, Germany).Participants wore shorts, socks, and training shoes and rested for 15 minutes while thermistors and sensors were attached.Aural and skin thermistors were continuously recorded via a data logger (Squirrel 1020 series, Cambridge, UK).Aural temperature was measured by an aural thermistor (Grant, Cambridge, UK) inserted into the subject's auditory canal and securely taped into position and insulated with cotton wool.Skin thermistors (Grant, Cambridge, UK) were attached to standard anatomical landmarks for the upper arm, back, chest, thigh, and calf.An additional thermistor was applied to the second toe on the right foot with the shoe and sock replaced.Thermistors were attached to the skin using strips of water permeable tape (3M Transpore, Loughborough, UK).Mean skin temperature (T ms ) was calculated using the equation of Ramanathan (1964) where: T ms = 0.3(T chest )+ 0.3 (T upper arm ) + 0.2(T calf ) + 0.2 (T thigh ).Heat fl ow was measured at the calf and the upper arm using heat fl ow sensors (Data Harvest Easy Sense Advanced, Bedfordshire, UK) and was recorded continuously.Sensors were attached adjacent to the skin thermistors on the calf and upper arm and were attached using water permeable tape according to the manufacturer's guidelines.
During each submaximal trial expired gas was continuously analysed.Baseline data for all measures were obtained during the seated rest for fi ve minutes prior to exercise and a resting capillary blood sample (80μl) taken from the earlobe for measurement of blood lactate (Analox GM7, London, UK).Strain gauge plethysmography was employed to measure relative volume changes of the calf during arm exercise.Lower leg blood fl ow was measured at rest using venous occlusion plethysmography and standard procedures (Hopman, Verheifen & Binkhorst, 1993).The strain gauge plethysmograph (Hokanson EC6,

Preliminary Tests
The peak physiological responses obtained during the incremental test for V  O 2peak are shown in Table 1.

Physiological responses during exercise and passive recovery
The physiological responses at the cessation of each submaximal arm exercise trial are shown in Table 2.A difference between exercise intensities was observed for both HR and V  O 2 during exercise and recovery (main effect; P<0.05).During passive re-covery, HR decreased but remained higher (P<0.01)than rest in all but the 45% W peak trial.Blood lactate increased linearly with exercise intensity (P<0.05).Body mass decreased in all trials (-0.1 ±0.1, -0.2 ±0.1, -0.1 ±0.1 and -0.1 ±0.1kg for 45, 60, 75 and 90% W peak respectively) with no signifi cant differences between the trials (P>0.05).

Aural Temperature during Exercise and Passive Recovery
A main effect for exercise intensity was observed for aural temperature (P<0.05).Post hoc analysis revealed differences between 45 and 90% W peak trials and, 60 and 90% W peak trials.At the end of exercise in all trials aural temperature had decreased from resting values by -0.04 ±0.21, -0.03 ±0.20, -0.13 ±0.08 and -0.14 ±0.24°C for 45, 60, 75 and 90% respectively.

Upper Body Skin Temperatures during Exercise and Passive Recovery
Upper arm skin temperature initially decreased during the fi rst minute of exercise (P<0.05; Figure 1).
Immediately post exercise upper arm skin temperature increased for all trials (P<0.05;main effect for time).There was a main effect for exercise intensity with a greater increase in temperature occurring during the higher intensity trials (75 and 90% W peak ) when compared to the lower intensity trials (45 and 60% W peak ).A main effect for time was observed (P<0.05) for chest temperature which decreased during exercise followed by an increase post exercise for all trials.There was no difference between exercise intensities for chest temperature (P>0.05).Back skin temperature remained relatively constant throughout exercise and passive recovery for each exercise intensity (P>0.05).

Lower Body Skin Temperatures during Exercise and Passive Recovery
Thigh skin temperature decreased during exercise and passive recovery (main effect for time P<0.05).Differences were observed between 60 and 75% W peak trials, 60 and 90% W peak trials (main effect for intensity; P<0.05).During exercise calf temperature remained constant however, during recovery calf temperature decreased in all trials (P<0.05), with a greater decrease in temperature during recovery from the higher exercise intensities (75 and 90% W peak ; However, there was a tendency for further increases in heat fl ow during exercise but not signifi cant in any trial.On the cessation of exercise heat fl ow decreased (P<0.05) and remained constant throughout passive recovery.Heat fl ow during exercise and recovery at 90% W peak was greater than the other three intensities (P<0.05).No further differences were noted between trials.In contrast to the upper arm, heat fl ow from the calf remained constant throughout exercise and passive recovery at all exercise intensities (P>0.05; Figure 3b).However, heat fl ow was lower during the 45% W peak when compared to the other three intensities at rest and throughout both exercise and passive recovery (main effect for intensity; P<0.05).

Calf Volume and Blood Flow during Exercise and Passive Recovery
When compared to rest, calf volume decreased (P<0.05) during exercise at each exercise intensity (-0.7 ±0.8, -1.4 ±0.9, -1.2 ±0.6, and -1.6 ±0.7% respectively; Figure 4).Differences were observed between 45% and 60% W peak , and 45 and 90% W peak (P<0.05).Calf volume returned to baseline within fi ve minutes of passive recovery in all trials.Blood in arm skin temperature, most likely due to the greater metabolic heat production than 40 and 60% W peak trials.Consequently, greater local heat storage is likely to have occurred as changes in convective air currents would have been the same for each trial (i.e.all exercise performed at the same movement speed; 70rev.min - ).Conversely, the initial decrease in arm temperature during exercise was probably due to increased heat fl ow as a result of convection current generation on the initiation of movement (Mitchell, 1977).Sawka et al., (1984) suggested that heat fl ow values were representative of skin conductance and therefore provide a relative index of cutaneous blood fl ow.Therefore, in the present study blood fl ow to the upper arm must have increased based on the increased heat fl ow.Absolute arm heat fl ow was similar for the 45, 60 and 75% trials but greater for the 90% trial suggesting a greater increase in skin blood fl ow at the higher exercise intensity.This may repre-fect for time; P<0.05).There was also a main effect for trial, with blood fl ow being signifi cantly lower during the 45%W peak trial compared to the other tri-als (P<0.05), with no other differences occurring (P>0.05).

DISCUSSION
The principal aim of the study was to determine the effects of exercise intensity on calf volume using strain gauge plethysmography as a means of examining the redistribution of blood during arm exercise.The main fi nding of the study was a decrease in calf volume during arm exercise of intensities up to 60%W peak after which point calf volume did not signifi cantly decrease further.This decrease in calf volume was accompanied by an increase in calf blood fl ow from rest to the end of exercise.

Upper Body Thermoregulatory Responses
During the fi rst 5 minu tes of passive recovery there was a signifi cant increase in arm skin temperature in all trials due to decreased convective air currents and resultant decrease in arm heat fl ow.The 75 and 90% W peak trials demonstrated greater increases temperature during exercise, but as previously mentioned the exercise duration was considerably longer (60-90 min).It was suggested in this instance that the decrease in calf temperature during prolonged arm exercise at constant exercise intensity (60% V  O 2peak ) and environmental temperature was due to decreased blood fl ow to the calf.Since the calf is relatively metabolically inactive during upper body exercise, longer durations of exercise may be required for signifi cant changes in calf skin temperature to occur due to reduced blood fl ow and changes in thermal state.Therefore examining the relationship between calf volume and skin temperature over greater durations of exercise would contribute to our understanding of calf blood fl ow and skin temperature relationship.
Calf volume immediately began to return towards baseline once exercise ceased suggesting blood fl ow to the muscle was returning.This could be potentially explained by a rapid vasodilation at the calf occurring immediately at the cessation of exercise causing a rapid increase in blood fl ow (hyperaemia) to the calf.During the present study calf volume was observed to decrease during exercise which was most likely due to increased muscle sympathetic nerve activity (Saito et al. 1993) causing vasoconstriction in the calf.As soon as exercise ceased a rapid vasodilation occurred causing a rapid increase in calf blood fl ow as blood returned to the calf.This is demonstrated by the fact that blood fl ow was greatest on the fi rst measurement post exercise compared to the second and third measurements however this only occurred during the 75 and 90%W peak trials.Muscle sympathetic nerve activity (MSNA) is exercise dependent (Saito et al. 1993) therefore the MSNA was only suffi cient during exercise in the 75 and 90%W Peak trials to cause this rapid hyperaemia as exercise ended.Blood fl ow decreased towards baseline values for the remainder of passive recovery, following this initial increase.
Several studies have determined the average volume of the calf and results have varied between 1.7 -3.4L (Hargens, 1983;Convertino et al. 1989;Moore & Thornton, 1987).Therefore using these approximate volumes the amount of blood being redirected away from the calf during upper body exercise in this study can be estimated.Calf volume decreased between 0.7-1.6%during exercise in all studies which equates to about 12-54 ml of blood being redirected away from the calf.Therefore, the results sent the exercise intensity where direct heat transfer from the contracting muscle was greater.Indeed, if heat fl ow does represent skin blood fl ow (Sawka et al., 1984) then this would have allowed greater heat transfer from the upperarm to the environment in the 90% trial resulting in a similar absolute increase in skin temperature to the other trials.

Lower Body Thermoregulatory Respons
A decrease in calf volume was observed during all exercise intensities, however, after 60% W peak no further decreases were observed.Hopman, Verheifen & Binkhorst (1993) measured calf volume during 10 min of arm exercise at 50% W max and observed a decrease in volume of 0.1% per minute at 5 min of exercise, which is the same rate of decrease in limb volume per minute as during the 45% W peak trial in the present study.However, the present study has expanded upon these data in observing that the decrease in calf volume was not linear with exercise intensity, and therefore does not express the same intensity dependent decrease as visceral blood fl ow.This would most likely be due to any further decrease in leg blood fl ow with higher exercise intensities resulting in a relatively small fl ow rate which may compromise oxygen delivery to the lower limb.The most likely role of calf vasoconstriction during the initial stages of exercise is to help maintain cardiovascular stability.
In contrast to the present study's results and Hopman, Verheifen & Binkhorst (1993), Theisen et al., (2001aTheisen et al., ( , 2001b) ) observed that during arm exercise at 50% W peak and 80% W peak skin blood fl ow of the calf increased during exercise.These authors measured skin blood fl ow using Laser Doppler Flowmetry which only measures blood fl ow up to 1.5mm below the skin (Johnson et al., 1984).+Therefore, the decrease in calf volume observed in the present study using plethysmography, may be due to a reduction in predominantly muscle blood fl ow rather than blood fl ow to the superfi cial skin layers (Saumet et al., 1988).This is supported by Seals (1989) who suggested that during handgrip exercise skin blood fl ow increases while muscle blood fl ow to the calf decreases.
Although there was a decrease in calf volume during exercise, there was no signifi cant difference in calf temperature or calf heat fl ow.Previous research (Price & Campbell, 1997)

CONCLUSION
The results of this study suggest a redistribution of blood from the relatively inactive lower body during arm exercise of intensities up to 60%W peak after which point calf volume does not signifi cantly decrease further.Calf blood fl ow immediately at the end of exercise was greater than that at rest.The most likely explanation is that at the end of exercise rapid vasodilation occurred in the calf causing an increase in blood fl ow thus calf volume returned to baseline levels within 5 minutes of exercise ceasing.The calf volume decrease is therefore likely a result of vasoconstriction reducing blood pooling in the leg due to an increase in muscle sympathetic nerve activity.Reducing the venous pooling that is occurring during upper body exercise could substantially improve upper body exercise performance and may benefi t sports which have a large upper body component such as kayaking.
Future research should focus upon the differentiation between calf muscle and calf skin blood fl ow during and following arm exercise, and the effects of longer durations of arm exercise on calf volume and skin temperature in order to examine further the thermoregulatory responses during upper body exercise.demonstrate that the lower body appears to have a small contribution not only to thermoregulation but also to cardiovascular stability.
The decrease in calf skin temperature post exercise may possibly be explained by counter current heat exchange within vascular bundles of the leg.Blood fl ow to the arteries in the calf region post exercise would deliver warm blood from the core with heat most likely being transferred to the adjacent and cooler blood in the veins leaving the calf.The arterial blood would therefore be cooled, which on delivery to the cutaneous circulation of the leg would result in a decrease in skin temperature.Consequently fl ow and nutrient supply can be restored without any increase in limb temperature.

Figure 1
Figure 1 Mean (±SEM) relative changes in upper arm skin temperature during each exercise trial at 45, 60, 75 and 90% W peak and passive recovery (n=9).EX denotes exercise period.

Figure 2
Figure 2 Mean (±SEM) relative changes in calf skin temperatures during each exercise trial at 45, 60, 75 and 90% W peak and passive recovery (n=9).EX denotes exercise period.

Table 2
Mean (±s) physiological responses at the cessation of arm exercise during each submaximal trial (n=9) *Signifi cant (P<0.05)main effect between exercise intensities.