Introduction

Human body temperature is highly regulated.1 Afferent signals from various body sites are integrated in the hypothalamus where appropriate effector responses regulate core temperature around a given set point.2 The major effector responses that regulate heat storage include vasomotor changes, sweating, shivering and changes in blood flow, with the intensity of these effector responses being proportional to the core temperature.

The temperature of arterial blood reaching the hypothalamus is typically represented as the core temperature and reflects the balance between heat production and heat loss. To estimate the core temperature, temperature at sites other than the hypothalamus are recorded, including the rectum, oseophagus and tympanic membrane. Evidence suggests that temperature measured in the oseophagus at rest and during exercise is a valid estimate of arterial blood temperature and therefore temperature of the blood reaching the hypothalamus.3,4,5,6,7 Rectal temperature, while having slower response times, does also track arterial blood temperature.8,9,10

Passive heating is typically described as heat exposure without exercise11 and can include warm water immersion (hot baths).12 Immersion in water at 39°C can seriously challenge efferent heat loss responses because of the high thermal conductivity of water and a limited skin area available for evaporative or convective heat loss. Prolonged immersion (60 min), in 39°C water can result in a rapid increase in heat storage or core temperature (Tcore). Indeed, warm water immersion or hydrotherapy is frequently used as a therapeutic modality for patients with a spinal cord lesion, often with little regard to the thermal conductivity of water and heat balance implications.

During dynamic exercise in a thermoneutral environment, an additional thermal load proportional to the intensity of the exercise is imposed which may result in a higher core temperature and more intense effector responses than passive heating alone.

Traumatic tetraplegia results in motor, sensory and autonomic deficits that are dependent upon the site and extent of spinal cord damage. Complete spinal cord lesions above the fifth cervical segment, will result in the loss of motor and sensory innervation to some regions of the arms and all regions of the legs, hips and trunk. The loss of the sympathetic division of the autonomic nervous system will also affect cardio-acceleration and redistribution of blood flow.13,14

When people with tetraplegia exercise at high relative intensities in thermo-neutral environments, the absolute heat production from exercise is small because of the reduced amount of innervated and active muscle mass.15 Despite limited sweating and evaporative capacity, people who are tetraplegic competing in a marathon event under neutral environmental conditions can expect convection to be the main avenue for heat loss. Accordingly, with a reduced innervated muscle mass and with the wheelchair moving at speeds of approximately 5.0 m·s−1, the rise in core temperature over 2 h would be expected to be 0.50–1.00°C.15

This may not be the case when people with tetraplegia use hydrotherapy and are immersed to the nipple line in warm water. The high thermal conductivity of water favours a high rate of heat gain, dependent partially on water temperature and, at 39°C can be characterised by a 1.0–3.0°C rise in core temperature within 60 min.

Given the widespread use of exercise and hydrotherapy after spinal cord injury, the aim of the present experiment was to compare the thermoregulatory responses of four highly trained men with tetraplegia during active heating via 40 min of exercise at 65% V̇O2 peak and during passive heating via 60 min of immersion in 39°C water.

Method

Four physically trained men with tetraplegia (competitors in international wheelchair marathons) volunteered to participate in the study which had been approved by an institutional ethics committee. After giving consent, each participant visited the laboratory on three occasions.

The first visit was to provide familiarisation followed by an incremental wheelchair exercise test to exhaustion on a motorised treadmill to determine peak oxygen consumption (V̇O2 peak). A second visit either involved 40 min of exercise at 65% V̇O2 peak or 60 min of warm (39°C) water immersion. A third visit was the alternate condition. Visitations were separated by two to three days, and were randomly allocated with each subject completing each condition.

At the first visit, each participant was weighed wearing a swimming costume and an emptied catheter bag. Skinfold thickness was measured at four sites (triceps, biceps, subscapular and suprailiac) on each subject on the right side. Electrocardiograph electrodes were placed in the CM5 placement to monitor and record heart rate and rhythm throughout the incremental wheelchair test.

After a two min seated rest period, each participant performed an incremental exercise test by pushing their track wheelchair on a motor driven treadmill until exhaustion. Participants began at a speed of 0.5 km·h−1 with increments of 0.5 km·h−1 and 0.5% grade every minute. Throughout the incremental wheelchair exercise test, participants breathed through a Hans Rudolph valve (no 2700) and expired gas passed through a calibrated pneumotach (Hewlett-Packard model HP 47304A) into a baffled mixing chamber. Oxygen and carbon dioxide analysers (Rapox and Capnograph) continuously sampled mixed expired gas from the distal end of the mixing chamber. Every 30 s calculations were made of O2 uptake (V̇O2) CO2 production (V̇CO2) and minute ventilation (V̇E). Each analyser was calibrated before and after each exercise test using a mixture of precision reference gases.

Heart rate (HR) was monitored throughout and recorded every minute. All experiments were conducted in a controlled environment (23°C Dry Bulb, 18°C Wet Bulb).

Exercise at 65% V̇O2 peak

A thin, flexible, calibrated rectal probe (AD590) was inserted approximately 14 cm beyond the anal sphincter to estimate core temperature (Trec). Skin temperature (Tsk) was recorded from the forehead and the calf using AD590 technology.

A sweat capsule, using the principle of dewpoint hygrometry,16 was secured to the forehead of each participant and sealed with a thin layer of paraffin jelly at the skin-capsule interface. Changes in dewpoint temperature were used to detect the onset of sweating, and calculate sweat rate.

A surflo catheter was inserted into either a vein on the dorsum of the hand, or the lower third of the extensor aspect of the forearm with patency maintained by flushing with 15 IU.ml−1 of sterile heparinised saline. Venous blood was withdrawn at the beginning of the pre-exercise period (−20 min), immediately before exercise (0 min) and 30–90 s after the 65% V̇O2 peak test finished, and analysed for haemoglobin (Hb) and haematocrit (Hct). Relative changes in plasma volume were calculated according to the formula of Greenleaf et al (1979)17. Separated plasma was analyzed for noradrenaline (NAD) and adrenalin (AD) using high performance liquid chromatography (HPLC).

After a 20 min seated pre-exercise rest period, each participant pushed their wheelchair at approximately 65% V̇O2 peak on the motor driven treadmill for up to 40 min. V̇O2, V̇CO2 and HR were measured at 15–20 and 30–35 min of the exercise test as described for Visit 1.

Warm water immersion (39°C)

Rectal and skin temperature probes, sweat capsule, surflo catheter and ECG electrodes were placed as previously described. Venous blood was withdrawn 20 min before immersion (−20 min), just prior to immersion (0 min) and just before coming out of the pool (60 min). Venous blood was analyzed as previously described.

Each participant sat in a wheelchair in a thermoneutral environment (23°C DB; 18°C WB) while pre-immersion values (−20,0 min) were obtained. Once the 0 min seated pre-immersion venous blood sample was obtained, the participant was wheeled to the next room and transferred in the sitting position to a lightweight plastic chair and lowered into the 39°C water to the nipple line. Trec, Tsk and forehead dew point temperature were monitored continuously and recorded each 30 s. HR was monitored continuously and recorded every 2 min. The participant remained in the pool for up to 60 min, or until Trec reached 38.5°C.

Results

Participants' descriptive statistics, including sporting achievements, are presented in Table 1. All participants became tetraplegic as a result of trauma, and had established themselves as elite wheelchair athletes.

Table 1 Personal and anthropometric characteristics of participants
Full size table

The results from the incremental treadmill wheelchair test to exhaustion are presented in Table 2. The V̇O2 peak (l·min−1, ml·min−1·kg−1) and peak heart rate (HR) (b·min−1) were 1.14±0.16 l·min−1, 16.7±1.8 ml·kg−1·min−1 and 99±4 b·min−1 respectively.

Table 2 Peak physiological data from incremental wheelchair test to exhaustion
Full size table

Heart rate and rectal temperature (Trec) recorded for each participant during the exercise test at 65% V̇O2 peak and during water immersion is shown in Figure 1.

Figure 1
figure 1

Heart rate and rectal temperature (Trec) during exercise and immersion

Full size image

Pre-immersion heart rate (0 min) was 67±4 b· min−1 rising to 75±4 b·min−1 and 87±3 b·min−1 after 40 and 60 min, respectively. The pre-exercise rate heart rate was 68±3 b·min−1 rising to 91±5 b·min−1 after 40 min of exercise at 60–65% V̇O2 peak. Rectal temperature (Trec) rose from 35.97±0.30°C at rest to 36.46±0.40°C after 40 min and to 37.32±0.51°C after 60 min of immersion. The Trec increased from 36.42±0.20°C at rest to 36.67±0.19°C after 40 min of exercise.

Calf and head skin temperature obtained during exercise and water immersion are shown in Figure 2.

Figure 2
figure 2

Calf and head skin temperatures (Tcalf, Tskin) during exercise and immersion

Full size image

Participants 1, 2 and 4 demonstrated no centrally driven forehead sweating during exercise or immersion. Participant 3 (C 5/6) commenced sweating after 20 min of exercise (0.1 mg·min−1·cm2) and reached 0.4 mg·min−1·cm−2 after 35 and 40 min of exercise. This same participant (no.3) commenced sweating after 5 min of immersion (0.2 mg·min−1·cm−2) and maintained this sweat rate until the 50 min after which sweat rate increased to 0.3 mg·min−1·cm−2 until the end of water immersion (60 min). For participant 3, sweating onset during exercise and pool immersion occurred at Trec 36.8°C and 36.12°C, respectively.

Plasma adrenalin (AD) and noradrenalin (NAD) concentrations measured during water immersion and exercise are presented in Table 3.

Table 3 Plasma noradrenalin (NAD) and adrenalin (AD) concentration, haemoglobin (Hb), and haematocrit (Hct), during exercise (Ex) and water immersion (WI) (mean±SE)
Full size table

Haemoglobin and haematocrit values are presented in Table 3. Haemoconcentration occurred during 40 min of exercise (Δ% PV:-8.6±2.3) and haemodilution occurred throughout 60 min of water immersion (Δ% PV:+2.2±1.8). Changes in plasma volume volume during immersion and exercise are shown in Figure 3.

Figure 3
figure 3

Changes in plasma volume during exercise and immersion

Full size image

Discussion

The participants in the present study were classified as having clinically complete tetraplegia and were successful elite athletes. The absolute (l·min−1) and relative (ml·kg−1·min−1) V̇O2 peak for the present participants, is consistent with values reported for other athletes with tetraplegia18 and reflects not only their highly trained state but also the reduced amount of innervated muscle mass. Despite the low absolute value, V̇O2 peak (l·min−1) was approximately 50% higher than similarly aged but sedentary persons with tetraplegia.19 The maximum heart rate and minute ventilation rate recorded at exhaustion during the incremental exercise test are consistent with the physiological consequences of cervical spinal cord injury and with values reported for other tetraplegic athletes.18 The maximum minute ventilation reflects the reduced level of innervation to the intercostal and other accessory respiratory muscles resulting from injury to the cervical spinal cord.

Trec recorded prior to exercise at 65% V̇O2 peak (36.42±0.2°C) and prior to immersion (35.97±0.3°C) is consistent with the values reported for persons with tetraplegia.15

The low resting Trec, typically found in persons with tetraplegia does, provide an added margin of safety when exposed to hot, but not cold environments. The present study indicates that a rise in Trec of 2.5–3.0°C would need to occur before raising the spectre of heat stress/heat stroke. In those situations where resting core temperature may be elevated due to bladder infection, travel immunisation or prior hydrotherapy a shorter immersion time will be required before reaching 38.5–39.0°C. The continued use of hydrotherapy with people who are tetraplegic should include consideration of their initial core temperature, time of immersion and water temperature.

By comparison, the rise in Trec after 40 min of exercise at 60–65% V̇O2 max was 0.35±0.12°C and reflects a low heat storage resulting primarily from a reduced absolute metabolic heat production (eg 15 kJ·min−1) and the raised convective heat loss by propelling a wheelchair at approximately 5 m·s−1.

The Tcalf and Thead recorded at rest and prior to immersion and exercise were consistent with values reported by others.15 The direction and magnitude of change in Tcalf and Thead primarily reflects the nature of the two stressors used to raise Tcore. The temperature recorded at the calf during immersion would be primarily a reflection of the water temperature rather than a centrally driven increase in skin blood flow.

During 40 min of exercise Tcalf remained essentially unchanged and is consistent with the results of other trained men with tetraplegia.15 It has been suggested6,20 that an intact pathway from above C5 to at least T11 is required to mediate active vasodilation in the lower limb. Accordingly, participants in the present study, because of their cervical cord lesions, could not be expected to actively redistribute bloodflow to the skin and raise Tcalf. Skin blood flow is an important contributor to skin temperature21 and on the basis of Tcalf recorded during exercise it is reasonable to conclude that in the present participants there was no centrally mediated vasodilation in the calf.

The consistency of Thead recorded during exercise reflects the absence of sweating and evaporation and the convective effects of pushing a wheelchair on a motor driven treadmill at approximately 10 km·h−1.

The amplitude and direction of change in heart rate during exercise is consistent with the results of others.15 No previous studies report heart rate changes during immersion in warm water in people with tetraplegia. Compared to exercise, the lower heart rate recorded after the end of immersion was accompanied by a lower plasma noradrenalin concentration and an expanded plasma volume. Using impedance methodology to estimate cardiac output, immersion has been associated with increases in cardiac output of 25% and 35% in participants with tetraplegia after 60 and 120 min.22 Whether the decrease in HR observed during immersion in the present study was accompanied by an increase in stroke volume is not known, as cardiac output was not measured.

The pre-exercise expansion in plasma volume found in the present participants has also been reported in other adults with paraplegia.23 It has been suggested that the pre-exercise plasma volume expansion in persons with paraplegia may be a reflection of an exercise anticipatory sympathetic response affecting vascular constriction and favouring inward filtration and plasma volume expansion.23 Given that all participants in the present study were assessed as having clinically complete cervical cord lesions, it is difficult to attribute the pre-exercise plasma volume expansion to a centrally derived sympathetic vascular constriction. The expansion of plasma volume during immersion is consistent with the results of others using subjects who are able bodied24 and has been attributed to the depth related pressure generated by immersion in water.

The reduction in plasma volume (−8.6±2.3%) that occurred after 40 min of exercise at 65% V̇O2 max was larger than the plasma volume reduction reported for adults with paraplegia (T4–T12) after 40 min of exercise at 50% V̇O2 peak (−2.5%) but similar to the reduction (−9%) in men with paraplegia (T5–T12) after 40 min of exercise at 60–65% V̇O2 peak.23,25

The reduction in plasma volume during exercise has been ascribed to various factors such as decreased lymph flow, methodological problems related to changes in posture, raised capillary hydrostatic pressure and sweating.26,27,28 It was expected that posture, capillary hydrostatic pressure and sweating would have minimal effect on the plasma volume reduction observed during 40 min of exercise in the present study. All participants remained seated throughout the exercise and immersion, the increase in capillary hydrostatic pressure was expected to be small, and only one participant exhibited modest sweating (0.4 mg·min−1·cm−2) after 40 min of exercise.

The seated posture, low absolute V̇O2/cardiac output and the absence of consistent central control of sweating provides little guidance as to the mechanism for the plasma volume reduction in tetraplegics during 40 min of exercise at 65% V̇O2 max.

In summary, the present study found a higher rate of heat gain (ΔTrec) and an increased plasma volume expansion after 40 min of water immersion (39°C) than after 40 min of exercise at 65% V̇O2 max in highly trained men with tetraplegia.