The mechanism(s) for the changes in cardiac function during heat stress remain unknown. This study tested two unique hypotheses. First, sympathetic innervation to the heart is required for increases in cardiac systolic function during heat stress. This was accomplished by comparing responses during heat stress between paraplegics versus tetraplegics, with tetraplegics having reduced/absent cardiac sympathetic innervation. Second, stimulation of skin thermoreceptors contributes to cardiovascular adjustments that occur during heat stress in humans. This was accomplished by comparing responses during leg only heating between paraplegic versus able-bodied individuals. Nine healthy able-bodied, nine paraplegics, and eight tetraplegics participated in this study. Lower body (i.e., nonsensed area for para/tetraplegics) was heated until esophageal temperature had increased by ∼1.0°C. Echocardiographic indexes of diastolic and systolic function were performed before and at the end of heat stress. The heat stress increased cardiac output in all groups, but the magnitude of this increase was attenuated in the tetraplegics relative to the able-bodied (1.3 ± 0.4 vs. 2.3 ± 1.0 l/min; P < 0.05). Diastolic function was maintained in all groups. Indexes of left atrial and ventricular systolic function were enhanced in the able-bodied, but did not change in tetraplegics, while these changes in paraplegics were attenuated relative to the able-bodied. These data suggest that the cardiac sympathetic innervation is required to achieve normal increases in cardiac systolic function during heat stress but not required to maintain diastolic function during this exposure. Second, elevated systolic function during heat stress primarily occurs as a result of increases in internal temperature, although stimulation of skin thermoreceptors may contribute.
- cardiac systolic and diastolic function
- skin thermal receptors
heat stress causes a significant strain on the human cardiovascular system (18). In healthy individuals heat stress increases cardiac output, which offsets reductions in systemic vascular resistance, resulting in little to no changes in arterial pressure (18). This increase in cardiac output primarily occurs via increases in heart rate since stroke volume does not change or may only slightly increase/decrease (1, 4, 13, 15, 16, 19, 22). Cutaneous vasodilation accompanying heat stress decreases ventricular filling pressure, ventricular volume, and central blood volume (4, 16, 19, 22). Without compensation, decreases in ventricular filling pressure will reduce stroke volume, although this generally does not occur during heat stress due to increases in cardiac systolic function coupled with maintained diastolic function (1, 4, 15).
Heat stress-induced increases in heart rate occur primarily through autonomic innervation of the heart, although a relatively small component of this response can be due to direct heating of the heart (6, 10). In contrast, the effect of heat itself on cardiac contractility (i.e., in the absence of autonomic innervation) in humans remains unknown, although heating of isolated rat hearts did not improve indexes of contractility (12). To that end, the first objective of this project tested the hypothesis that sympathetic innervation to the heart is required for heat stress-induced increases in left ventricular systolic function in humans. This objective was achieved by evaluating such responses during whole-body heat stress from individuals with absent (or markedly reduced) sympathetic innervation to the heart secondary to a cervical spinal cord injury (CSCI, i.e., tetraplegics) relative to responses from individuals with a thoracic spinal cord injury (TSCI) having preserved cardiac sympathetic innervation (i.e., paraplegics).
Second, integration of thermal afferent signals (e.g., changes in internal or skin temperatures) at the preoptic area in the hypothalamus (i.e., thermoregulatory center) are the primary controllers of efferent thermoregulatory responses, such as sweating and cutaneous vasodilation/constriction (8, 20). However, the contribution of skin afferent signals, relative to internal temperature afferent signals, responsible for elevations in heart rate and systolic cardiac function during heat stress in human are unknown. To that end, the second objective of this study tested the hypothesis that skin thermoreceptors contribute to cardiovascular adjustments that occur during heat stress in humans. This objective was accomplished by comparing cardiovascular responses to heating of sensate skin (i.e., the legs of able-bodied individuals) relative to heating insensate skin (i.e., legs of TSCI), while controlling for increases in core body temperature between these two groups.
Thus, by evaluation of heat stress-induced responses between able-bodied, TSCI, and CSCI groups, greater insight could be obtained regarding the mechanisms responsible for cardiac changes associated with that exposure in able-bodied individuals.
Nine healthy able-bodied males (AB), nine males with paraplegia (TSCI), and eight males with tetraplegia (CSCI) participated in this study. Both TSCI and CSCI subjects had a traumatic complete spinal cord injury and were functionally nonambulatory (American Spinal Injury Association grade A). In addition to the clinical diagnosis report, motor and sensory examinations were performed in their first visit. Subjects′ characteristics, inclusive of lesion location, time since injury, and medications, are presented in Table 1. Most of the injured volunteers actively participated in sports such as wheelchair basketball or wheelchair marathons and consistently performed some form of exercise training, with the exception of one TSCI and three CSCI volunteers. Subjects were excluded if they had comorbid neurological/medical conditions or were taking medications that primarily affected cardiac function. Subjects were informed of the purpose and risks of the study before providing their informed written consent. The protocol and consent were approved by Wakayama Medical University (No. 1091). Subjects refrained from alcohol, caffeine, and exercise for 24 h before the study. The protocol was performed in accordance with the Declaration of Helsinki.
Instrumentation and measurements.
Upon arrival to the laboratory a medical doctor assessed the patients, including reconfirming the location of the lesion through review of medical records and physical assessment. The border between sensed and nonsensed skin was assessed via patient feedback by a pressure stimulus and by placement of a warmed syringe containing 50°C water against the skin at different locations. Each subject inserted a thermocouple via the nasal passage to a distance equivalent to one-fourth the subject's height for the measurement of esophageal temperature. Skin temperatures were measured at nine sites (forehead, chest, upper back, middle of front and back, abdomen, lower back, thigh, and lower leg), while mean skin temperature was calculated using six of these sites (chest, upper back, abdomen, lower back, thigh, and lower leg) with corresponding ratios as presented by Taylor et al. (21). Mean upper body skin temperature was calculated from the average of the upper body sites (i.e., chest, upper back, middle of front and back, abdomen, lower back).
For TSCI subjects, the chest and back thermocouples were located in sensed area, while the abdomen, lower back, thigh, and lower leg thermocouples were placed on nonsensed area. For CSCI, all thermocouples, except the forehead thermocouple, were placed on nonsensed areas. Each subject was fitted with a two-piece water-perfused tube-lined suit (Med-Eng, Ottawa, Canada) that covered the entire body except for the head, face, hands, one forearm, and feet. This suit permitted the control of skin and internal temperatures by adjustment of the temperature of the water perfusing the suit. Heart rate was continuously obtained from an electrocardiogram (Nihon Kohden; BSM-2401, Tokyo, Japan). Blood pressure was continuously recorded from a finger using the Penaz method (Portapres, Finapres Medical Systems, the Netherlands) along with intermittent brachial artery blood pressures. Skin blood flow was measured at the chest and thigh via laser-Doppler flowmetry (Omegaflo Flo-C1, Tokyo, Japan), with these probes placed under the water perfused suit.
Blood samples (20 ml) were collected from the antecubital vein for estimated changes in plasma volume due to heat stress from hemoglobin and hematocrit values (7).
Echocardiographic images were obtained using commercially available ultrasound equipment (LOGIQ7, GE Healthcare, Tokyo, Japan). All echocardiography examinations were performed with the subject in the left lateral decubitus position. Measurements were obtained during quiet guided expiration. Images were obtained by two experienced sonographers and stored on the ultrasound hard drive and were later exported for offline analysis by an experienced sonographer using commercially available software (GE Healthcare).
Tissue Doppler imaging.
Measurements of septal and lateral mitral annular early diastolic (E′), late diastolic (A′), and systolic (S′) velocities were obtained to provide indexes of left ventricular diastolic function, left atrial systolic function, and left ventricular systolic function, respectively, via standard tissue Doppler imaging techniques (17). Additionally, isovolumic acceleration, which is another index of left ventricular systolic function, was determined from the slope of the presystolic velocity curve at the septal and lateral mitral annulus. Tissue Doppler measurements were obtained from the apical four-chamber view with a 4.0-mm sample volume positioned at the junction of the septal mitral annulus and the left ventricular wall, as well as the junction of the lateral mitral annulus and the left ventricular wall. Although these parameters were measured from both the septal and lateral regions, averaged responses from these regions are reported.
Mitral inflow velocities.
Mitral inflow velocities were assessed from the apical four-chamber view using pulsed wave Doppler with a sample volume of 2.0 mm positioned over the mitral valve leaflet tips. Peak inflow velocities were obtained during the early phase of left ventricular relaxation (E) and during left atrial contraction (A).
Isovolumetric relaxation time.
Isovolumetric relaxation time (IVRT) represents the time interval between aortic outflow during systole and the opening of the mitral valve during diastole and is commonly used as an index of left ventricular relaxation. IVRT was determined using a five-chamber apical view with the sample volume set at 4.0 mm. Because of complications obtaining IVRT in one able-bodied subject, the data analyzed for this variable are representative of eight subjects for that group.
Respiratory and metabolic variables.
Standard respiratory and metabolic data were obtained by using an automatic breath-by-breath respiratory gas analyzing system (ARCO200-MET; Arcosystem, Chiba, Japan). Cardiac output was indexed via the carbon dioxide (CO2) rebreathing technique, with CO2 content in mixed venous blood estimated using the CO2 rebreathing equilibrium technique (9).
Following instrumentation, each subject rested quietly in the supine position for ∼30 min while normothermic water (33°C) circulated through both upper and lower body parts of the water-perfused suit. A blood sample was obtained at this time followed by baseline thermal and hemodynamic data. The aforementioned echocardiographic measures were then obtained, followed by rebreathing cardiac output measurements. Subjects were then exposed to a lower body heat stress by perfusing 50°C water through only the lower body part of the suit, whereas slightly warmed water (36°C) circulated through the upper body part of the suit until core temperature was elevated by ∼1°C. In pilot studies, we identified that in able-bodied and paraplegic individuals lower body passive heat stress slightly increased upper body skin temperature. Because of this, 36°C water was perfused in the upper body garment to minimize any differences in skin temperature between groups. Lower body heating was used as the primary heating source, resulting in an absence of sensed elevations in skin temperature in the spinal cord injured subjects. Once this target internal temperature was achieved, the temperature of the water circulating through the lower body suit was decreased to ∼47°C to limit further increases in core temperature while heat stress data were obtained.
P values <0.05 were considered statistically significant.
Thermoregulatory and hemodynamic variables.
Esophageal temperature was similar between groups before the heat stress (Table 2) and was increased to the same extent (i.e., 1.0 ± 0.1°C in AB, 1.0 ± 0.1°C in TSCI, and 1.1 ± 0.1°C in CSCI) to the heat stress. Lower body passive heating similarly increased mean lower body skin temperatures (average of thigh and lower leg skin temperatures) to 39.8 ± 0.7°C (AB), 39.7 ± 1.4°C (TSCI), and 39.2 ± 1.3°C (CSCI). The heat stress increased cutaneous vascular conductance at the chest in AB (0.29 ± 0.04 to 1.13 ± 0.36 au/mmHg) and TSCI (0.37 ± 0.19 to 0.99 ± 0.43 au/mmHg), but not in CSCI (0.37 ± 0.16 to 0.45 ± 0.25 au/mmHg), whereas the magnitude of this increase was greater in AB than in TSCI (Table 3). Likely due to a direct effect of local heating, thigh skin blood flow increased similarly between groups.
Before the heat stress cardiac output was lower in CSCI while heart rate, stroke volume and mean arterial pressure were not significantly different between groups (Table 2). Cardiac output and heart rate increased during heat stress in all groups, whereas stroke volume did not change. However, the magnitude of the increase in cardiac output was significantly greater in AB relative to in CSCI, but not relative to TSCI (Table 3). Moreover, the magnitude of the increase in heart rate was different between groups. Mean arterial pressure was slightly decreased by the heat stress in TSCI but did not change in the other groups. The relative reduction in plasma volume from the rest until just before measurement of cardiac echo was greater in AB than other groups (Table 3).
Indexes of cardiac function.
Before heat stress, indexes of ventricular systolic function [isovolumic acceleration (IVA); S′, mitral annular systolic velocity; Fig. 1], atrial systolic function (A, mitral inflow velocity during left atrial contraction; A′, late diastolic mitral annular velocity; Fig. 2), and diastolic function (IVRT; E, mitral inflow velocity during the early phase of left ventricular relaxation; E′, early diastolic mitral annular velocity; Table 4) were similar between groups. Also E/E′ and E/A ratios were similar between group (Table 4). During heat stress, indexes of diastolic function (E and E′) were similarly maintained between groups, whereas IVRT similarly decreased in all groups (AB, 14.3 ± 5.6 ms; TSCI, 15.8 ± 13.8 ms; CSCI, 13.2 ± 10.4 ms).
Heat stress increased indexes of ventricular systolic function (S′ and IVA; Fig. 1) in AB and TSCI, whereas neither changed in CSCI, suggesting that cardiac sympathetic innervation is required for enhanced systolic function in hyperthermia. Notably, the magnitude of the increase in S′ due to heat stress was significantly greater in AB relative to in TSCI, whereas the increase in IVA was similar between AB and TSCI. Indexes of atrial systolic function (A and A′; Fig. 2) also did not change in CSCI during heat stress, whereas both increased during heat stress in AB and only A increased in TSCI. However, the magnitude of the increase in A was greater in AB relative to both TSCI and CSCI.
The primary observations of this study are as follows: 1) indexes of diastolic function are generally maintained during heat stressed regardless of group, 2) indexes of ventricular systolic function increased due to heat stress in AB and TSCI but not in CSCI, and 3) indexes of atrial systolic function increased in AB but generally did not change in TSCI or CSCI. Regardless of these varied results, cardiac output increased during heat stress in all groups, although the magnitude of this increase was smaller in CSCI. Together these findings suggest that 1) in all groups heat stress increases cardiac output, the extent of which varied between groups, through a combination of increases in heart rate and maintained diastolic function; 2) systolic function was not affected by heat stress in CSCI, whereas this variable increased in both AB and TSCI, indicating that sympathetic innervation is required to increase systolic function during heat stress; 3) elevated systolic function during heat stress primarily occurs as a result of increases in internal temperature, although stimulation of skin thermoreceptors may contribute.
Cardiac responses in tetraplegia.
The first objective of this project was to evaluate whether sympathetic innervation to the heart is required for appropriate cardiac responses to heat stress. This objective was accomplished through comparing cardiac responses between TSCI and CSCI subjects, the latter of which lack sympathetic innervation to the heart, while both groups were heated from insensate areas. Despite this lack of innervation, cardiac output increased during heat stress in CSCI, although the magnitude was smaller than the other groups. The increase in cardiac output in CSCI likely occurred secondary to increases in heart rate, presumably via withdrawal of vagal tone, given the absence of increases in atrial or ventricular systolic function in this group to heat stress. Dela et al. (5) reported that cardiac output in tetraplegia can increase secondary to increases in heart rate and stroke volume during either active or passive exercise. The muscle pump during those exercises contributed to the increase in stroke volume, although such a pump would not be functional during passive heat stress. On the contrary, passive heat stress decreases venous return secondary to cutaneous vasodilation (2–4, 11), which likewise occurred even in CSCI due to cutaneous vasodilation associated with local heating of the legs. In all groups the early diastolic blood and tissue velocities were maintained during heat stress, whereas IVRT was reduced, with the latter variable being sensitive to changes in heart rate and systolic function (14). Given an absent sympathetic innervation to the heart in CSCI, the observed change in IVRT in this group was most likely caused by increases in heart rate secondary to reduced cardiac vagal activity. Finally, this group showed a marked absence of increases in indexes of atrial and ventricular systolic function to heat stress, consistent with cardiac denervation. When taken together, these observations, coupled with increases in these indexes in TSCI, suggest that direct heating of the heart (i.e., in the absence of cardiac sympathetic innervation) does not contribute to heat-induced increases in atrial or ventricular systolic function.
Contribution of cutaneous thermoreceptor afferents in mediating cardiac responses to heat stress.
The second objective of this study was to evaluate the influence of skin thermoreceptors in mediating cardiac responses to heat stress. This objective was accomplished by comparing cardiac responses between AB and TSCI subjects during lower body heating, recognizing that in the TSCI subjects the central nervous system would not receive afferent information indicative of elevated skin temperatures from insensate region. Heat stress increased indexes of atrial and ventricular systolic function in both groups; however, the magnitude of these responses was generally smaller in TSCI relative to AB. Despite these observations, there was no difference in heat stress-induced increases in cardiac output between these groups. Based upon these observations, skin thermoreceptors contribute to some extent in mediating increases in indexes of atrial and ventricular systolic function during heating, but these responses were not sufficient to significantly alter cardiac output.
Alternatively, the magnitude of the increase in chest skin blood flow was smaller in TSCI relative to AB. Given that skin blood flow is controlled by the integration of central and peripheral thermoreceptors, coupled with the increase in internal temperature being the same between these groups, the attenuated increase in chest skin blood flow in TSCI may be due to the absence of peripheral thermoception from the area primarily being heated (i.e., the legs). Moreover, it is noteworthy that the reduction in plasma volume to heating was smaller in TSCI than in AB. The most likely explanation for this observation is less sweating occurred in TSCI, likely due to an absence of sweating below the level of the injury. Given this observation, coupled with a proposed smaller reduction in central blood volume in TSCI due to less cutaneous vasodilation, the strain on the human cardiovascular system during the heat stress in TSCI subjects might not be the same as in AB subjects, resulting in smaller increment of cardiac systolic function, especially in atrial systolic function.
Perspectives and limitations of the study.
By the comparison of responses between the evaluated groups to whole-body heat stress, these findings provide insight into the controlling mechanisms governing cardiac responses during whole-body heat stress in AB individuals. Increases in indexes of cardiac systolic function are primarily driven by increases in core temperature during a hyperthermic challenge, while afferent information from the skin may contribute. Based upon the present observations, such cardiac changes do not occur secondary to changing temperature of cardiac tissue independent of sympathetic/parasympathetic innervation. However, these findings should be interpreted with the understanding that possible differences in central volume changes (i.e., cardiac preload) associated with the heat stress, as well as differences in heart rate responses, between groups may contribute to the reported findings independent of cardiac sympathetic innervation and afferent sensory information.
The cardiovascular system adapts to a myriad of stimuli through a combination of neural and nonneurally mediated responses. One such stimulus is whole-body heat stress, the physiological adaptations of which are particularly important given progressive increases in global surface temperatures. In this study we observed differential cardiac responses to heat stress between AB, TSCI, and CSCI subjects. These differential responses are proposed to be due to a combination of an absence of cardiac sympathetic innervation in the CSCI subjects and the absence of cutaneous thermal afferent input from heated region in both CSCI and TSCI. Thus cardiac sympathetic innervation likely contributes to changes in cardiac systolic function during heat stress. Second, elevated internal temperature is the primary driver for many cardiovascular responses to heat stress, whereas skin afferents likely have a less prominent role in mediating some of these responses.
This study was funded by the Research Foundation of NachiKatsuura City, the Nakatomi Foundation, and National Heart, Lung, Blood Institute Grant HL-61388.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: M.S., C.G.C., and F.T. conception and design of research; M.S., Y.U., T.K., K.K., T.I., T.N., C.G.C., and F.T. performed experiments; M.S., Y.U., T.K., K.K., T.I., T.N., and C.G.C. analyzed data; M.S., Y.U., T.K., K.K., T.I., T.N., and C.G.C. interpreted results of experiments; M.S., T.K., and K.K. prepared figures; M.S., Y.U., C.G.C., and F.T. drafted manuscript; M.S., C.G.C., and F.T. edited and revised manuscript; M.S., Y.U., T.K., K.K., T.I., T.N., C.G.C., and F.T. approved final version of manuscript.
We appreciate the time and effort expended by the volunteer subjects. We also thank the staff of the Department of Rehabilitation and Institute of Sports Science and Environmental Physiology, Wakayama Medical University.
- Copyright © 2015 the American Physiological Society