Share this article on:

Upright Posture Reduces Thermogenesis and Augments Core Hypothermia

Nakajima, Yasufumi MD, PhD*,; Takamata, Akira PhD†,; Ito, Tomoyuki MD†,; Sessler, Daniel I. MD‡,; Kitamura, Yoshihiro MD, PhD*,; Shimosato, Goshun MD*,; Taniguchi, Satoshi MD*,; Matsuyama, Hiroki MD*,; Tanaka, Yoshifumi MD, PhD*, and; Mizobe, Toshiki MD, PhD*

doi: 10.1213/00000539-200206000-00053

We recently reported that baroreceptor-mediated reflexes modulate thermoregulatory vasoconstriction during lower abdominal surgery. Accordingly, we examined the hypothesis that postural differences and the related alterations in baroreceptor loading similarly modulate the thermogenic (i.e., shivering) response to hypothermia in humans. In healthy humans (n = 7), cold saline was infused IV (30 mL/kg at 4°C) for 30 min to decrease core temperature. Each participant was studied on 2 separate days, once lying supine and once sitting upright. Tympanic membrane temperature and oxygen consumption were monitored for 40 min after each saline infusion. The decrease in core temperature upon completion of the infusion in the upright posture position was 1.24°C ± 0.07°C, which was significantly greater than the 1.02°C ± 0.06°C seen in the supine position. The core temperature was reduced by 0.59°C ± 0.07°C in the upright position but only by 0.37°C ± 0.05°C in the supine position when the increase in oxygen consumption signaling thermogenic shivering occurred. Thus, the threshold temperature for thermogenesis was significantly less in the upright than the supine position. The gain of the thermogenic response did not differ significantly between the positions (363 ± 69 mL · min−1 · °C−1 for upright and 480 ± 80 mL · min−1 · °C−1 for supine). The skin temperature gradient was significantly larger in the upright than in the supine posture , suggesting that the peripheral vasoconstriction was augmented by upright posture. Plasma norepinephrine concentrations increased in response to cold saline infusion under both conditions, but the increase was significantly larger in the upright than in the supine posture. Baroreceptor unloading thus augments the peripheral vasoconstrictor and catecholamine response to core hypothermia but simultaneously reduces thermogenesis, which consequently aggravated the core temperature decrease in the upright posture.

*Departments of Anesthesiology and †Physiology, Kyoto Prefectural University of Medicine, Kyoto, Japan; and ‡University of Louisville, KY, and Ludwig Boltzmann Institute, University of Vienna, Austria

None of the authors has a financial relationship with any company related to this research.

Supported, in part, by the Japanese Ministry of Education, Science, Culture, and Sports, NIH Grant GM 58273 (Bethesda, MD), the Joseph Drown Foundation (Los Angeles, CA), and the Commonwealth of Kentucky Research Challenge Trust Fund (Louisville, KY).

January 28, 2002.

Address correspondence and reprint requests to Yasufumi Nakajima, MD, PhD, Department of Anesthesiology, Kyoto Prefectural University of Medicine, Kawaramachi Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan. Address e-mail to

Thermoregulatory responses are influenced by nonthermal factors including signals from the cardiovascular system (1–3). For example, several studies suggest that the cutaneous circulation in humans during heat stress is subservient not only to thermoregulatory responses, but also to baroreflexes (4–6). These reports conclude that baroreceptor unloading during heat stress inhibits active thermoregulatory cutaneous vasodilation. The sweating response during exercise is also attenuated by baroreceptor unloading, as is active vasodilation. Both effects are rapidly reversed by the release of baroreceptor unloading (6). Baroreceptor-mediated inhibition of thermoregulatory reflexes during heat stress thus influences both active vasodilation and sweating. These results suggest that the interaction between the baroreceptor and thermoregulatory systems occurs proximal to the effector organs and presumably within the central nervous system.

We reported that baroreceptor loading/unloading altered the decrease in core body temperature by modulating peripheral vasoconstriction during lower abdominal surgery in anesthetized patients. However, the metabolic response could not be evaluated in that study because muscle relaxants given to facilitate surgery prevented thermoregulatory shivering (7). Thus, it remains unclear whether the augmentation of thermoregulatory defenses against hypothermia induced by baroreceptor unloading is specific for peripheral vasoconstriction or if it is a more general thermoregulatory response that includes thermogenesis.

Blood pressure manipulation and sympathetic nervous system activation reportedly modulate thermoregulatory responses in animals. Hohtola et al. (8) reported that both increases and decreases in blood pressure inhibit shivering in pigeons. Roth et al. (9) reported that increased peripheral sympathetic nervous activity, whether induced directly or indirectly, inhibits noradrenergic neurons in the lower brainstem and reduces the hypothalamic thermoregulatory set point. However, little information is available about the effect of baroreflex control on the shivering response in humans. We therefore evaluated the effect of postural differences, which affect baroreceptor activity, on the vasomotor, hormonal, and thermogenic response to core hypothermia in unanesthetized humans. Specifically, we tested the hypothesis that baroreceptor unloading augments both the vasomotor and thermogenic responses to core hypothermia.

Back to Top | Article Outline


This study was approved by the Review Board on Human Experiments, Kyoto Prefectural University of Medicine, Japan. Written informed consent was obtained from seven healthy men. Their average age was 29 ± 2 yr (mean ± sem), body weight was 67 ± 1 kg, and height was 174 ± 2 cm. Each volunteer participated on 2 different study days separated by at least 6 days. To avoid circadian variations, the studies were scheduled at the same time each day. Each of the subjects wore a pair of thin cotton trunks during the studies.

Participants refrained from salty food and beverages containing caffeine or alcohol the night before the experiments. They were instructed to eat a light breakfast and drink water if they desired. The volunteers reported to the laboratory at 9:00 am. After voiding, they entered an environmental chamber maintained at 23°C with a relative humidity near 30%.

After injecting a local anesthetic, a 16-gauge 30-cm-long catheter (ES-04400, Arrow Japan, Ltd, Tokyo, Japan) was inserted into the cephalic or other basilic vein; this catheter permitted infusion of cold fluid without pain or a cold sensation. An 18-gauge IV catheter was inserted into an antecubital vein in the contralateral (noninfused) arm. On the first study day, volunteers were randomly assigned to either lie in the supine position on a bed or to sit upright against a backrest. On the second study day, the alternate position was used. The designated position was maintained throughout each experiment. After a 15-min-long rest period, IV fluid at 4°C was infused at a rate of 60 mL · kg−1 · h−1 for 30 min.

Core temperature was measured at the tympanic membrane using Mon-a-Therm® thermocouples (Mallinckrodt Anesthesiology Product, Inc, St Louis, MO). The aural probes were inserted by the volunteers until they felt the thermocouple touch the tympanic membrane; appropriate placement was confirmed when volunteers easily detected a gentle rubbing of the attached wire. The aural canal was occluded with cotton and taped in place.

Skin temperature was measured at the forehead, chest, back, forearm, thigh, and calf. Forearm skin temperature was measured on the noninfused arm. Mean skin temperature was calculated from the body surface area distribution and thermal sensitivity of each skin area using the following equation (10):MATHwhere Tsk is mean skin temperature, Tfh is forehead, Tch is chest, Tab is abdomen, Tua is upper arm, Tfa is forearm, Tth is thigh, and Tca is calf skin temperatures.



Heart rate (HR) was monitored from an electrocardiogram, and blood pressure was measured at 5-min intervals in the noninfused arm (model STBT-780, Colin, Nagoya, Japan). Mean arterial blood pressure (MAP) was calculated as (SBP − DBP)/3 + DPB, where SBP is systolic blood pressure and DBP is diastolic blood pressure.

Nonshivering thermogenesis contributes minimally in adult humans (11,12). Consequently, thermoregulatory increases in metabolic rate result almost completely from shivering. Therefore, oxygen consumption (V̇o2) is an accurate measure of shivering thermogenesis and was determined and recorded at 1-min intervals with a flow/gas analyzer (Aeromonitor AE260, Minato, Tokyo, Japan).

Blood samples were collected at 4 time points: 0 min (before the infusion), 15 min (midinfusion), 30 min (at the end of the infusion), and 40 min (at the end of the experiment). The samples collected at 0 min, 30 min, and 40 min were obtained from the central venous catheter, whereas the sample obtained at 15 min was collected from the IV catheter inserted into the noninfused arm.

Hematocrits (Hct) were determined immediately; changes in plasma volume were calculated from Hct according to the following equation (13):MATHwhere the subscript A indicates before the infusion (control) and B indicates after the treatment (experimental).



Aliquots for the measurement of catecholamine concentrations were centrifuged at 4°C, and the plasma was stored at −80°C until assayed. Catecholamine concentrations were determined by high-performance liquid chromatography (HLC-725CA, Tosoh, Tokyo, Japan) with an electrochemical detector after alumina extraction. The detection limit of the assay was 8 pg/mL, and the intra- and interassay coefficients of variation were <5%. Plasma renin activity was determined using an angiotensin I radioimmunoassay kit (Renin RIABEAD, Dainabot, Japan), where the level of detection was 0.1 ng · mL−1 · h−1, and the intra- and interassay coefficients of variation were 4.5% and 5.5%, respectively.

We defined the core-temperature threshold for the thermogenic response as the change in core temperature triggering a rapid increase in V̇o2. The threshold was individually determined for each posture in each subject by the same inspector who was blinded to the experimental conditions. Thermal responsiveness (gain) for thermogenesis was defined by the relationship between V̇o2 and core temperature after the threshold. The gain for each subject was individually calculated using least squares linear regression. Skin temperature gradient was considered an index of peripheral vasoconstriction (14).

All morphometric data, thermal responsiveness, and thresholds were analyzed by two-tailed paired t-tests. The effects of postural differences and time on the cardiovascular, thermogenic, and hormonal responses were analyzed by a general linear regression model for analysis of variance with repeated measures (two within and zero between factors), which was followed by a multiple comparison test with Fisher’s least significant difference test. All data are reported as mean ± sem;P < 0.05 was considered statistically significant.

Back to Top | Article Outline


MAP, HR, V̇o2, minute expired ventilation, respiratory rate, core temperature, mean skin temperature, and Hct were similar at baseline for each posture (Table 1). At the completion of the cold saline infusion, plasma volume was increased by 9.1% ± 0.6% when the volunteers were supine and by 9.0% ± 0.9% when they were upright (P = not significant). By the end of the studies (40 min), the plasma volume expansion was 6.0% ± 1.2% in the supine posture and 6.2% ± 0.9% when the volunteers were upright (P = not significant).

Table 1

Table 1

MAP increased to a peak value of 107 ± 7 mm Hg with each treatment during the cold saline infusion with no significant difference between the two postures. HR increased significantly more with the supine posture (86 ± 8 bpm) than with the upright posture (76 ± 5 bpm;P < 0.01). Respiratory rates increased significantly to 22.7 ± 2.7 breaths/min in the supine posture and 23.8 ± 3.1 breaths/min in the upright posture; however, these were not statistically different (Fig. 1).

Figure 1

Figure 1

Core cooling decreased core temperature with each posture. However, the decrease was larger in the upright than the supine posture (−1.24°C ± 0.07°C versus −1.02°C ± 0.06°C;P < 0.01;Fig. 2). Core cooling similarly decreased mean skin temperature during each treatment. However, the decrease was again larger when the volunteers were upright (−0.63°C ± 0.08°C versus −0.29°C ± 0.10°C;P < 0.01). Statistically significant differences in the decrease of skin temperature at the end of the experiment were observed between the upright versus the supine posture, especially for the forehead (−0.86°C ± 0.30°C versus −0.62°C ± 0.16°C), forearm (−1.00°C ± 0.16°C versus −0.04°C ± 0.36°C), thigh (−0.55°C ± 0.17°C versus −0.11°C ± 0.15°C), and calf (−1.04°C ± 0.22°C versus −0.61°C ± 0.17°C). The skin temperature gradient between the chest and forearm—an index of peripheral vasoconstriction—was thus larger with the upright than the supine posture (P < 0.05;Fig. 2). The thermogenic response was significantly reduced in the upright position (P < 0.05;Fig. 2).

Figure 2

Figure 2

The threshold temperature for thermogenesis was significantly lower with the upright than the supine posture. Shivering occurred at 0.59°C ± 0.07°C below control for the upright posture versus 0.37°C ± 0.05°C for the supine posture (P < 0.01). The gain of the thermogenic response was 363 ± 69 mL · min−1 · °C−1 when the volunteers were upright and 480 ± 80 mL · min−1 · °C−1 when they were supine; these values did not differ significantly (Fig. 3).

Figure 3

Figure 3

We were unable to obtain blood samples in three subjects at 15 min because they were so intensely vasoconstricted. Average values at this time point were computed from the remaining volunteers. Plasma norepinephrine concentrations at baseline were 0.15 ± 0.02 ng/mL in the supine posture and 0.26 ± 0.02 ng/mL with the upright posture (P = not significant). Although plasma norepinephrine concentrations were increased because of the cold saline infusion during each treatment, the increase was significantly larger with the upright than the supine posture (P < 0.05). Plasma epinephrine concentrations for each posture were similar at baseline and did not change during central core cooling. Significant differences in plasma renin activity were observed before the saline infusion. The values then converged during the fluid administration but showed significant differences between the groups (P < 0.05;Fig. 4).

Figure 4

Figure 4

Back to Top | Article Outline


Thermoregulation is controlled by multiple feedback systems, such as baroreflexes and osmoregulation, during heat stress and exercise; the final goal of this integration seems to be the maintenance of arterial blood pressure (1–6,15,16). A previous report, for example, indicates that baroreceptor unloading downregulates both sweating and active vasodilation. This observation led us to speculate that signals from baroreceptors interact with the thermoregulatory center directly or at least well proximal to the thermoregulatory effectors (6). To the extent that this theory applies to cold defenses, one might expect that the upright posture would augment both peripheral vasoconstriction and shivering. Although, our main finding was that the thermogenic response to hypothermia was attenuated by the upright posture, whereas peripheral vasoconstriction was augmented. This indicates that cardiovascular regulation affects the thermogenic and the vasoconstrictor efferents distal to the common thermoregulatory center and presumably at different sites.

Our current results contrast with those we reported in which baroreceptor unloading attenuated perioperative hypothermia by enhancing thermoregulatory peripheral vasoconstriction in anesthetized patients (7), although the critical distinction is that our current volunteers were unanesthetized and permitted to shiver. In contrast, our previous results were obtained in surgical patients who were paralyzed and anesthetized. Thus, the effects of thermoregulatory vasoconstriction on core temperature dominated in the surgical patients, whereas thermogenesis proved to be most important in the conscious states. The effects of posture on core temperature thus depend critically on the circumstances of the study.

The decreased threshold for thermogenesis and the smaller increase in V̇o2 as a function of time in the upright position were associated with a smaller increase in HR. Interestingly, plasma norepinephrine concentration was larger in the upright position, indicating that the overall sympathetic nervous activity was increased in the upright position. Thus, attenuation of the HR response during body core cooling in the upright position might result from a reduced cardiac response to sympathetic activation; that is, a smaller increase in sympathetic outflow to the heart compared with the overall increase may be because of regional heterogeneity of sympathetic outflow or to withdrawal of vagal influence (3,17,18). However, it seems more plausible that HR was simply controlled to match metabolic demand. HR is highly and linearly correlated with V̇o2 during exercise (3) and may be regulated during body core cooling by the mechanism similar to that during exercise.

Mean skin-surface temperature contributes 20% to thermoregulatory control during body core cooling, with the remaining 80% being provided by core temperature (19). The change in mean skin temperature was significantly larger by 0.34°C in the upright than in the supine posture. This most likely indicates that the downward shift in the threshold for thermogenesis (in terms of core temperature) was actually underestimated by 0.07°C in the upright posture.

The thermogenic response to body core cooling was attenuated during the upright position, and the increase in plasma norepinephrine during the cold saline infusion was larger in the upright position. The greater norepinephrine response, or larger sympathetic nerve activity, during the body core cooling might be involved in the mechanism of the attenuated thermogenic response in the upright position. For example, shivering is inhibited in proportion to exercise intensity, a stimulus well known to increase the sympathetic nervous activity and plasma norepinephrine (11). In another study, the thermoregulatory cold defense response was shifted to a lower mean body temperature in warm-adapted animals after IM injection of norepinephrine in amounts equal to those released during cold adaptation. The authors concluded that high peripheral sympathetic nervous system activity, or an increase in systemic norepinephrine, directly or indirectly inhibits noradrenergic neurons in the lower brainstem. Thus, reduced adrenergic input from the brainstem to the hypothalamus in turn decreases the hypothalamic thresholds triggering cold defenses. Taken together, it is possible that peripherally-released norepinephrine, or increased sympathetic nervous activity, impairs centrally-induced cold defense or thermogenesis in humans.

We adopted the cold saline infusion method to investigate the effects of posture on the thermogenic response to core cooling. The increase in MAP during body core cooling was similar between the positions, suggesting that the level of arterial baroreceptor loading induced by the infusion was similar in each position. Although we did not measure central venous pressure, the hormonal data suggest that the level of baroreceptor loading (presumably cardiopulmonary baroreceptors) differed with each posture. Furthermore, increased norepinephrine concentration and plasma renin activity in the upright posture during the preinfusion period was maintained during the fluid administration. We therefore conclude that differences in baroreceptor activation are the primary explanation for the observed differences in thermoregulatory control.

We reported the baroreceptor modulation of peripheral vasoconstriction during lower abdominal surgery (7). However, we could not address the thermogenic response because of the use of muscle relaxants during surgery. Therefore, this basic study is instructive for managing thermocare, especially for postoperative patients who need to be carefully controlled in fluid balance or need to change posture.

In summary, we found that baroreceptor unloading augments the peripheral vasoconstrictor and catecholamine response to core hypothermia but simultaneously reduces the thermogenic response. These results suggest that the thermoregulatory and baroreceptor systems interact at locations distal to the common thermoregulatory control center.

The authors are grateful to Professor Hiroshi Nose, Department of Sports Medicine, Shinshu University School of Medicine, Matsumoto, Japan, for his help and advice in the present study. Mallinckrodt Anesthesiology Products, Inc (St Louis, MO) donated the thermocouples we used.

Back to Top | Article Outline


1. Johnson JM. Nonthermoregulatory control of human skin blood flow. J Appl Physiol 1986; 61: 1613–22.
2. Nadel ER, Fortney SM, Wenger CB. Effect of hydration state of circulatory and thermal regulations. J Appl Physiol 1980; 49: 715–21.
3. Rowell LB. Human circulation: regulation during physical stress. New York: Oxford University Press, 1986.
4. Crandall CG, Johnson JM, Kosiba WA, Kellogg DL Jr. Baroreceptor control of the cutaneous active vasodilator system. J Appl Physiol 1996; 81: 2192–8.
5. Kellogg DL Jr, Johnson JM, Kosiba WA. Baroreflex control of the cutaneous active vasodilator system in humans. Circ Res 1990; 66: 1420–6.
6. Mack G, Nishiyasu T, Shi X. Baroreceptor modulation of cutaneous vasodilator and sudomotor responses to thermal stress in humans. J Physiol (Lond) 1995;483:537–47.
7. Nakajima Y, Mizobe T, Takamata A, Tanaka Y. Baroreflex modulation of peripheral vasoconstriction during progressive hypothermia in anesthetized humans. Am J Physiol 2000; 279: R1430–6.
8. Hohtola E, Saarela S, Hissa R. Effect of blood pressure manipulations on shivering thermogenesis in the pigeon. Acta Physiol Scand 1980; 110: 277–83.
9. Roth J, Zeisberger E, Schwandt HJ. Influence of increased catecholamine levels in blood plasma during cold-adaptation and intramuscular infusion on thresholds of thermoregulatory reactions in guinea-pigs. J Comp Physiol [B] 1988;157:855–63.
10. Nadel ER, Mitchell JW, Stolwijk JA. Differential thermal sensitivity in the human skin. Pflügers Arch 1973;340:71–6.
11. Hong SI, Nadel ER. Thermogenic control during exercise in a cold environment. J Appl Physiol 1979; 47: 1084–9.
12. Ozaki M, Kurz A, Sessler DI, et al. Thermoregulatory thresholds during epidural and spinal anesthesia. Anesthesiology 1994; 81: 282–8.
13. Van Beaumont W. Evaluation of hemoconcentration from hematocrit measurements. J Appl Physiol 1972; 32: 712–3.
14. Rubinstein EH, Sessler DI. Skin-surface temperature gradients correlate with fingertip blood flow in humans. Anesthesiology 1990; 73: 541–5.
15. Kenney WL, Tankersley CG, Newswanger DL, Puhl SM. α1-Adrenergic blockade does not alter control of skin blood flow during exercise. Am J Physiol 1991; 260: H855–61.
16. Nakajima Y, Nose H, Takamata A. Plasma hyperosmolality and arterial pressure regulation during heating in dehydrated and awake rats. Am J Physiol 1998; 275: R1703–11.
17. Abboud FM, Heistad DD, Mark AL, Schmid PG. Reflex control of the peripheral circulation. Prog Cardiovasc Dis 1976; 18: 371–403.
18. Kaye DM, Jennings GL, Dart AM, Esler MD. Differential effect of acute baroreceptor unloading on cardiac and systemic tone in congestive heart failure. J Am Coll Cardiol 1998; 31: 583–7.
19. Cheng C, Matsukawa T, Sessler DI, et al. Increasing mean skin temperature linearly reduces the core-temperature thresholds for vasoconstriction and shivering in humans. Anesthesiology 1995; 82: 1160–8.
© 2002 International Anesthesia Research Society