autonomic nervous system fun”>American Journal of Emergency Medicine (2013) 31, 375-380

Original Contribution

The role of autonomic nervous system function in hypothermia-mediated sepsis protection

Yun-Te Chang MD, MPH ^a,b,d, Shue-Ren Wann MD ^a,?, Jung-Shun Tsai PhD ^b, Chih-Hsiang Kao MD ^a, Po-Tsang Lee MD, PhD ^c, Neng-Chyan Huang MD ^a,

Cheng-Chang Yen MD ^a, Mu-Shun Huang MD ^d, Hong-Tai Chang MD ^a

^aDepartment of Emergency Medicine, Kaohsiung Veterans General Hospital, Kaohsiung City 81362, Taiwan, ROC

^bYuh-Ing Junior College of Health Care and Management, Kaohsiung City 80776, Taiwan, ROC

^cDepartment of Internal Medicine, Kaohsiung Veterans General Hospital, Kaohsiung City 81362, Taiwan, ROC

^dSchool of Medicine, National Yang-Ming University, Taipei City 11221, Taiwan, ROC

Received 14 August 2012; accepted 21 August 2012

Abstract

Objective: The objective of this study is to determine whether hypothermia will lessen decreases in Heart rate variability and improve outcome in a rat model of sepsis.

Methods: Thirty-six male Sprague-Dawley rats were randomized into 3 groups: control, low sepsis, and high sepsis groups. These groups were each subdivided into a normothermia (37?C) (n = 6) and a hypothermia group (34?C) (n = 6). Cyclophosphamide (100 mg/kg) was administered 5 days before Staphylococcus aureus injection to produce conditions in which sepsis could be induced reliably. Hypothermic rats received temperature reduction for 1 hour post injection. Electrocardiogram was recorded before, after, and 1 day after staphylococcal injection, and the low frequency, high frequency (HF), and LF/HF ratio measurements of heart rate variability and the frequencies of arrhythmia were recorded. The effects of time, sepsis severity, and hypothermia on these variables were analyzed using a multivariate generalized estimation equation mode.

Results: Four deaths occurred in the normothermic group, and none, in the hypothermic group. Sepsis of both low and high severity increased low frequency and HF 1 day after sepsis induction. Hypothermia significantly decreased HF in low, but not high sepsis severity.

Conclusions: Hypothermia decreased mortality in septic rats. The influence of hypothermia on HF depended on the severity of the sepsis.

(C) 2013

Introduction

Sepsis is the leading cause of death in the hospital, with a mortality rate of 30% to 50% [1,2] and approximately 4.3%

* Corresponding author. Tel.: +886 7 346 8342; fax: +886 7 3468343.

E-mail address: [email protected] (S.-R. Wann).

in the emergency department (ED) [3]. Usually seen initially in the ED, it is characterized by cardiovascular and Multiple organ dysfunction, and rapid, appropriate therapy will improve the outcome and prognosis.

heart rate variability analysis is a Noninvasive tool to evaluate autonomic nervous modulation of the cardiovascular system [4,5]. In sepsis, a decrease in HRV is associated with increased mortality risk [6,7]. A

0735-6757/$ - see front matter (C) 2013 http://dx.doi.org/10.1016/j.ajem.2012.08.028

significant reduction in the low-frequency (LF) compo- nent of HRV occurs in septic patients, and this reduction has been linked to a decrease in sympathetic nervous system activity [8,9].

One of the most common bacterial infections in sepsis is infection with gram-negative aerobic rods. Septic shock caused by Gram-negative organisms has been associated with autonomic cardiovascular dysfunction [10,11]. Endotoxin, a lipopolysaccharide released by a constituent of the gram- negative bacterial cell wall, causes the release of inflammatory cytokines from monocytes [12,13]. Endotoxin and cytokines are thought to contribute to the myocardial depression seen in sepsis [14].

Hypothermia has been reported to have an organ- protective effect in various pathologic situations [15,16] and may delay the induction of Proinflammatory cytokines [17,18]. However, the mechanisms causing depressed HRV and hypercytokinemia in sepsis are complex and not well understood, and their clinical significance has not yet been determined and needs further study.

We, therefore, investigated the effect of hypothermia on HRV during mild and severe sepsis in a rat sepsis model. Because some studies have questioned the clinical relevance of models using sepsis induced by endotoxin injection [19,20], we used subcutaneous Staphylococcus aureus injection to induce sepsis. We proposed to document a Rodent model of sepsis, to use 2 different amounts of injected S aureus (10⁹ and 10¹⁰ colony-forming units/mL) to induce sepsis, to record the effects of hypothermia on HRV during sepsis, and to clarify the effects of poor outcome and sepsis severity. The rat model in use was modified from an established model for investigation of sepsis [21]. Our hypothesis predicts that application of hypothermia will decrease the extent of septic damage and lessen the sepsis-induced decrease in HRV in this rat model.

Materials and methods

Animal preparation

Thirty-six male breeder Sprague-Dawley rats, aged 6 to 8 months, weighing 450 to 550 g, were provided by the BioLASCO Taiwan Co, Ltd (Taipei, Taiwan). The Institutional Animal Care and Use Committee of Kaoshiung Veterans General Hospital approved the experimental protocol, and the care and the handling of the animals followed the health guidelines.

Preparation of bacteria

S aureus, which had been isolated from a patient’s blood during sepsis was transferred to tubes containing 1.0 mL of Brucella broth with 10% glycerol and stored at -70?C until use. Twenty-four hours before injection, the stored bacteria samples were thawed at Room temperature for 1 hour and then incubated for 5 hours at 37?C. The broth was subsequently re-streaked

onto media plates of blood agar and incubated overnight at 37?C. Approximately 2 hours before injection, the bacteria were harvested from the agar and suspended in sterile saline. The suspension was washed 3 times and centrifuged at 3500 rpm for 20 minutes. The resulting bacterial pellet was resuspended in sterile saline and diluted to achieve an optical density comparable with McFarland Turbidity Standard No. 1. The original yield was approximate 10¹¹ organisms per milliliter. The viability of the organisms was verified by plating on blood agar after serial 10-fold dilutions and incubation for 18 hours at 37?C before colony counting.

Experimental protocols and measurements

Preliminary feasibility studies established that intraperi- toneal injection of cyclophosphamide (100 mg/kg) produced no gross injury to internal organs. Injected animals demonstrated decreased food intake and weight loss that was reversed by the fifth day, at which time the white blood cell count had decreased to approximately 1000/uL, thus creating a period of immunocompromised status during which sepsis could be induced by bacterial injection.

Thirty-six SD rats weighing 450 to 550 g were given intraperitoneal injections of cyclophosphamide (100 mg/ kg) 5 days before bacterial or saline injections. Eighteen of these rats were randomly assigned to be controlled at core temperature 37?C and were further randomly divided into 3 subgroups (control, low sepsis, and high sepsis, with 6 rats in each group). Another 18 were randomly assigned to be controlled at core temperature 34?C and were also randomly divided into 3 subgroups (control, low sepsis, and high sepsis; with 6 animals in each group). Twelve rats (mild sepsis group) received subcutaneous injections of 0.3-mL saline containing 10⁹ colony-forming units, 12 (severe sepsis group) received similar injections contain- ing 10¹⁰ colony-forming units/mL of S aureus, and 12 animals (control group) received an equivalent volume of sterile saline.

At the start of the experiment, all animals were anesthetized by Intramuscular injection of 20 to 40 mg/kg Zoletil 50 (tiletamine and zolazepam, 1:1 ratio) under monitoring. Subcutaneous injections of 0.3 mL of sterile saline or saline containing either 10⁹ or 10¹⁰ colony-forming units of S aureus were administered at a site close to the midline of the back, approximately 3.5 cm proximal to the coccyx. After bacterial inoculation, body temperature was lowered in the hypothermic group with ice surrounding the body of the rat. After achieving target temperature, animals were continuously monitored for 1 hour before natural recovery from anesthesia. On the next day (the first day after bacterial injection), animals were again anesthetized for 30- minute monitoring at 24 hours, followed by natural recovery. Survival for 72 hours (the third day) was recorded.

A 2-lead electrocardiogram (ECG) was used for contin- uously monitoring and recording during the procedure.

Control	0.0% (0/6)	.037	0.0% (0/6)	NA
Low	16.7% (1/6)		0.0% (0/6)
High	50.0% (3/6)		0.0% (0/6)

Animals breathed spontaneously in room air and, after the ECG recording was completed, were moved into a cage with a heating lamp until wake-up. Body temperature was

Table 1 Effect of sepsis severity on mortality rate in normothermic and hypothermic rats

Sepsis severity level

Normothermia group (37?C)

Mortality rate

P a

Hypothermia group (34?C)

Mortality rate

P a

Abbreviations: NA, not applicable because the mortality rate for all sepsis severity levels was zero. Deaths occurred 36, 51, 69, and 70 hours after bacterial inoculation.

^a Cochran-Armitage trend test.

measured with a thermocouple microprobe (#9030-12-34; Columbus Instruments, Columbus, OH) that was advanced through the right femoral artery into the thoracic aorta and was controlled to the predetermined temperature +-0.5? using a heating lamp or ice bath. The following parameters were recorded: respiratory frequency, heart rhythm, HRV, and changes in ECG. The ECG in each animal was recorded continuously for 1 hour at each of the following time points: before bacteria inoculation, after bacterial inoculation (0-40 minutes), 1 day after bacterial inoculation (24 hours). Chart Modules software (ADInstruments Pty Ltd, New South Wales, Australia) was used for ECG and HRV analysis on-line and off-line. At the end of the day 3 after bacteria/saline injection, all animals were euthanized with an intraperitoneal injection of pentobarbital sodium (150 mg/kg) and were examined

at autopsy.

Fig. Plots the time course of GEE-fitted means with SEs for LF, HF, and LF/HF for 3 sepsis severity levels in normothermia and hypothermia.

Statistical analysis

Three factors were designated independent variables: temperature (hypothermia, normothermia), sepsis severity (none, low, high), and time (before bacterial or saline injection, immediately after, and day 1 after injection). The effects of these 3 factors on the dependent variables of interest were investigated using a multivariate generalized estimation equation (GEE) model to accommodate repeated measurements of rats across different time points. Interac- tions between sepsis severity level and temperature, temperature and time course, and sepsis severity level and time course were also investigated. For continuous depen- dent variables, such LF, high frequency (HF), and LF/HF ratio, the GEE model with identity link function was used to provide adjusted mean difference and 95% confidence interval (CI) compared with the appropriate reference group. For binary dependent variables, such arrhythmia, the GEE model with logit link function was used to provide an adjusted odds ratio and 95% CI, compared with the appropriate reference group. The mortality trend across sepsis severity level was examined by the Cochran-Armitage trend test. Statistical analyses were performed with SAS

Table 2 Multivariate GEE model for LF, HF, and LF/HF ratio

software version 9.2 (SAS Institute Inc, Cary, NC), and 2- tailed P b .05 indicated statistical significance.

Results

Mortality

A total of 36 rats were used in the analysis. Four rats died in the normothermia group, and a significant trend related to severity of sepsis was seen in this group (no sepsis 0%, low sepsis severity 16.7%, high sepsis severity 50.0%, P = .037, Table 1). All rats that died were examined by autopsy to confirm sepsis-related death. During the autopsy, we searched for evidence of injuries that would indicate that the death was due to complica- tions from the surgical procedures or was due to infection. Visual inspection showed that rats dying from infection presented with serious abscesses at the bacterial inocula- tion site in the dorsal muscles as well as necrotic changes in multiple organs including heart, lungs, intestines, and kidneys. No rat in the hypothermia group died, regardless of sepsis level.

	LF			HF			LF/HF
	? (95% CI)	P		? (95% CI)	P		? (95% CI)	P

Effect of temperature
Normothermia (37 ?C)	Reference group	-	Reference group	-	Reference group	-
Hypothermia (34 ?C)	4.59 (-2.26 to 11.44)	.189	2.35 (0.44-4.26)	.016 ?	-0.01 (-1.94 to 1.91)	.988
Effect of sepsis
Control	Reference group	-	Reference group	-	Reference group	-
Low	2.03 (-6.28 to 10.34)	.632	1.73 (-0.53 to 4.00)	0.134	-0.88 (-2.54 to 0.78)	.300
High	-4.98 (-11.91 to 1.95)	.159	1.89 (-0.78 to 4.55)	.165	-1.81 (-3.51 to -0.11)	.037 ?
Interaction of temperature and sepsis
Low vs control, normothermia	Reference group	-	Reference group	-	Reference group	-
Low vs control, hypothermia	-4.79 (-12.67 to 3.09)	.234	-3.66 (-6.14 to -1.17)	.004 ?	0.89 (-0.84 to 2.61)	.313
High vs control, normothermia	Reference group	-	Reference group	-	Reference group	-
High vs control, hypothermia	3.04 (-7.37 to 13.44)	.567	-2.61 (-5.85 to 0.63)	.114	1.11 (-0.63 to 2.85)	.212
Effect of time
Before	Reference group	-	Reference group	-	Reference group	-
After	0.27 (-5.72 to 6.26)	.929	-0.21 (-1.38 to 0.96)	.722	0.18 (-1.19 to 1.55)	.796
Day 1	-2.43 (-8.82 to 3.95)	.455	-0.55 (-2.00 to 0.90)	.459	-0.49 (-1.57 to 0.60)	.381
Interaction of temperature and time
Hypothermia vs normothermia, before	Reference group	-	Reference group	-	Reference group	-
Hypothermia vs normothermia, after	-1.67 (-8.02 to 4.68)	.606	-0.61 (-2.01 to 0.78)	.389	-0.43 (-1.67 to 0.81)	.498
Hypothermia vs normothermia, day 1	0.02 (-6.74 to 6.77)	.996	-1.26 (-3.08 to 0.57)	.177	0.51 (-0.54 to 1.57)	.340
Interaction of sepsis and time
Low vs control, before	Reference group	-	Reference group	-	Reference group	-
Low vs control, after	-0.67 (-9.84 to 8.49)	.886	0.59 (-1.34 to 2.52)	.549	-0.02 (-1.73 to 1.69)	.981
Low vs control, day 1	10.24 (1.60-18.88)	.020 ?	4.37 (2.47-6.27)	b.0001 ?	0.34 (-1.15 to 1.82)	.655
High vs control, before	Reference group	-	Reference group	-	Reference group	-
High vs control, after	0.86 (-7.11 to 8.83)	.833	-0.30 (-1.87 to 1.27)	.708	0.13 (-1.60 to 1.87)	.880
High vs control, day 1	14.68 (6.46-22.89)	.001 ?	4.83 (2.41-7.25)	b.0001 ?	0.82 (-0.44 to 2.09)	.203
?, adjusted mean difference compared with the reference group. * P b .05.

Table 3 Multivariate GEE model for arrhythmia

Arrhythmia aOR (95% CI)

Effect of temperature

P

Normothermia (37?C) Hypothermia (34?C) Effect of sepsis Control

Low High

Referent group

3.66 (0.29-45.82)

-

.314

Referent group

2.21 (0.20-23.85)

3.67 (0.39-34.33)

-

.514

.255

Interaction of temperature and sepsis

Low vs control, normothermia Low vs control, hypothermia High vs control, normothermia High vs control, hypothermia Effect of time

Before After Day 1

Abbreviation: aOR, adjusted odds ratio.

Referent group

0.45 (0.02-12.03)

Referent group

0.27 (0.01-6.49)

-

.636

-

.422

Referent group

2.10 (0.66-6.72)

0.17 (0.02-1.68)

-

.211

.131

Low frequency, HF, and LF/HF

Following the inoculation of bacteria, both LF and HF increased significantly 1 day later in the normothermia and hypothermia groups in both sepsis levels (Fig.). Table 2 shows the effects of sepsis severity; temperature; and time on the 3 HRV parameters, LF, HF, and the LF/HF ratio. These 3 parameters did not change with time when the rats were treated as a whole and not grouped according to sepsis severity or normothermia/hypothermia. However, when rats were grouped according to sepsis severity, high sepsis severity significantly increased the LF/HF ratio, and sepsis of both low and high severity significantly increased LF and HF compared with no sepsis on day 1 after bacterial injection.

Hypothermia significantly increased HF when rats were divided into hypothermia and normothermia groups without regard to time or sepsis severity. When sepsis severity was included in the grouping, hypothermia decreased HF significantly in low, but not in high sepsis.

Low-frequency and HF changes with time after bacterial injection were similar in normothermic and hypothermic rats.

Arrhythmia

Table 3 shows the effect of sepsis severity level, temperature, and time course on arrhythmia. No significant effects on arrhythmia were seen with any of the 3 independent variables.

Discussion

In the current study, we investigated whether hypothermia lessened the HRV decrease seen in sepsis and improved

outcome, using 2 levels of S aureus-induced sepsis in a rat model. Both low and high severity sepsis increased LF and HF by 1 day after bacterial injection. Hypothermia significantly increased HF when considered without regard to sepsis severity and, when sepsis severity was included in the analysis, low, but not high, severity sepsis significantly decreased HF. No significant differences in arrhythmia were seen, whether rats were grouped by sepsis level, temperature, or time of recording.

Our observations that LF increased in mild and severe sepsis are at variance with reports by others that LF is decreased in patients with sepsis [7-9]. We believe that this difference is related to differences in the times that the measurements were taken because we measured LF at 1 day (ie, early in the development of sepsis), and Annane et al [22] have reported that LF decreased significantly in patients in the late, but not the early stages of sepsis.

Cardiovascular reflexes and the autonomic nervous system may be involved in the cardiovascular dysfunction seen in sepsis patients. Heart rate variability is decreased in patients with severe illness or injury, indicating a lessening of autonomic nervous system control of the cardiovascular system [6,23]. Spectral HRV analysis has been used to assess autonomic nervous modulation in critically ill patients. A reduction in the LF component of HRV in seriously ill patients has been reported in a number of studies [24-26]. The LF component has been linked to sympathetic alteration of the heart rate, especially during conditions of stress, and the HF component, to respiration and to parasympathetic nervous system mediation through cardiac vagal activity. Changes in both components of HRV have been shown to be correlated to sepsis in animal and human studies [27-29], and inflammatory cytokine levels have been reported to be increased 48 hours before the onset of sepsis [30]. Our data showed a significant change in the LF/HF ratio in severe sepsis but no significant effect of sepsis on LF or HF levels themselves. A decrease in the LF/HF ratio is considered to indicate a lessening of the ability of the autonomic nervous system (ANS) to control heart rate [4].

The induction of hypothermia has been reported to have a protective effect in a number of pathologic situations [15- 18]. A lessened need for oxygen and a decrease in the production of Inflammatory mediators have both been suggested as being responsible for this protection. The only effect of hypothermia on HRV in our study was a change in the HF component in low severity sepsis. No effect was seen on LF, the component related to Sympathetic activity. Because (due to manpower constraints) we did not record HRV at times later than 24 hours after infection and the deaths occurred 36, 51, 69, and 70 hours after infection, we could not analyze whether there was any correlation between changes in HRV and death. This type of data would have to be collected under another study design.

The mortality rate for sepsis has been estimated to be 30% to 50% [1,2]. The mortality rate for endotoxin injection sepsis in rats is also very high [19,20], but some studies have

observed that hypothermic patients in septic shock have a higher mortality rate than febrile septic shock patients [31- 33]. In this study, rats with severe sepsis showed the highest mortality, whereas those with mild sepsis had lower mortality. However, no deaths occurred in hypothermic rats, either those with severe or those with mild sepsis.

Some studies have speculated that endotoxin-induced systemic inflammation and respiratory sinus arrhythmia can result in increased Vagal tone [6,34]. We used Zoletil-50 to induce anesthesia, and this anesthetic may result in an increase in respiratory frequency. In addition, we used live staphylococcal inoculation, and this bacterial species pro- duces superantigens and other toxins but not the endotoxin produced by Gram-negative bacteria.

There are some limitations in our study. We performed this study in a rodent species under conditions of general anesthesia. More samples to increase statistical power are needed for further study. In addition, we were unable to provide ECGs at the time of the animal’s death due to unavailability of continuous monitoring. We determined the cause of death as sepsis in accordance with the procedures received but, owing to the lack of monitoring, could not determine whether or not the animals died of arrhythmia.

References

1. Dellinger P. Cardiovascular management of septic shock. Crit Care Med 2003;31:946-55.
2. Friedman G, Silva E, Vincent JL. Has the mortality of septic shock changed with time? Crit Care Med 1998;26:2078-86.
3. Smith W, Pheley A, Collier R, et al. Severe sepsis in the emergency department and its association with a complicated clinical course. Acad Emerg Med 1998;5:1169-76.
4. Kleiger RE, Stein PK, Bigger Jr JT. Heart rate variability: measurement and clinical utility. Ann Noninvasive Electrocardiol 2005;10:88-101.
5. Kamath MV, Fallen EL. power spectral analysis of heart rate variability: a noninvasive signature of cardiac autonomic function. Crit Rev Biomed Eng 1993;21:245-311.
6. Goldstein B, Kempski MH, Stair D, et al. Autonomic modulation of heart rate variability during endotoxin shock in rabbits. Crit Care Med 1995;23:1694-702.
7. Vasilios E, Papaioannou M, Dragoumanis C, et al. Relation of heart rate variability to serum levels of C-reactive protein, interleukin 6, and 10 in patients with sepsis and septic shock. J Crit Care 2009;24: 6251-7.
8. Tateishi Y, Oda S, Nakamura M, et al. Depressed heart rate variability is associated with high IL-6 blood level and decline in the blood pressure in septic patients. Shock 2007;28:549-53.
9. Pontet J, Contreras P, Curbelo A, et al. Heart rate variability as early marker of multiple organ dysfunction syndrome in septic patient. J Crit Care 2003;18:156-63.
10. Bone RC. The pathogenesis of sepsis. Ann Intern Med 1991;115: 457-69.
11. Koyama S. Central impairment of sympathetic outflow during hemorrhagic shock and endotoxin shock. Prog Clin Biol Res 1988; 264:181-90.
12. Lee JY, Zhao L, HS Youn, et al. Saturated fatty acid activates but polyunsaturated fatty acid inhibits Toll-like receptor 2 dimerized with Toll-like receptor 6 or 1. J Biol Chem 2004;279:16971-9.
13. Lee JY, Plakidas A, Lee WH, et al. Differential modulation of Toll-like receptors by fatty acids: preferential inhibition by n-3 polyunsaturated fatty acids. J Lipid Res 2003;44:479-86.
14. Quezado ZM, Natanson C. Systemic hemodynamic abnormalities and vasopressor therapy in sepsis and septic shock. Am J Kidney Dis 1992; 20:214-22.
15. Gundersen Y, Vaagenes P, Pharo A, et al. Moderate hypothermia blunts the inflammatory response and reduces organ injury after acute haemorrhage. Acta Anaesthesiol Scand 2001;45:994-1001.
16. Vazquez-Jimenez JF, Qing M, Hermanns B, et al. Moderate hypothermia during cardiopulmonary bypass reduces myocardial cell damage and myocardial cell death related to cardiac surgery. J Am Coll Cardiol 2001;38:1216-23.
17. Kimura A, Sakurada S, Ohkuni H, et al. Moderate hypothermia delays proinflammatory cytokine production of human peripheral blood mononuclear cells. Crit Care Med 2002;30:1499-502.
18. Soukup J, Zauner A, Doppenberg EM, et al. Relationship between brain temperature, brain chemistry and oxygen delivery after severe human head injury: the effects of Mild hypothermia. Neurol Res 2003; 24:161-8.
19. Russell JA. Management of sepsis. N Engl J Med 2006;355:1699-713.
20. Vincent JL, Abraham E. The last 100 years of sepsis. Am J Respir Crit

Care Med 2006;173:256-63.

1. Chang YT, Wann SR, Hsieh KH, et al. The effects of lazaroid U- 74389G in a rat sepsis model. Inflamm Res 2011;60:29-35.
2. Annane D, Trabold F, Sharshar T, et al. Inappropriate sympathetic activation at onset of shock. Am J Respir Crit Care Med 1999;160: 458-65.
3. ESC, NAPSE: Heart rate variability: Standards of measurements, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and the North American society of Pacing and Electrophysiology. Circulation 1996;93:1043-65.
4. Chen W-L, Chen J-H, Huang C-C, et al. Heart rate variability measure as predictors of in-hospital mortality in ED patients with sepsis. Am J Emerg Med 2008;26:395-401.
5. Garrard CS, Kontoyannis DA, Piepoli M. Spectral analysis of heart rate variability in the sepsis syndrome. Clin Auton Res 1993;3:5-13.
6. Yien HW, Hseu SS, Lee LC, et al. Spectral analysis of systemic arterial pressure and heart rate signals as a Prognostic tool for the prediction of patient outcome in the intensive care it. Crit Care Med 1997;25:258-66.
7. Goldstein B, Fiser DH, Kelly MM, et al. Decomplexification in critical illness and injury: relationship between heart rate variability, severity of illness, and outcome. Crit Care Med 1998;26:352-7.
8. Korach M, Sharshar T, Jarrin I, et al. Cardiac variability in critically ill adults: influence of sepsis. Crit Care Med 2001;29:1380-5.
9. Toweill D, Sonnenthal K, Kimberly B, et al. Linear and nonlinear analysis of hemodynamic signals during sepsis and septic shock. Crit Care Med 2000;28:2051-7.
10. Kuster H, Weiss M, Willeitner AE, et al. Interleukin-1 receptor antagonist and interleukin-6 for early diagnosis of Neonatal sepsis 2 days before clinical manifestation. Lancet 1998;352:1271-7.
11. Marik PE, Zaloga GP. Hypothermia and cytokines in septic shock. Intensive Care Med 2000;26:716-21.
12. Arons MM, Wheeler AP, Bernard GR, et al. Effects of ibuprofen on the physiology and survival of hypothermic sepsis. Crit Care Med 1999;27:699-707.
13. Clemmer TP, Fisher CJ, Bone RC, et al. Hypothermia in the sepsis syndrome and clinical outcome. Crit Care Med 1992;20:1395-401.
14. Borovikova LV, Ivanova S, Zhang M, et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 2000;6785:458-62.