Abstract

Background

Sepsis is characterized by an increase in nitric oxide (NO) production, hemodynamic dysfunction and multiple organ failure. Propofol, a commonly used anesthetic in the intensive care unit for sedation and hypnosis, is thought to exert a protective effect on NO overproduction by inhibiting the expression of inducible NO synthase (iNOS) in sepsis. The aim of this study was to examine the influence of propofol on the temporal changes in the power density of frequency components of systemic arterial pressure (SAP) variability in rats subject to Escherichia coli lipopolysaccharide (LPS)-induced endotoxemia and to delineate the role of iNOS.

Methods

Twenty-four Sprague-Dawley rats were used in this study. Arterial catheters were inserted for direct SAP monitoring, and SAP signals were subjected to online real-time power spectral analysis via fast Fourier transformation. Animals were divided into four groups to receive different dosages of propofol infusion (15 or 30 mg/kg/hr) with or without the iNOS synthase inhibitor, S-methylisothiourea (SMT, 2 mg/kg). Changes in the power density of high-frequency (BHF), low-frequency (BLF) or very-low-frequency (BVLF) components of the SAP spectrum in each group were analyzed.

Results

Systemic injection of LPS resulted in a transient decrease, followed by a rebound increase and a second-phase decrease in the power density of BHF, BLF and BVLF of the SAP spectrum. Compared with low dose (15 mg/kg/hr), high-dose (30 mg/kg/hr) propofol anesthesia profoundly suppressed the power density of BHF, BLF and BVLF of SAP signals during LPS-induced endotoxemia. Such suppression was not affected by SMT pretreatment. SMT treatment, on the other hand, augmented the rebound increase in the power density of BHF and BLF, and ameliorated the second phase decrease in the power density of BHF in the endotoxemic rats that were maintained under low dose propofol anesthesia.

Conclusion

We concluded that iNOS-induced NO might be involved in the manifestation of the BHF and BLF components of the SAP spectrum during endotoxemia when low-dose propofol is used, and this effect of NO is blunted when high-dose propofol is administered.

Keywords

blood pressure: arterial; endotoxemia; nitric oxide synthase; propofol; spectrum analysis;

1. Introduction

Spectral analysis of systemic arterial pressure (SAP) variability was reported to be a useful tool to evaluate outcome in critical patients with sepsis, severe brain injury or organophosphate poisoning.¹⁻³ Different frequency components of SAP signals represent the different influences from the autonomic nervous system on the cardiovascular system,⁴ and changes in these spectral components have been demonstrated to be able to predict mortality in patients with multiple organ dysfunction syndrome under sepsis or other conditions.⁵ Four frequency components can be resolved in the lower end of the SAP spectrum. These include the very-high-frequency (BVHF), highfrequency (BHF), low-frequency (BLF) and very-lowfrequency (BVLF) components.⁴ In patients who succumb to sepsis, a progressive reduction or loss in the power density of the BLF and BVLF components of SAP signals invariably precede death.⁵ Conversely, an increase in the power density of both components over time might signify a positive prognosis.^4,5

Sepsis carries a high mortality rate and is the leading cause of death in critically ill patients.⁶ Overproduction of nitric oxide (NO), mainly from inducible NO synthase (iNOS), accounts primarily for cardiovascular dysfunction associated with sepsis.⁷ In addition, NO also affects the autonomic regulation of the peripheral circulation during sepsis.⁸ For patients in intensive care units, sedative drugs are commonly given to reduce suffering from stress or discomfort. Propofol is a commonly used anesthetic sedative that is used for this purpose. Whether or not various doses of propofol differentially affect the manifestation of the SAP spectrum during sepsis, and the role played by iNOS-induced NO, however, are not fully understood. This study was therefore performed to investigate the effect of different doses of propofol anesthesia on the manifestation of the frequency components of SAP signals in a rodent model of sepsis induced by systemic injection of Escherichia coli lipopolysaccharide (LPS), and to decipher the role that is played by iNOS-induced NO. Our results indicate that high-dose propofol blunted the autonomic regulation of cardiovascular function during sepsis in an iNOS-independent mechanism. On the other hand, iNOS-induced NO may be involved in changes in the power density of the BHF and BLF components of SAP signals in septic animals maintained under low-lose propofol.

2. Material and Methods

2.1. Animal preparation

Adult male Sprague-Dawley rats weighing 260−450 g (purchased from the National Experimental Animal Center, Taiwan) were used in this study. All experimental procedures were carried out pursuant to the guidelines of our institutional animal care committee. Rats were anesthetized initially with pentobarbital sodium (50 mg/kg, i.p.) for performing experimental preliminaries,⁹ which included intubation of the trachea to facilitate ventilation and cannulation of the femoral artery and vein for SAP measurement and drug administration. All invasive procedures were performed under surgical plane of anesthesia as indicated by the absence of withdrawal reflex to hindpaw pinch. Pulsatile SAP or mean SAP (MSAP) and heart rate (HR) were recorded on a polygraph. Animals were mechanically ventilated to maintain an end-tidal CO₂ within 4−5%, as monitored by a capnograph. All data were collected at rectal temperature of 37 ± 0.5ºC. At the end of each experiment, rats were killed with an intravenous injection of an overdose of pentobarbital sodium (100 mg/kg).

2.2. Power spectral analysis of systemic arterial pressure signals

The recorded SAP signals were simultaneously subject to online power spectral analysis.¹⁰ We were particularly interested in the BHF (0.8−2.4 Hz), BLF (0.25−0.8 Hz) and BVLF (0−0.25 Hz) components of SAP signals. These frequency components of SAP signals were reported to reflect, respectively, the respiratory pumping mechanism, the synchronizing influence of baroreceptor afferent, and the prevalence of sympathetic neurogenic vasomotor tone.⁴ The power densities of these spectral components were displayed during the experiment, along with SAP, MSAP and HR, in an online and real-time manner.

2.3. Induction of experimental endotoxemia

Experimental endotoxemia in the rats was induced by intravenous injection of Escherichia coli lipopolysaccharide (LPS, 15 mg/kg, serotype 0111:B4).¹¹ Injection of the same amount of 0.9% saline served as the vehicle and volume control. The temporal changes in mean SAP, HR and power density of the vasomotor components of SAP signals were designatively followed for 6 hours.

2.4. Experimental protocols

After a 30-minute period of stable hemodynamics following the completion of general preparation, the animals received continuous intravenous infusion of propofol (15 or 30 mg/kg/hr) for maintenance of anesthesia and were randomly assigned to receive LPS (endotoxemia group) or saline (control group) injection 60 minutes after propofol infusion. In each experimental group, a specific iNOS blocker, SMT (2 mg/kg), or normal saline was administered intravenously 30 minutes before the induction of endotoxemia.

2.5. Statistical analysis

The results are expressed as mean ± standard error of mean (SEM). One-way analysis of variance (ANOVA) and t test were used for the statistical evaluation of differences among means. A value of p< 0.05 was considered to be statistically significant.

3. Results

3.1. Effect of different doses of propofol on temporal changes in the power density of the frequency components of SAP signals during experimental endotoxemia

Intravenous injection of LPS (15 mg/kg) resulted in temporal changes in the frequency components of SAP signals characterized by an immediate and significant decrease in SAP and the power density of the BHF, BLF and BVLF components of the SAP spectrum (phase I), followed by a cessation of fall with a reverse significant increase in these frequency components (phase II) and a secondary decrease in SAP, HR and the power density of the BHF, BLF and BVLF components of SAP signals (phase III) (Figures 1 and 2). In contrast with low-dose propofol (15 mg/kg/hr) anesthesia (Figure 1A), the temporal changes in the power density of the BHF, BLF and BVLF components of SAP signals during LPS-induced experimental endotoxemia in rats were significantly blunted under high-dose propofol (30 mg/kg/hr) anesthesia (Figure 2A), although the temporal profile was not affected. In particular, the amplitude of phase II increase in the BHF component (Figure 3), as well as duration of phase II changes in the power density of the BHF (Figure 3), BLF (Figure 4) and BVLF (Figure 5) components of SAP signals were significantly lessened in the endotoxemic animals under the influence of highdose propofol anesthesia. Control injection of saline had no effect on the power density of the BHF, BLF and BVLF components of SAP signals (data not shown). In addition, propofol (either 15 or 30 mg/kg/hr) infusion alone did not change the power density of the identical components of the SAP spectrum in the saline-treated rats (Figures 3−5).

3.2. Effect of SMT treatment on temporal changes in the power density of the frequency components of SAP signals during experimental endotoxemia

In contrast with saline pretreatment which had no discernible effect on temporal changes in the power density of the frequency components of SAP signals in the rats after LPS injection under the influence of low- or high-dose propofol anesthesia, pretreatment with the iNOS inhibitor, SMT (2 mg/kg), significantly prolonged the duration of phase II increase in the power density of the BHF component (Figure 3) and potentiated the amplitude of phase II increase in the power density of the BLF (Figure 4), but not the BVLF component (Figure 5) of SAP signals during experimental endotoxemia in the rats under lowdose propofol anesthesia. The same treatment, on the other hand, had no effect on the amplitude or duration of temporal changes in the power density of these frequency components of SAP signals in the endotoxemic rats under maintenance of high-dose propofol anesthesia (Figures 3−5).

Download full-size image

Figure 1 Representative original tracings show temporal changes in systemic arterial pressure (SAP), heart rate (HR), and the power density of the high-frequency (BHF), low-frequency (BLF) and very-low-frequency (BVLF) components of SAP signals after systemic injection of Escherichia coli lipopolysaccharide (LPS, 15 mg/kg) in rats under lowdose propofol anesthesia (15 mg/kg/hr) with or without S-methylisothiourea (SMT, 2 mg/kg) pretreatment. Upward arrow indicates the time point of saline or SMT pretreatment. Downward arrow indicates the time point when LPS was injected.

Download full-size image

Figure 2 Representative original tracings show temporal changes in systemic arterial pressure (SAP), heart rate (HR), and the power density of the high-frequency (BHF), low-frequency (BLF) and very-low-frequency (BVLF) components of SAP signals after systemic injection of Escherichia coli lipopolysaccharide (LPS, 15 mg/kg) in rats under highdose propofol anesthesia (30 mg/kg/hr) with or without S-methylisothiourea (SMT, 2 mg/kg) pretreatment. Upward arrow indicates the time point of saline or SMT pretreatment. Downward arrow indicates the time point when LPS was injected.

Download full-size image

Figure 3 Time-course changes in the power density of the high-frequency component of systemic arterial pressure spectrum after systemic injection of Escherichia coli lipopolysaccharide (LPS) to animals maintained under: (A) lowdose (15 mg/kg/hr) propofol anesthesia or (B) high-dose (30 mg/kg/hr) propofol anesthesia with or without S-methylisothiourea (SMT, 2 mg/kg) pretreatment. Values are mean ± SEM (n= 6 in each group). *p< 0.05 vs. non-LPS control; † p< 0.05 vs. SMT group.

Download full-size image

Figure 4 Time-course changes in the power density of the low-frequency component of systemic arterial pressure spectrum after systemic injection of Escherichia coli lipopolysaccharide (LPS) to animals maintained under: (A) lowdose (15 mg/kg/hr) propofol anesthesia or (B) high-dose (30 mg/kg/hr) propofol anesthesia with or without S-methylisothiourea (SMT, 2 mg/kg) pretreatment. Values are mean ± SEM (n= 6 in each group). *p< 0.05 vs. non-LPS control; † p< 0.05 vs. SMT group.

Download full-size image

Figure 5 Time-course changes in the power density of the very-low-frequency component of systemic arterial pressure spectrum after systemic injection of Escherichia coli lipopolysaccharide (LPS) to animals maintained under: (A) low-dose (15 mg/kg/hr) propofol anesthesia or (B) high-dose (30 mg/kg/hr) propofol anesthesia with or without S-methylisothiourea (SMT, 2 mg/kg) pretreatment. Values are mean ± SEM (n= 6 in each group). *p< 0.05 vs. non-LPS control; † p< 0.05 vs. SMT group.

4. Discussion

We provide in vivo evidence to demonstrate that septic shock may be associated with a marked reduction in the autonomic regulation of cardiovascular function. This effect was more pronounced in animals under maintenance of high-dose propofol anesthesia and might involve generation of iNOS-derived NO.

Power spectral analysis of SAP signals has been validated as an efficient method for evaluation of sympathovagal balance and autonomic influence on cardiovascular function.³ Current interpretation of the physiological significance of the spectral components of SAP signals stipulates that the HF component of SAP reflects the influence of respiratory pumping mechanism on the heart and vessels.^10,12,13 The LF component is related to the synchronizing influence of baroreceptor afferents, and the VLF component arises from the non-oscillatory perturbations of regional vasculature.^14,15 Both the LF and VLF components of SAP signals are linked to neurogenic vasomotor tone.⁴ All these spectral components of SAP signals are subjected to modification under pathological conditions.⁴ Consistent with previous reports,^4,16 we found that the temporal changes in the HF, LF and VLF components of SAP signals were characterized by an immediate and significant decrease in the power density of the HF, LF and VLF components of SAP spectrum (phase I), followed by a significant increase in these frequency components (phase II) and a secondary decrease in the power density of the HF, LF and VLF components of SAP signals (phase III) in the rats subjected to experimental endotoxemia. These results indicate that the autonomic regulation of hemodynamic parameters is subjected to modulation by pathological condition of endotoxemia. Our observations that phase III decrease in neurogenic sympathetic modulation of the cardiovascular system preceded hemodynamic shock in LPS-challenged animals (Figures 1 and 2) argue for a role of the dysfunction of the autonomic nervous system in the onset of shock in sepsis.^4,17 In this regard, phase III reduction in the LF and VLF components of SAP signals is intimately associated with failure in brainstem control of vascular vaso-motor activity, resulting in fatal cardiovascular de-pression during sepsis.^1,4 Treatments that prolong phase II increase or abrogate phase III decrease in the same frequency components, on the other hand, ameliorate cardiovascular depression and increase survival rate of sepsis.¹⁸

Patients with endotoxemia often require drugs for sedation and analgesia in the intensive care unit. Several anesthetics, such as ketamine,¹⁹ dexmedetomidine²⁰ and propofol,^20,21 have been used in patients with septic shock for these purposes. The beneficial effects of propofol in patients with sepsis have been demonstrated in many aspects including anti-inflammation²² and anti-oxidation.²³ These beneficial effects of propofol, however, are promoted via a wide dosing range. For example, Kwak et al reported that low-dose (5 mg/kg/hr) propofol reduces endotoxin-induced acute lung injury in rabbits subjected to LPS endotoxin.²⁴ High-dose (50 mg/kg/ hr) propofol, on the other hand, is required to rescue diaphragmatic dysfunction in septic hamsters.²⁵ Effects of different doses of propofol on the autonomic regulation of cardiovascular function during sepsis have not yet been determined. As such, one of the major contributions of this study is to demonstrate that, in addition to its effect on SAP and HR, different concentrations of propofol can differentially modulate the influence of the respiratory pump on the heart and vessels, and the autonomic control of circulation during different phases of endotoxemia. As revealed by power spectral analysis of the SAP signals, both amplitude and duration of phase II increase in the HF, LF and VLF components of SAP signals were profoundly inhibited by high-dose propofol. These results indicate that high-dose propofol suppressed profoundly the influence of respiratory pumping mechanism on the heart and vessels, as well as neurogenic sympathetic outflow to the vasculature. Since these frequency components of SAP signals originate from the medulla oblongata,²⁶ it is conceivable that high-dose propofol inhibits these frequency components via its direct depressive action on neural activities in this brain region. In support of this suggestion, propofol dosedependently inhibits spontaneous neuronal activity in the rostral ventrolateral medulla,²⁷ the origin of the LF and VLF components of SAP signals is rational.⁵

Diverse molecular mechanisms of cellular damage by endotoxins contribute to cardiovascular depression during sepsis, of which overt production of NO in the cardiovascular and pulmonary organs are of particular importance.^7,28 Overproduction of NO via activation of iNOS contributes to fatal cardiovascular depression and mortality during late-stage endotoxemia.^29,30 Inhibition of iNOS-derived NO, on the other hand, preserves cardiovascular function,³⁰ increases the LF component of SAP signals,⁸ and protects animals against sepsis.³⁰ Although a vast amount of evidence supports the pivotal role of NO in cardiovascular depression, its effect on the autonomic regulation of hemodynamic function during endotoxemia has not been fully defined. As such, another major contribution of the present study was to demonstrate that the interaction of iNOS-derived NO and propofol may modulate the influence of the autonomic nervous system, in particular the respiratory pumping mechanism and the baroreceptor afferents, in the regulation of cardiovascular function systems during sepsis. Potentiation of amplitude and/or duration of phase II increase in the HF and LF components of SAP signals by SMT treatment in septic animals under low-dose propofol indicates that this anesthetic may interact with iNOS-derived NO to suppress the respiration pumping mechanism and vagosympathetic balance to perturb hemodynamic stability during sepsis. Since propofol and SMT alone did not affect the baseline power density of the HF, LF and VLF components of SAP signals before LPS or saline injection, it is reasonable to stipulate that alterations in temporal changes in the power density of SAP spectrum during endotoxemia in animals subjected to propofol infusion alone or with additional SMT treatment are the result of interaction of propofol and iNOS. NO produced by iNOS was reported to suppress the power density of these frequency components.⁸ High-dose propofol has been shown to suppress the production of NO.^25,31 This mechanism may explain why the suppression of the power density of the HF, LF and VLF components of SAP signals by high-dose propofol was not affected by SMT treatment. We realized that in addition to iNOS, propofol could affect the production of proinflammatory cytokines and free radicals.²³ The significance of cytokines and free radicals in temporal changes in the power density of the same components of SAP spectrum during endotoxemia, however, awaits further investigation. In this regard, we reported recently that the antioxidant effect of propofol might not be involved in protection against cardiovascular depression during late-stage endotoxemia.³² We also realized the importance of inclusion of a control group that did not receive propofol infusion. Since the induced endotoxemia in this study was carried out in the anesthetized animals, and propofol alone should have no effect on the baseline power density of the various components of SAP spectrum, we therefore reasoned that the effect of propofol on the manifestation of SAP spectrum during endotoxemia resulted mainly from the effect of anesthetic but not the vehicle.

In conclusion, our study demonstrated that in sepsis, high-dose (30 mg/kg/hr) propofol suppressed the power density of the HF, LF and VLF components of SAP spectrum via an iNOS-independent mechanism. Low-dose (15 mg/kg/hr) propofol, on the other hand, exerted an iNOS-produced suppression of the HF and LF components of SAP spectrum during sepsis. Therefore, the results of the present study suggest that interaction of propofol and iNOS-derived NO may contribute to impairment of the autonomic regulation of cardiovascular function during endotoxemia.Physicians should, therefore, pay extra attention to the interpretation of SAP variability in septic patients if propofol anesthetic is used.

Acknowledgments

This work was supported in part by Kaohsiung Veterans General Hospital (VGHKS93-09), Taiwan. We thank Dr Alice Y.W. Chang for her valuable and constructive comments on the study.