Abstract
Background
Sepsis is associated with high mortality and septic patients sometimes need sedation to attenuate anxiety and agitation. Recently, it has been suggested that midazolam, a benzodiazepine, modulates the responses to cytokines released from macrophages in vitro. The purpose of the present study was to evaluate the effect of midazolam on hemodynamics and organic function in rats with septic syndromes induced by intravenous endotoxin.
Methods
Wistar rats were randomly allocated to three groups and treated intravenously as follows: (1) control group, in which the rats were given saline vehicle (1 mL/kg) at time 0; (2) LPS group, in which the rats were given Escherichia coli lipopolysaccharide (10 mg/kg) in an infusion over 10 minutes; (3) LPS + Midazolam group, in which the rats were given E. coli LPS infusion as in the LPS group followed immediately by infusion of midazolam, given at a rate of 1 μg/kg per hour. All hemodynamic and biochemical parameters were measured during the 4-hour observation period.
Results
The injection of LPS led to hypotension, tachycardia, and vascular hyporeactivity to vasoconstrictors. In addition, initial hyperglycemia, delayed hypoglycemia, hypoalbuminemia, elevated serum indicators of hepatic and renal injury and high mortality were also observed in rats treated with LPS. However, midazolam given in the designated dosage did not offer any modulatory effects on hemodynamic responses, multiple organic dysfunction and survival rate in rats with endotoxemia.
Conclusion
No deleterious or beneficial effects on hemodynamics and organ dysfunction were observed in the endotoxemic rats treated with midazolam at the designated dosage. However, previous reports which showed substantial benefits from the inhibitory effects of midazolam on proinflammatory factors in vitro should be subjected to further elucidation.
Keywords
benzodiazepines; hemodynamics; multiple organ failur; esepsis; survival rate;
1. Introduction
Sepsis is a syndrome of heterogeneity, caused by a systemic inflammatory response and is usually associated with the development of progressive damage of multiple organs. Eventually, multiple organic dysfunction syndrome (MODS) becomes a serious clinical entity and is a common cause of death in the critically ill patient. Sepsis is clinically characterized by profound hypotension, vascular hyporeactivity to vasoconstrictors, and often culminates in MODS. Lipopolysaccharide (LPS), embedded in the outer wall of Gram-negative bacteria, appears to be the key culprit involved in the interaction with the reticuloendothelial system in the initiation of sepsis.1,2 In sepsis, endotoxin induces production of a variety of chemical mediators, including cytokines, nitric oxide (NO) and reactive free radicals. These mediators contribute to the shock-like septic state and end-organ injury.
Benzodiazepines are commonly used as an anesthetic adjunct in certain surgical procedures to eradicate early septic foci and provide an optimal level of sedation for patients in the intensive care unit. In addition, it is known that benzodiazepines suppress LPS-induced interleukin-1 (IL-1), IL-6 and tumor necrosis factor (TNF) activity in mouse macrophages via the peripheral benzodiazepine receptors.3,4 Meanwhile, midazolam, a shorter-acting benzodiazepine derivative, not only modulates the production of cytokines, but also decreases the gen eration of NO and inhibits the function of neutrophils.5−7 Therefore, midazolam may have clinical implications in high-risk and immunocompromised patients.
Besides symptomatic treatment, blocking specific steps in the septic cascade has been considered to be of potential benefit. Previous studies showed that propofol, another intravenous anesthetic, ameliorates inflammatory mediators, attenuates organic dysfunction and improves mortality rate in animals with endotoxic shock.8,9 Therefore, we examined the effect of midazolam on the hemodynamics, organ injury/dysfunction (i.e. liver and kidney) and survival rate in animals subjected to endotoxemia.
2. Methods
2.1. In vivo experiments
All experimental protocols were approved by the Institutional Animal Care and Use Committee of National Defense Medical Center pursuant to the Helsinki Declaration and internationally accepted principles for the care and use of experimental animals. Male Wistar rats weighing 280−350 g were purchased from the National Laboratory Animal Center of Taiwan and housed under a 12-hour light/ dark cycle at a controlled temperature (21 ± 2ºC) with free access to food and tap water. All tested rats were anesthetized with intraperitoneal sodium pentobarbital (50 mg/kg). Polyethylene catheters were placed in the left jugular vein for the administration of drugs and in the right carotid artery for the measurement of hemodynamics and blood sampling. The distal part of the catheters was tunneled under the skin and externalized through an incision in the back of the neck.
After surgical cut-down for catheter access was completed, the rats were fasted overnight (about 12 hours) for recovery but allowed water ad libitum. The animals were randomly allocated to one of three groups and treated with different agents intravenously (i.v.) as follows: (1) control group was given vehicle (1 mL/kg of saline, at time 0); (2) LPS group was given Escherichia coli lipopolysaccharide (LPS; E. coli serotype 0127:B8, L3127) 10 mg/kg, infusion over 10 minutes; (3) LPS + Midazolam group was given E. coli LPS given as in group LPS plus midazolam 1 μg/kg per hour, immediately after LPS infusion.
The experiments were then performed on pairs of conscious rats and all groups were treated in parallel. The arterial catheter was connected to a pressure transducer (P23ID; Statham, Oxnard, CA, USA) for the measurement of phasic arterial pressure, mean arterial pressure (MAP) and heart rate, all of which were recorded on a multichannel recorder (MacLab/4e; AD Instruments Pty. Ltd., Castle Hill, Australia). After recording baseline hemodynamic parameters, animals were given norepinephrine (NE; 1 μg/kg, i.v.) to examine their vascular reactivity. In order to normalize the baseline value of pressor responses to NE of all groups, we calculated the value of pressor responses to NE in the resting state (i.e. time 0) in each group as 100%.
All hemodynamics (i.e. MAP and heart rate) and the pressor responses to NE were reassessed hourly after vehicle or LPS injection. Arterial blood samples (0.3 mL) were obtained at baseline (time 0) and specified times (at 1, 2, and 4 hours) throughout the whole procedure. Each volume of blood removed was immediately replaced by the injection of an equal volume of sterile saline. At the end of each experiment, the surviving rats were euthanized by intravenous administration of an overdose of sodium pentobarbital.
2.2. Quantification of organ dysfunction
Blood (10 μL) was immediately analyzed for blood glucose (ONE TOUCH test strips; Lifescan, Milpitas, CA, USA). The remaining blood was then centrifuged for 3 minutes at 7000g and 70 μL of the serum was taken for measuring glutamine-oxaloacetic transaminase (GOT), glutamine-pyruvic transaminase (GPT), albumin, blood urea nitrogen (BUN), and creatinine (Fuji DRI-CHEM 3030; Fuji Photo Film Co., Tokyo, Japan).
2.3. Statistical analysis
The data are expressed as mean ± standard error of the mean. Statistical analysis was performed by one-way analysis of variance (ANOVA), followed by post hoc Scheffé’s test. Survival analysis was carried out using the Kaplan−Meier method, and comparisons between groups were made using the log-rank test. Statistical significance was accepted at p< 0.05.
3. Results
3.1. Effects of midazolam on survival in endotoxin-treated rats
During the whole 4-hour experimental study, 12 out of 20 rats (60%) survived in the LPS group, 10 out of 15 rats (66.7%) survived in the LPS + Midazolam group, and all eight rats survived in the control group (Figure 1). All LPS-treated animals presented with characteristics of severe sepsis, including ocular discharge, piloerection and diarrhea. In addition, signs of respiratory distress such as nasal flaring, tachycardia and hyperapnea were found at the end of the experiment.
3.2. Effects of midazolam on hemodynamic parameters in endotoxin-treated rats
The mean baseline values of MAP were between 119 ± 4 mmHg and 126 ± 6 mmHg in all studied animal groups, and there was no significant difference among groups. Administration of LPS produced a biphasic fall in MAP. The injection of LPS resulted in an initial fall in MAP from 125 ± 5 mmHg to 88 ± 7 mmHg at 1 hour, then a partial recovery at 2−3 hours, and a further decrease at 4 hours (76 ± 7 mmHg) was seen (Figure 2A). The infusion of midazolam did not cause significant changes of the initial and late hypotension after LPS. In the control group, there was no significant change in MAP during the experimental period.
The mean baseline values of heart rate were between 359 ± 20 beats/minute and 385 ± 18 beats/ minute in all studied animal groups, and there was no significant difference among groups. LPS caused a significant increase in heart rate from 385 ± 18 beats/minute to 593 ± 33 beats/minute at 2 hours and then a progressive fall occurred (Figure 2B). Interestingly, the LPS-induced tachycardia was slightly, but not significantly, enhanced by midazolam at 1 hour after LPS treatment. In the control group, there was no significant change in heart rate during the experimental period.
LPS caused a substantial attenuation of the pressor responses elicited by NE (Figure 2C). The infusion of midazolam did not cause a significant change in the NE-induced pressor responses after LPS. In the control group, there was no significant change in the NE-induced pressor responses during the experimental period.
3.3. Effects of midazolam on organ functional indices in endotoxintreated rats
The mean baseline values of blood glucose were between 98 ± 5 mg/dL and 108 ± 3 mg/dL in all studied animal groups, and there was no significant difference among groups. The administration of LPS significantly increased blood glucose from 105 ± 5 mg/ dL to 170 ± 7 mg/dL within 1 hour, with a subsequent decline below the initial baseline level to 58 ± 8 mg/dL (Figure 3A). However, the LPS-induced initial hyperglycemia and delayed hypoglycemia were not attenuated by treatment with midazolam. In the control group, there was no significant change in blood glucose during the experimental period.
The administration of LPS caused a significant increase in the serum activity of GOT, GPT, BUN and creatinine (Figures 3B−E), whereas it caused a significant decrease in albumin level (Figure 3F). The treatment of LPS rats with midazolam did not attenuate the changes in plasma levels of these biochemical parameters. In the control group, no significant changes in these parameters during the experimental period were observed.
4. Discussion
Sepsis is the cause of 30−50% of deaths in intensive care units despite recent advances in critical care therapy.10,11 The overall mortality rate of sepsis ranges from 30% to 100% depending on the number of organ systems involved.12 In a rat model of endotoxic shock after 4-hour endotoxemia, we demonstrated that administration of 10 mg/kg of E. coli LPS to rats could lead to a biphasic fall in blood pressure, tachycardia, and vascular hyporeactivity to NE. In addition, initial hyperglycemia, delayed hypoglycemia, hypoalbuminemia (impairment of vascular permeability), increases in serum levels of GOT and GPT (biomarkers of hepatocellular injury), BUN and creatinine (biomarkers of renal injury), and high mortality were also observed in rats treated with LPS.
Most clinical effects of benzodiazepines mainly arise from their action on the central nervous system, but an interesting and noticeable peripheral effect of these compounds is vasodilation, probably mediated through the release of NO from the endothelium and the activation of the BKCa K+ channel.13,14 Midazolam, like other benzodiazepines, lowers systemic vascular resistance and blood pressure especially in patients with hypovolemia and heart disease.15−17 Therefore, it is important to evaluate whether midazolam could cause hemodynamic instability and tissue perfusion impairment in critical patients. However, the data showed that midazolam did not cause a significant change in the initial and late hypotension after LPS treatment. Furthermore, the infusion of midazolam that partially enhanced the LPS-induced tachycardia was perhaps the result of compensation to vasodilation.
Midazolam, like other benzodiazepines, is also known to modulate the immune function via peripheral benzodiazepine binding sites.4,18 Several studies have revealed that midazolam suppresses activated macrophages and inhibits the synthesis and release of IL-1, IL-6, and TNF-α in vitro. 19−21 In addition to inhibition of cytokine production, midazolam also suppresses the expression of LPSstimulated inducible NO synthase, cyclooxygenase-2 and superoxide anion production, possibly through inhibition of nuclear factor (NF)-κB and the p38 mitogen-activated protein kinase pathway.22 However, in a human study, midazolam infusion did not modulate cytokine production in septic patients.23 Moreover, it did not affect the critical patient’s gastric intramucosal pH, an indicator of tissue perfusion. Here, we found that midazolam did not affect the LPS-induced metabolic disturbance (e.g. initial hyperglycemia and delayed hypoglycemia) and multiple organ dysfunction (hepatic and renal dysfunction, and hypoalbuminemia). The data failed to demonstrate that midazolam in the designated dosage used had any harmful or protective effect on the survival of rats with LPS-induced sepsis/ septic shock. This is, however, not consistent with a previous study which showed that mice treated with diazepam or chloride methyldiazepam had an improvement in survival to experimental Salmonella typhimurium infections after 3 months of treatment.24
In addition to vasodilation and hypotension, the use of midazolam carries the risk of respiratory suppression and lower ventilatory response to CO2. 25 Previous trials with benzodiazepines have shown an increased incidence of respiratory suppression in patients with septic shock.17 Therefore, the administration of high-potency midazolam necessitates careful titration to avoid overdose and apnea. In particular, the hepatic and renal dysfunction induced by endotoxin may lead to prolonged sedation and respiratory suppression in animals receiving midazolam because of the accumulation of a conjugated metabolite, α-hydroxymidazolam.26−28 Thus, respiratory failure induced by endotoxin, as seen in our study, may be the main cause of high mortality, suggesting that the dosage of intravenous midazolam must be titrated. The optimum dose of midazolam used in critically ill patients varies widely and the doses listed on the approved product literature also differ between countries.29−31 According to the patient’s condition, continuous infusion of midazolam at 0.03−0.4 mg/kg per hour is recommended.32 In our previous study, midazolam dosages of 0.05 mg/kg and 0.005 mg/kg per hour were used, but nearly all the rats were dead before the end of a 4-hour observation period. Thus, we chose a lower dosage (1 μg/kg per hour) to evaluate the effect of midazolam on conscious endotoxic rats.
In conclusion, the use of midazolam at the studied dose in conscious rats with endotoxemia had neither harmful nor beneficial effects on hemodynamic change and organ dysfunction, although a few previous reports have shown substantial benefits from its inhibition of in vitro production of proinflammatory factors.