Effects of Dexmedetomidine on Intestinal Microcirculation and Intestinal Epithelial Barrier in Endotoxemic Rats

DEXMEDETOMIDINE is a potent α2-adrenoceptor agonist¹  and widely used sedative in mechanically ventilated patients in intensive care units.²  Pandharipande et al.³  reported that the duration of mechanical ventilation was shorter and the mortality was lower in septic patients receiving dexmedetomidine than in those receiving a lorazepam-based sedation regimen. Although it was an a priori–designed subgroup analysis of the Maximizing Efficacy of Targeted Sedation and Reducing Neurological Dysfunction trial,⁴  the authors suggested that their findings might have resulted from the antiinflammatory effects and organ-protective properties of dexmedetomidine.^5,6  Moreover, Memiş et al.⁷  reported that dexmedetomidine infusion reduces cytokine production in critically ill septic patients. Dexmedetomidine has been reported to enhance macrophagic activity⁸  and reduce inflammation and mortality in endotoxemia and polymicrobial sepsis animal models in several studies.^9–11 

Several studies have revealed sepsis-related microcirculatory dysfunction, namely, decreased perfused vessel density,¹²  microthrombosis,¹³  shunting,¹⁴  endothelial dysfunction, and capillary hyperpermeability.¹⁵  Severe microcirculatory dysfunction and ischemic injury in the highly vascularized intestine may cause enteric bacterial translocation and systemic inflammatory responses, eventually leading to multiple organ dysfunction and death.^16,17  Endothelial cells play a key role in sepsis pathogenesis and microcirculatory dysfunction,^18–20  and an endothelial cell–specific molecule-1 (endocan, a 50-kDa dermatan sulfate proteoglycan) is expressed on endothelial cells and is acknowledged as a suitable marker of endothelial dysfunction.^21,22  Furthermore, Frey and Kesel²³  reported that splanchnic perfusion decreases early in sepsis. In our previous study, dexmedetomidine attenuated intestinal microcirculation alteration in a rat model of surgical stress and pain stimulation.²⁴  In the current study, we aimed to test the first hypothesis that the antiinflammatory effect of dexmedetomidine can alleviate intestinal microcirculatory dysfunction and reduces endothelial dysfunction in endotoxemic rats. Moreover, intestinal mucosal ischemia can cause mucosal cell death–dependent barrier disruption and subsequent bacterial translocation,^25–27  and some investigators have suggested that the gut is a source of multiple organ failure.^28,29  Dexmedetomidine administration before ischemia can attenuate intestinal injury induced by intestinal ischemia–reperfusion in rats.³⁰  Furthermore, perioperative administration of dexmedetomidine has been shown to attenuate intestinal injury in patients.³¹  Our previous studies have shown that epithelial cell death is partly responsible for tight junctional damage and bacterial translocation.^32,33  We speculate that if intestinal microcirculation could be preserved, epithelial cell death due to ischemia might also be prevented. Therefore, we aimed to test the second hypothesis that restoration of intestinal microcirculation may reduce the severity of epithelial cell death and intestinal barrier damage.

The primary objective of this study was to determine whether dexmedetomidine can reduce intestinal microcirculatory and endothelial dysfunction in endotoxemic rats by using a sidestream dark-field (SDF) videomicroscope, a tissue oxygen monitor, a full-field laser perfusion imager,^34,35  and an endocan assay kit. The secondary objective of this study was to investigate the integrity of the intestinal barrier by using the Ussing chamber permeability assay and lumen-to-blood passage of gadodiamide by magnetic resonance imaging (MRI).³⁶  In addition, epithelial cell death, tight junctional damage, and bacterial translocation to extraintestinal organs were investigated.

A total of 92 male Wistar rats (BioLASCO Taiwan Co., Taiwan) weighing 250 ± 50 g were used in this study, which was approved by the Animal Care and Use Committee of Laboratory Animal Center (No. 20130367; College of Medicine, National Taiwan University, Taipei, Taiwan, Republic of China). The rats were maintained on a 12-h light–dark cycle and had ad libitum access to water and food. The study was divided into three parts. Part A was designed to investigate the effects of dexmedetomidine on the microcirculation of the terminal ileum in endotoxemic rats and to determine the biomarkers of endothelial dysfunction; 40 rats were used. Part B was designed to determine the effect of dexmedetomidine on the intestinal epithelial permeability by ex vivo assay, bacterial translocation from the gut lumen to the liver and spleen; 40 rats were used. Part C was designed to investigate the effect of dexmedetomidine on the intestinal epithelial permeability by in vivo assay, tight junctional damage, and mucosal cell death levels; 12 rats were used. Considering the high variation of the severity of sepsis in cecal ligation and puncture model, we chose endotoxemic model for this experiment based on our experience from a previous study.

In parts A and B, the rats were randomly assigned to the following four groups (fig. 1). (1) Rats receiving 1.5 and 1 ml/kg 0.9% saline intravenously at 0 and 120 min, respectively, comprised the Sham group. Saline (0.9%) was continuously infused at 2 ml · kg⁻¹ · h⁻¹ as a maintenance fluid supplement through the external jugular vein catheter by using an infusion pump (KDS101; KD Scientific, USA). (2) Rats receiving 15 and 10 mg/kg lipopolysaccharide ( Escherichia coli, serotype O127:B8; Sigma, USA; 10 mg/ml in 0.9% saline) intravenously at 0 and 120 min, respectively, and continuous IV infusion of 0.9% saline at 2 ml · kg⁻¹ · h⁻¹ comprised the lipopolysaccharide group. (3) Rats receiving 1.5 and 1 ml/kg 0.9% saline intravenously at 0 and 120 min, respectively, and continuous IV infusion of dexmedetomidine (Precedex; Hospira, USA) at 5 μg · kg⁻¹ · h⁻¹ in 0.9% saline at 2 ml · kg⁻¹ · h⁻¹ from 0 to 240 min comprised the dexmedetomidine group. (4) Rats receiving 15 and 10 mg/kg lipopolysaccharide intravenously at 0 and 120 min, respectively, and continuous IV infusion of dexmedetomidine at 5 μg · kg⁻¹ · h⁻¹ in 0.9% saline at 2 ml · kg⁻¹ · h⁻¹ from 0 to 240 min comprised the lipopolysaccharide + dexmedetomidine group. The dosage of dexmedetomidine was determined by our pilot study and the studies by Taniguchi et al.⁹  and Zhang et al.³⁰  In part C, the rats were randomly assigned to the following three groups: the Sham group, the lipopolysaccharide group, and the lipopolysaccharide + dexmedetomidine group.

Initial anesthesia, weighing, and shaving of abdominal hair were performed as described in our previous study.³⁵  Atropine sulfate (0.05 mg/kg in 5 ml/kg 0.9% saline; Taiwan Biotech Co., Ltd., Taiwan) was administered subcutaneously. A 14-gauge catheter was inserted into the trachea via tracheostomy and connected to the anesthesia machine. The rats spontaneously breathed, and subsequent anesthesia was maintained with 1.0% isoflurane in oxygen (0.6 l/min). Polyethylene catheters were inserted into the right common carotid artery and external jugular vein. Arterial blood pressure and heart rate were continuously monitored using Biopac Student Lab PRO 3.7 software and MP45 hardware (BIOPAC Systems Inc., USA). Rectal temperature was continuously monitored. After the abdominal wall was disinfected, a 3-cm-long midline laparotomy was performed using sterile instruments. After locating the cecum, a segment of the terminal ileum (approximately 6 to 10 cm proximal to the ileocecal valve) was carefully exteriorized, and the mesenteric arteries were examined to avoid stretching or kinking. A 2-cm-long incision was made at the antimesenteric site of the intestinal lumen by using a high-frequency desiccator (Aaron 900; Bovie Medical, USA). Three intestinal sites (the mucosa, serosal muscular layer, and Peyer patch) were located to examine the microcirculation. The rats were observed for a 15-min equilibration period.³⁵ 

After 15 min, the microcirculation in the 40 rats of the four groups was examined using SDF videomicroscope (MicroScan; MicroVision Medical, The Netherlands) on the mucosa, serosal muscular layer, and Peyer patch. A tissue oxygen monitor (moorVMS-OXY; Moor Instruments, United Kingdom) was used for measuring the relative concentration of oxygenated hemoglobin (oxyHb), relative concentration of deoxygenated hemoglobin (deoxyHb), and tissue oxygen saturation (StO₂) of the mucosa, serosal muscular layer, and Peyer patch. After the examinations, the time point of the experiment was set as 0 min, and the rats received treatment according to their grouping and treatment protocol. The mucosa, serosal muscular layer, and Peyer patch of the rats were continuously examined using a full-field laser perfusion imager (moorFLPI; Moor Instruments Ltd.) for 240 min. At the end of the experiment, SDF videomicroscopy and tissue oxygen monitoring were performed again, and blood was withdrawn from the right common carotid artery catheter for laboratory analysis. The rats were euthanized through exsanguination under anesthesia.

The total small vessel (less than 20 μm) density, blood flow classification of small vessels, perfused small vessel density (PSVD), proportion of perfused small vessels (PPV), microvascular flow index (MFI) score, and heterogeneity index were investigated according to the suggestion of the round table conference for evaluating microcirculation.³⁷  The blood flow of each small vessel was classified by a semiquantitative method as follows: (0) absent (either no flow or filled with microthrombosis), (1) intermittent flow (presence of flow for less than 50% of the observation time), (2) sluggish flow, and (3) continuous flow.³⁷  The setting of videomicroscope was performed as described in our previous study.³⁵  Three continuous 10-s image sequences were recorded for each measured site. The images were analyzed using automated analysis software (AVA 3.0; Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands), and data on the three images were averaged for statistical analysis. The measurement and calculation of these microcirculation parameter were performed as described in our previous study.³⁵  The analyses were performed by a single investigator who was blinded to the grouping.

The moorVMS-OXY operates on the basis of the theory of white light reflectance spectroscopy. Measurements using this instrument rely on spectrophotometric principles that associate light absorption with chromophore concentrations. Measurements are performed using probes, which are placed in contact with the tissue at the measurement site. Optical fibers are used to deliver illumination light to the tissue and collect reflectance light for processing, and analysis is performed using spectra in the visible wavelength range between 500 and 650 nm. OxyHb is displayed as an arbitrary unit. The StO₂ reading is displayed as a percentage of the oxyHb divided by the relative total hemoglobin concentration (total Hb = oxyHb + deoxyHb).

The intensity of microvascular blood flow was measured using a full-field laser perfusion imager by employing the laser speckle contrast imaging technique, as described in our previous study.³⁴  Three regions of interest (ROIs) were set at the mode of single-point measurement: ROI 1 was selected on the mucosa, ROI 2 was selected on the serosal muscular layer, and ROI 3 was selected on the Peyer patch. The microvascular blood flow intensities of each ROI were recorded as flux with the perfusion unit. The changes of microvascular blood flow intensity were expressed as percentage changes of blood flux from the baseline (0 min) and were subsequently used for statistical analysis.

Serum endocan levels were measured using the Rat-Endocan ELISA Kit (Lunginnov SAS, France) according to the manufacturer’s instructions.

Rats were anesthetized with urethane (1.2 g/kg in 5 ml/kg 0.9% saline, intraperitoneal injection; Sigma), and 0.05 mg/kg atropine sulfate in 5 ml/kg 0.9% saline was administered subcutaneously. Tracheostomy was performed, and the rats spontaneously breathed. A polyethylene catheter was inserted into the external jugular vein for subsequent IV treatments. The rats were observed for a 15-min equilibration period, and the time point of the experiment was set as 0 min. The rats received treatment according to their grouping and treatment protocol. After 240 min, aseptic laparotomy was performed through a midline incision by using sterile instruments. The spleen and liver were removed from the rats, and the tissue weights were determined. After locating the cecum, two 3-cm segments of the terminal ileum (approximately 6 to 9 cm and 12 to 15 cm proximal to the ileocecal valve) were carefully removed, and the luminal contents were gently rinsed out using phosphate-buffered saline.

Muscle-stripped intestinal segments (2 cm) were opened along the mesenteric border and mounted on Ussing chambers (World Precision Instruments, USA). The chamber was prepared as described in our previous study.^36,38  A circulating water bath maintained the buffer temperature at 37°C. The potential difference (mV) was measured, and the short-circuit current (Isc, μA/cm²) of the tissue was determined in a voltage-clamp mode on line. The tissue was subjected to 1-mV, 1-s pulses in 5-min intervals, and the Isc change caused by the pulses was used for calculating the intestinal tissue conductance (mS/cm²) according to Ohm law.³⁹  A greater conductance indicated higher intestinal permeability. The level of mucosal-to-serosal flux of dextran conjugated to fluorescein isothiocyanate (FITC) (dextran-FITC, molecular weight [MW] = 4 kDa; Sigma) was used to determine intestinal epithelial permeability.⁴⁰  Tissues mounted on the Ussing chambers were allowed to equilibrate until the Isc stabilized, before the dextran probe was added to the mucosal buffer at a final concentration of 500 μM. Serosal buffer samples (250 μl) were collected at 0, 30, 60, 90, and 120 min after placement of dextran probe and were replaced with Krebs buffer. A multimode plate reader was used to determine the fluorescence units of dextran-FITC at excitation/emission = 490/530 nm, and the concentration (nM) was calculated.⁴¹  The mucosal-to-serosal flux levels of dextran-FITC at 30, 60, 90, and 120 min minus the baseline level (0 min) were expressed as an indicator of the intestinal epithelial permeability.

The tissues were homogenized and sonicated in sterile phosphate-buffered saline with a ratio of 1 mg to 10 μl.⁴²  Undiluted lysate (200 μl) was inoculated onto fresh blood agar plates (Scientific Biotech Corp., Taiwan). After incubation at 37°C for 24 h, the number of bacterial colonies was enumerated and normalized to colony-forming units per gram of tissue.³⁸ 

Initial anesthesia and operation were performed as described in our previous study.³⁵  To evaluate the changes in intestinal permeability in vivo and the lumen-to-blood passage of a small molecule, a novel MRI-based assay was conducted by applying a contrast probe gadodiamide (Gadoscan, MW = 574 Da; InnoPharmax, Taiwan) to the lumen of a ligated jejunal sac immediately after saline or lipopolysaccharide injection. After opening the abdominal wall, an 8-cm jejunal sac was created starting 6 cm proximal to the ileocecal valve for probe injection. Gadodiamide was instilled into the lumen of ligated jejunal sac at 0.25 M in 1 ml volume immediately after saline or lipopolysaccharide injection. The blood samples were collected before (0 min) or after injection at 30, 60, 120, and 240 min. For measurement of serum gadodiamide concentration, a standard solution with known concentration (0.5 M) of gadodiamide was diluted by a two-fold series ranging from 2⁻¹⁸ to 2⁻²⁹ M with 0.9% saline. Each serum sample and each standard solution were, respectively, collected in a 200-μl polypropylene tube. All samples were assayed in duplicates. We used 1.5-Tesla MRI (GE Signa Excite; General Electric, USA) for measuring the signal intensity of the serum. These test tubes were placed in a 96-well rack in a homemade water tank and then subjected to MRI scan. Images were obtained in the transverse and axial plane of the entire tube using 1.5-mm-thick slices. We selected Fast spin echo with T1-weighted sequence for imaging. Imaging parameters were as follows: repetition time = 400 ms, echo time = 11.8 ms, field of view = 14 × 14 cm, and number of excitations = 6. Matrix size was 256 × 192, and the number of slice was 10. The total scan time was 20 min.

The signal intensity of each sample solution was measured by circling a region of 4 mm² in MRI image using the workstation provided by GE Healthcare (United Kingdom). The signal intensity of serum samples was divided by that of background noise and expressed as signal-to-noise ratio. The signal-to-noise ratio of the standard solutions of gadodiamide was plotted against the concentration thereof to establish a standard curve. The gadodiamide concentration in serum samples at each time point was calculated by using the standard curve.

Intestinal mucosal proteins were extracted with complete radioimmunoprecipitation buffer and subjected to sodium dodecyl sulfate/polyacrylamide gel electrophoresis (4 to 13% polyacrylamide) as described.^42–44  The resolved proteins were then electrotransferred onto polyvinylidene difluoride or nitrocellulose membranes in a semidry blotter. Blots were blocked with 5% (w/v) nonfat dry milk in Tris-buffered saline (TBS) or 5% (w/v) bovine serum albumin in TBS with Tween-20 (TBS-T; 0.1% [v/v] Tween-20 in TBS) for 1 h, washed with TBS-T, and incubated with a primary antibody at 4°C overnight. The membrane was washed and incubated with a secondary antibody for 1 h. After washing, the membranes were incubated with chemiluminescent solution and signals detected. The primary antibodies used included anti-occludin (1:1,000; Invitrogen, USA), anti-caspase-3 (1:2,000; Cell Signaling Technology, USA), anticleaved caspase-3 (1:2,000; Cell Signaling Technology), and anti-β-actin (1:10,000; Sigma). The secondary antibodies used were horseradish peroxidase–conjugated goat anti-mouse immunoglobulin G (1:2,000; Santa Cruz, USA) and anti-rabbit immunoglobulin G (1:1,000; Cell Signaling Technology).

Data were analyzed using statistical software (SPSS 20; IBM SPSS, USA). Based on previous experience, 10 rats per group were needed for testing the primary hypothesis regarding the changes of microvascular blood flow intensity. Continuous variables were presented as mean (SD). Mean differences at different time points were investigated using two-way repeated-measures ANOVA (with time and group factors), followed by the Tukey test for between-groups multiple comparison. Serum endocan levels and intestinal mucosa conductance were analyzed using one-way ANOVA followed by post hoc analysis using the Tukey test. All tests were two tailed. A P value less than 0.05 was considered significant. MFI scores and bacterial colony counts were presented as the median (interquartile range) and analyzed using the Mann–Whitney U test. While comparing bacterial colony counts, we defined three comparison pairs as follows: Sham versus lipopolysaccharide, Sham versus lipopolysaccharide + dexmedetomidine, and lipopolysaccharide versus lipopolysaccharide + dexmedetomidine. The Bonferroni correction was used, and a P value less than 0.017 was considered significant.

The heart rate was significantly increased in the lipopolysaccharide rats compared with the Sham controls, and dexmedetomidine attenuated lipopolysaccharide-induced tachycardia (table 1). The mean arterial pressure of endotoxemic rats was significantly lower than that of the Sham controls. No significant difference was observed between the lipopolysaccharide and lipopolysaccharide + dexmedetomidine groups (table 1), indicating that dexmedetomidine did not potentiate endotoxemia-related hypotension. The body temperature of endotoxemic rats was significantly higher than that of the Sham controls, and no significant difference was observed between the lipopolysaccharide and lipopolysaccharide + dexmedetomidine groups (table 1).

Absence of blood flow or microthrombosis in small vessels was observed in endotoxemic rats (fig. 2), and it resulted in lower PSVD and PPV. The PSVD and PPV of the mucosa, serosal muscular layer, and Peyer patch in the lipopolysaccharide group were significantly lower than those in the Sham and lipopolysaccharide + dexmedetomidine groups (table 2). The heterogeneity index of the mucosa was higher in the lipopolysaccharide group than in the lipopolysaccharide + dexmedetomidine group. At 240 min, the MFI scores of the lipopolysaccharide group were significantly lower than those of the lipopolysaccharide + dexmedetomidine group at the mucosa and serosal muscular layer, but not at the Peyer patch (table 2). The StO₂ values of the mucosa and Peyer patch in the lipopolysaccharide group were significantly lower than those in the Sham and lipopolysaccharide + dexmedetomidine groups; however, the difference in StO₂ value at the serosal muscular layer was not significant between the lipopolysaccharide and lipopolysaccharide + dexmedetomidine groups (table 3). At 240 min, the percentage changes of blood flux from the baseline were higher in the lipopolysaccharide group than in the lipopolysaccharide + dexmedetomidine group (mucosa: −33[16]% vs. −11[9]%, P = 0.001; serosal muscular layer: −16[10]% vs. −2[9]%, P = 0.021; and Peyer patch: −40[13]% vs. −20[11]%, P = 0.011) (fig. 3). The serum endocan levels of the lipopolysaccharide group were higher than those of the Sham and lipopolysaccharide + dexmedetomidine groups (table 4).

The intestinal tissue conductance of the lipopolysaccharide group was higher than that of the Sham controls, whereas the conductance values of the lipopolysaccharide + dexmedetomidine group were not significantly different from that of the Sham controls (table 4). The levels of mucosal-to-serosal flux of dextran-FITC (MW = 3,000 Da) were higher in the lipopolysaccharide group than in the Sham controls (fig. 4). However, the dextran flux was not reduced in the lipopolysaccharide + dexmedetomidine group (fig. 4). The plasma gadodiamide concentration was eight times higher in the lipopolysaccharide group (0.134 ± 0.112 μM) compared to sham group (0.016 ± 0.003 μM) after 240 min but did not reach statistical significance. A trend of decrease was in the plasma gadodiamide concentration of lipopolysaccharide + dexmedetomidine group (0.034 ± 0.010 μM) compared to lipopolysaccharide group (fig. 5). We next investigated alteration of tight junction molecules in rat intestinal mucosa. Western blotting results showed a significant reduction of the ratio of cleaved to full-length occludin in the lipopolysaccharide + dexmedetomidine group compared to the lipopolysaccharide group, suggesting amelioration of tight junctional damage (fig. 6). Lastly, the numbers of liver and spleen bacterial colonies of endotoxemic rats were statistically higher than those of Sham controls. Moreover, the number of spleen bacterial colonies in the lipopolysaccharide + dexmedetomidine group was significantly decreased compared to the lipopolysaccharide group (table 4).

The cell death levels in rat intestinal mucosa were determined by Western blots of cleaved caspase-3 (a hallmark of cell apoptosis). The ratio of cleaved to full-length caspase-3 was statistically higher in the lipopolysaccharide group than in the Sham controls (fig. 6). The ratio of cleaved to full-length caspase-3 in the lipopolysaccharide + dexmedetomidine group was not significantly different from that of Sham controls.

In this study, we provide evidence that dexmedetomidine can improve intestinal microcirculatory dysfunction, reduce endothelial dysfunction and excessive tachycardia, and attenuate intestinal epithelial cell death, tight junction damage, and intestinal bacterial translocation to the spleen in endotoxemic rats. Administering dexmedetomidine alleviates endotoxemia-related intestinal microcirculatory dysfunction, as evidenced by the higher PSVD values, blood flux, and MFI scores. In addition, dexmedetomidine reduced serum endocan level, suggesting its protective action against endothelial dysfunction. Moreover, our data demonstrated that lipopolysaccharide-induced mucosal cell apoptosis was attenuated by dexmedetomidine in rats, in association with reduction of occludin cleavage and bacterial influx into spleen. These findings suggest that dexmedetomidine, by improving microcirculation, may attenuate lipopolysaccharide-induced epithelial cell death as a means to protect against gut barrier dysfunction. The hypothesis and results of this study are summarized in figure 7.

Two major mechanisms support the finding that dexmedetomidine preserves the microcirculation in endotoxemic rats. First, the percentage changes of blood flux from the baseline at the three measured sites were lower in the lipopolysaccharide + dexmedetomidine group than in the lipopolysaccharide group. It evidences that the microcirculatory blood flow was higher in the lipopolysaccharide + dexmedetomidine group than in the lipopolysaccharide group. Second, PSVD and PPV were lower in the lipopolysaccharide group than in the lipopolysaccharide + dexmedetomidine group. It represented that more microthrombosis formation, absence of blood flow, or intermittent flow was noted in the small vessels of the lipopolysaccharide group. Excessive inflammation,^45,46  leukocyte adherence,^47–49  and endothelial cell injury^50,51  alter coagulation and fibrinolysis during endotoxemia or sepsis, and these alterations may cause disseminated intravascular coagulation. The antiinflammatory effects of dexmedetomidine were proven by several studies^5,7,9–11 ; moreover, Miranda et al.⁵²  demonstrated that dexmedetomidine reduces lipopolysaccharide-induced leukocyte adherence and leukocyte–endothelial interactions. Furthermore, in our study, the lower serum endocan level in the lipopolysaccharide + dexmedetomidine group than in the lipopolysaccharide group suggests that dexmedetomidine may reduce the severity of endothelial dysfunction in endotoxemic rats.

We found for the first time that dexmedetomidine can attenuate lipopolysaccharide-induced epithelial cell death as a means to protect against gut barrier dysfunction. The evidence is that the ratio of cleaved to full-length caspase-3 was higher in the lipopolysaccharide group than in the Sham controls. The ratio in the lipopolysaccharide + dexmedetomidine group was not significantly different from that of Sham controls. Moreover, there was a significant reduction in the ratio of cleaved to full-length occludin in the lipopolysaccharide + dexmedetomidine group compared to the lipopolysaccharide group, suggesting amelioration of tight junctional damage. The lipopolysaccharide-induced enteric bacterial translocation to spleen suggested impaired epithelial barrier integrity, which was reduced by dexmedetomidine. However, dexmedetomidine did not reduce the liver bacterial counts. We suspect that aside from limiting enteric bacterial translocation, dexmedetomidine might also reduce microbial counts in the bloodstream and spleen through enhancing microcirculation and infiltration of phagocytes to facilitate in bacterial clearance.⁵³ 

Bradycardia is a commonly reported adverse event occurring during dexmedetomidine administration, and there is a crucial concern that dexmedetomidine may impede compensated tachycardia and reduce cardiac output. We suggest that adequate tachycardia may increase cardiac output during early sepsis; nevertheless, excessive tachycardia may result in an opposite response. In our study, the antisympathetic effect of dexmedetomidine significantly attenuated excessive tachycardia, and microvascular blood flow was well preserved. Furthermore, Morelli et al.⁵⁴  reported that heart rate control using a titrated esmolol (a cardioselective β1-receptor blocker) infusion in septic shock patients was associated with the maintenance of stroke volume and preservation of microvascular blood flow. In addition to bradycardia, hypotension is another antisympathetic effect of dexmedetomidine and is the most commonly reported adverse event in critically ill patients.^55,56  In nonshock critically ill patients, hypotension is usually prevented by omitting the loading dose or resolved without intervention.^2,57  In septic shock patients, hypotension is usually prevented by omitting the loading dose, providing an adequate fluid supplement, or coadministering vasopressors. Another common clinical concern is that dexmedetomidine may potentiate the severity of hypotension. Our results suggest that the mean arterial pressure was not significantly different between endotoxemic rats with and those without dexmedetomidine infusion. Another serious concern is that dexmedetomidine may increase the infusion doses of vasopressors. Geloen et al.⁵⁸  demonstrated that dexmedetomidine increases the vasopressor response to norepinephrine in endotoxemic rats. To investigate the clinical effect of dexmedetomidine in sepsis patients, a multicenter clinical trial investigating 28-day mortality is ongoing in Japan (ClinicalTrials.gov Identifier: NCT01760967). Moreover, we are conducting a randomized controlled trial to investigate the effect of dexmedetomidine on sublingual microcirculation in severe sepsis patients (ClinicalTrials.gov Identifier: NCT02109965).

There are several limitations in this study. First, the results of this study are limited to the endotoxemic model. The effects of dexmedetomidine in the sepsis model require further investigation. Second, the observation time of this endotoxemic model is short. The effects of dexmedetomidine after 240 min require further investigation. Third, the investigation of intestinal microcirculation is focused on the terminal ileum. Further research is required to determine the effects of dexmedetomidine on the microcirculation of other regions of intestine. Fourth, the arterial and central venous blood gas analyses were absent in this study. Further investigation with blood gas analysis is required to determine the causes of observed difference in mucosal StO₂ between the lipopolysaccharide and the lipopolysaccharide + dexmedetomidine group.

In conclusion, dexmedetomidine attenuates intestinal microcirculatory alteration and attenuates elevation of endocan levels. In addition, dexmedetomidine reduces epithelial cell death, tight junctional damage, and enteric bacterial translocation to the spleen. We suggest that these benefits of dexmedetomidine in endotoxemic rats may lead to the development of novel therapeutic strategies for septic patients.

The authors thank all the participants of National Taiwan University Hospital Center of Microcirculation Medical Research. The authors also thank Roger Lien, B.S.E. (MicroStar Instruments, Taipei, Taiwan, Republic of China), for providing technical assistance during full-field laser perfusion imager examinations.

This work was supported, in part, by a grant from the National Science Council (NSC 103-2314-B-002-116), Taipei, Taiwan, Republic of China, and National Taiwan University Hospital (NTUH 103-A125), Taipei, Taiwan, Republic of China.

The authors declare no competing interests.