Marie-Reine Losser

Systemic and renal macro- and microcirculatory responses to arginine vasopressin in endotoxic rabbits*

Objective:Arginine vasopressin is being used increasingly to treat vasodilatory hypotension, although little is known of its effects on regional perfusion. Arginine vasopressin hemodynamic effects in physiology are mainly mediated through the V1a receptor on blood vessels. To investigate this further, we studied the effect of arginine vasopressin on systemic and renal blood flow in anesthetized, ventilated rabbits given either intravenous saline or endotoxin, and the impact of blocking V1a receptors. Design:Prospective, randomized, controlled study. Setting:Animal research laboratory. Subjects:Male White New Zealand rabbits. Interventions:Measurement was made of mean arterial blood pressure, aortic and renal blood flow velocities (pulsed Doppler), and renal cortical and medullary flow (laser Doppler). Measurements and Main Results:In a first series of animals, incremental intravenous boluses of arginine vasopressin ranging from 1 to 1000 ng were administered 90 mins postendotoxin or saline. In control rabbits (n = 9), increasing doses of arginine vasopressin elevated mean arterial blood pressure but reduced both aortic and renal blood flow velocity and renal cortical flow (p < .05). In endotoxic animals (n = 6), arginine vasopressin produced a similar increase in mean arterial blood pressure although aortic flow was maintained while renal blood flow velocity increased, mostly in its diastolic component (p < .05). Pretreatment with the V1a receptor antagonist in a second series of animals blunted all the effects observed in both control (n = 5) and endotoxic (n = 6) animals, suggesting that arginine vasopressin acted mainly through V1a subtype in this early phase of sepsis. Conclusions:Preservation of renal blood flow with arginine vasopressin during endotoxemia, in particular to the cortex, suggests it could be a promising agent for hemodynamic support during septic shock.

Cardiac and Renal Effects of Levosimendan, Arginine Vasopressin, and Norepinephrine in Lipopolysaccharide-treated Rabbits

Background:Because sepsis-induced myocardial dysfunction related to sepsis is at least partially related to a decrease in cardiac myofilament response to calcium, the use of the new myofilament-calcium sensitizer, levosimendan, has been proposed. In addition, arginine vasopressin is increasingly proposed as a vasopressor in septic patients, although data on its effects on cardiac function are still scarce. The aim of the current study was to assess, invasively and noninvasively, whether levosimendan, arginine vasopressin, and norepinephrine, either alone or combined, may modify sepsis-induced myocardial dysfunction and renal hemodynamics. Methods:Thirty-six hours after lipopolysaccharide or saline administration, rabbits were studied either after slight sedation for echocardiography or after general anesthesia with sodium pentobarbital for the following measurements: aortic flow velocity and maximum acceleration of blood flow in the ascending aorta and renal macrocirculation and microcirculation. Results:Levosimendan improved, within 30 min of administration, both maximum acceleration of blood flow by 20 ± 12% (n = 8; P < 0.05) and left ventricular shortening fraction by a similar extent. Furthermore, low doses of arginine vasopressin markedly deteriorated cardiac function via an afterload-independent mechanism, even when animals were pretreated with levosimendan, whereas norepinephrine showed no detrimental effects on cardiac function. The study also showed that norepinephrine often improved renal medullary blood flow, whereas arginine vasopressin consistently decreased it. Conclusion:Levosimendan and norepinephrine both exert beneficial effects in endotoxemic animals and should be further explored in human sepsis trials.

The Postoperative Effects of Halothane Versus Isoflurane on Hepatic Artery and Portal Vein Blood Flow in Humans

Animal studies have shown that halothane decreases total hepatic blood flow (THBF) by reducing both arterial (HABF) and portal (PVBF) inflow, whereas isoflurane appears to preserve them. In this study we assessed the effect of halothane and isoflurane on HABF and PVBF in surgical patients by using the pulsed Doppler technique. A validation study was conducted in six cynomolgus monkeys to compare the values of THBF obtained by the pulsed Doppler and indocyanine green clearance methods. Subsequently, six patients (ASA status I and II) undergoing elective open cholecystectomy were studied after surgery by using implanted pulsed Doppler probes. THBF and liver flow partition were compared during 1% halothane and 1.5% isoflurane (end-tidal concentrations). In the animal study, there was good agreement between the techniques (Bland and Altmann representation). In flunitrazepam-anesthetized patients, THBF was 1120 ± 284 mL/min. Compared with this baseline and for a similar mean arterial blood pressure decrease (10%), THBF was maintained with isoflurane, whereas it decreased by 36% (P < 0.05) under halothane. With isoflurane, PVBF increased (25%;P = 0.067) with a maintained HABF. With halothane, both PVBF (−44%;P < 0.05) and HABF (−20%;P < 0.05) were reduced. Halothane acted mainly as a vasoconstrictor of the hepatic circulation, whereas isoflurane was a vasodilator, confirming the beneficial effect of isoflurane on hepatic oxygen supply.

Intravenous Almitrine Bismesylate Reversibly Induces Lactic Acidosis and Hepatic Dysfunction in Patients with Acute Lung Injury

Background Intravenous almitrine, which augments hypoxic pulmonary vasoconstriction, is used for short‐term improvement of arterial oxygenation. However, recent research has suggested a potentially harmful effect on lactate metabolism and hepatic function. Methods Arterial oxygenation, hemodynamic parameters, plasma lactate, and hepatic function were monitored prospectively in 25 patients with acute lung injury (defined as a ratio of arterial oxygen pressure to inspiratory oxygen fraction <or= to 150 mmHg) who where treated with intravenous almitrine. In 21 of 25 patients, acute lung injury was related to primary lung lesions, including pneumonia, postcardiosurgical atelectasis, and lung contusions. Results Intravenous almitrine increased the ratio of arterial oxygen pressure to inspiratory oxygen fraction from 93 +/− 33 mmHg to 207 +/− 107 mmHg (mean +/− SD). In eight patients (three men), the plasma lactate concentration increased by an average of +3.5 +/− 1.8 mM, and the pH and bicarbonate concentration both decreased during the first 24 h of treatment. In this group of patients, the total bilirubin concentration was elevated before almitrine administration, and the results of other hepatic function tests, such as aspartate aminotransferase, alanine aminotransferase, and prothrombin time, were altered by almitrine administration. Therefore, intravenous almitrine was discontinued. Lactic acidosis and hepatic dysfunction improved. In the other 17 patients (14 men), the plasma lactate concentration and the hepatic function tests remained unaltered during intravenous almitrine therapy for >60 h. Univariate and multivariate analyses revealed that an abnormal plasma concentration of total bilirubin before almitrine administration and female gender were the two factors significantly linked with lactic acidosis during almitrine infusion. Conclusions This study confirms that intravenous almitrine greatly improves arterial oxygenation in patients with acute lung injury but may also induce lactic acidosis and hepatic dysfunction. The coexistence of lactic acidosis and hepatic dysfunction in the same patients strongly suggests that the liver is the primary source of intravenous almitrine‐induced lactic acidosis.

Nitric oxide involvement in the hemodynamic response to fluid resuscitation in endotoxic shock in rats.

Objective:Fluid loading is an essential part of cardiovascular resuscitation in septic shock. We hypothesized that fluid administration increases blood flow velocity and thus endothelial shear stress, causing the release of nitric oxide by the vascular endothelium. Because of endothelial dysfunction in sepsis, this mechanism would be less effective in septic animals. Fluid loading may have different effects in septic compared with control animals. Design:Prospective, randomized, controlled study. Setting:Animal research laboratory. Subjects:Male Sprague–Dawley rats. Interventions:We tested the involvement of nitric oxide in the fluid-induced cardiovascular response after administration of lipopolysaccharide (5 mg/kg, n = 10) or vehicle (control, n = 10) in rats subsequently randomized after 165 mins to receive l-NG-nitroarginine (7.5 mg/kg) or saline (n = 5 in each group). At 180 mins, all animals received hydroxyethyl starch (fluid loading, 15 mL/kg in 15 mins). Reversal of l-NG-nitroarginine was studied with an intravenous bolus of l-arginine (300 mg/kg). Measurements and Main Results:Lipopolysaccharide injection induced a hypokinetic shock (low blood pressure: −30% ± 9%, p < .05), low cardiac output (aortic pulsed-Doppler probe: −20% ± 8, p < .05), and unchanged systemic conductance, which turned into a hyperkinetic shock by fluid loading. Pretreatment with l-NG-nitroarginine totally abolished this fluid loading–induced vasodilation in control rats but only partially in lipopolysaccharide-treated rats, suggesting an altered endothelial response after lipopolysaccharide injection. Maximal aortic blood flow acceleration was used as an index of left ventricular systolic function. The improvement of maximal aortic blood flow acceleration observed during fluid loading in lipopolysaccharide-treated or control animals was blunted by l-NG-nitroarginine pretreatment, suggesting the involvement of nitric oxide in the myocardial response to fluid loading. Conclusions:These results suggest that the endothelium participates in the hemodynamic response to fluid loading in control rats, but less in rats with septic shock, secondary to an altered nitric oxide–dependent vasodilation.

Impact of fluid resuscitation with hypertonic-hydroxyethyl starch versus lactated ringer on hemorheology and microcirculation in hemorrhagic shock.

The choice of volume expander for fluid resuscitation in hemorrhagic shock is still debated. Changes in plasma viscosity (PV) are barely investigated while PV modulates functional capillary density, microcirculation and organ function. The present study evaluated the impact of 2 strategies of fluid resuscitation in hemorrhagic shock in pigs. Ten pigs were subjected to hemorrhagic shock and randomly assigned to a low viscosity fluid regimen (Lactated Ringers, LR) group or a high viscosity regimen (hypertonic-hydroxyethyl starch, HES) for volume resuscitation. Sublingual microcirculatory flow and tissue oxygen tension were assessed together with macro- and microcirculatory, biochemical and rheological variables at baseline, 30 minutes after hemorrhagic shock, immediately after reaching resuscitation endpoints (R-0), and 60 minutes after resuscitation (R-60). PV decreased similarly in both groups following resuscitation (from 1.36 [1.32-1.38] to 1.21 [1.21-1.23] for LR, and from 1.32 [1.31-1.32] to 1.20 [1.17-1.21] mPa.s for HES). No differences were found between the groups for other rheological variables, microcirculatory flow or tissue oxygen tension at R-0 and R-60. Despite a 6-fold difference in the volumes required to achieve blood flow endpoints, commercially available volume expanders had similar effects on rheological and microcirculatory variables, irrespective of their viscosity. Our findings are consistent with the absence of clinically relevant differences between crystalloid and colloid resuscitation of hemorrhagic shock.

Cerebral oximetry assessed by near-infrared spectrometry during preeclampsia: an observational study: impact of magnesium sulfate administration.

Objectives: To determine the regional cerebral oxygen saturation of hemoglobin (rcSO2) in severe preeclamptic parturients exhibiting neurologic symptoms compared with healthy pregnant women (control) and to describe the effects of MgSO4 infusion on rcSO2 and cerebral and systemic hemodynamic variables. Design: Prospective, observational study. Setting: Obstetric critical care unit in a university-affiliated hospital. Patients: Twenty severe preeclamptic parturients presenting with neurologic signs before any administration of MgSO4, and 20 control parturients. Intervention: Infusion of MgSO4 in severe preeclamptic patients. Measurements and Main Results: We measured rcSO2 using near-infrared spectroscopy, blood flow velocities of the middle cerebral artery, and cardiac output at baseline, 5 minutes, 1 hour, and 6 hours after the MgSO4 bolus (4 g), followed by continuous MgSO4 infusion (1 g/h). These measurements were also obtained in 20 control parturients at baseline and 6 hours. Baseline rcSO2 was significantly lower in the severe preeclamptic group: 61% (56–69) vs 66% (63–71) (p = 0.037). At inclusion, blood pressures were significantly higher in the severe preeclamptic group compared with the control group, whereas cardiac output and transcranial Doppler readings were similar. Five minutes after the MgSO4 bolus infusion, a median increase of 8.6% (3.2–18.1) in rcSO2 was observed (p = 0.007), reaching values of the control group that were maintained up to 6 hours. Blood pressures and systolic velocities of the middle cerebral artery significantly decreased (p < 0.01) after the MgSO4 bolus, whereas cardiac output did not change. The percentage increase in rcSO2 was negatively correlated to the mean blood pressure (r2 = 0.60, p < 0.0001). Conclusion: Cerebral oxygenation impairment can be detected by near-infrared spectroscopy monitoring in severe preeclamptic parturients. These results suggested the presence of disorders in cerebral microcirculation and/or changes in cerebral oxygenation. MgSO4 infusion in patients with severe preeclampsia restored rcSO2 to control levels with no systemic side effects. Further studies are needed to confirm the usefulness of near-infrared spectroscopy monitoring in patients with preeclampsia and to assess the action of other antihypertensive therapies on rcSO2.

Inhaled Nitric Oxide Fraction is Influenced by Both the Site and the Mode of Administration

Objective. Inhaled nitric oxide (NO) can be delivered continuouslyor sequentially (= during inspiration) at different locations of theventilation circuit. We have tested the influence of locations, modes of NOadministration and the ratio of the inspiratory time over the respiratorycycle time (I/I + E ratio) on the accuracy of NO fractions, delivered by 2devices: Opti-NO and Flowmeter. Methods.We used a simplified lungmodel consisting of a ventilation circuit with a Y piece, a tracheal tube, a150 ml dead-space volume and a 5 liter balloon. Three fractions (3, 6, 9 ppm)were administered continuously or sequentially, in controlled volume, in 4different sites on the inspiratory branch above the Y piece:i) justafter the water trap, ii) just before the Y piece; below the Y piece:iii) just after the Y piece, iv) into the endotracheal tube. In addition,different I/I + E ratios (25, 33, 50, 80%) were studied. The delivered NOfractions were measured in the balloon by chemiluminescence (CLD 700,Ecophysics). A linear regression analysis was used to test the relationshipbetween administered and measured NO fractions for the 3 fractions (3, 6 and9 ppm) in sequential and continuous modes. Intercept values were compared tozero and slopes to the identity line. Results.When NO was administeredin the continuous mode upstream the Y piece, NO fractions measured in theballoon corresponded to the administered fractions. In contrast, below the Ypiece, the measured NO fractions were significantly lower than theadministered NO fractions. In the sequential mode, above and below the Ypiece, the delivered NO fractions were within the manufacturers range.Conclusions.For the continuous NO delivery, locations above the Ypiece are mandatory. However, locations below the Y piece imposes a sequentialsystem, which can also be used for the sites located above the Y piece.

Modification of the hepatic hemodynamic response to acute changes in PaCO2 by nitric oxide synthase inhibition in rabbits.

BACKGROUND: Hypercapnia has been reported to modify liver circulation. The vascular regulations implicated in this response remain partly unknown. METHODS: Using anesthetized and ventilated rabbits, we designed this study to evaluate the hepatic artery and portal vein blood flow velocity adjustments (20 MHz pulsed Doppler) after changes in Paco2 (by varying the inspiratory fraction of CO2) and to assess the proper role of pH, independent of Paco2 changes, the role of portal vein CO2, and the effect of nitric oxide synthase inhibition on CO2-induced modifications of hepatic hemodynamics. RESULTS: Increasing Paco2 from 30.9 ± 5 mm Hg to 77 ± 11 mm Hg increased arterial blood pressure by 20% (P < 0.01) and hepatic artery blood flow velocity by 90% (P < 0.05) and decreased aortic blood flow velocity by 15% and portal vein blood flow velocity by 40% (both P < 0.05). Changes in pH (1 mL of 0.1 N hydrochloric acid infusion) or isolated changes in portal vein CO2 at constant Paco2 induced by CO2 insufflation in an open abdomen had no effect on hepatic hemodynamics. Pretreatment with a nitric oxide synthase inhibitor, N&ohgr;-nitro-l-arginine (2.5 mg/kg), blunted the systemic response to hypercapnia, whereas the portal modifications persisted, with a largely attenuated hepatic artery blood flow increase. CONCLUSIONS: CO2 per se acts on hepatic blood flow by its systemic effect, probably via chemoreflexes. Nitric oxide does not mediate hepatosplanchnic hemodynamic modifications to acute changes in Paco2 but may play a permissive role by regulating the amplitude of hepatic vascular response.

Immediate postoperative plasma neutrophil gelatinase-associated lipocalin to predict acute kidney injury after major open abdominal aortic surgery: A prospective observational study

INTRODUCTION Plasma neutrophil gelatinase-associated lipocalin (pNGAL) has been used as a biomarker in acute kidney injury (AKI). AKI is a common postoperative complication of aortic surgery. We sought to evaluate the performance of the immediately postoperative pNGAL level in comparison with the serum creatinine (SCr) level in predicting AKI and the need for renal replacement therapy (RRT). PATIENTS AND METHODS Prospective non-interventional study in a university hospital. Fifty patients undergoing elective or emergent major intra-abdominal aortic surgery were included. Comparisons between groups of patients with or without postoperative AKI, according to KDIGO staging, were made. Performance of NGAL was determined by examining the area under receiver operating characteristic (AUROC) curve. RESULTS The incidence of AKI was 36%. At H+2, pNGAL values in AKI and non-AKI patients, respectively, were 221 [133-278] versus 50 [50-90] ng/mL (P<0.0001), and SCr values were 115 [96-178] versus 90 [72-99] μmol/L (P<0.0008). The AUROC of pNGAL for prediction of AKI was 0.90 (95% CI: 0.81-0.98) with an optimal cutoff of 112ng/mL, a sensitivity of 83%, specificity of 84%, and positive and negative predictive values of 75% and 90%, respectively. SCr produced an AUROC curve of 0.79 (0.65-0.92) at a cutoff of 110μmol/L. The diagnostic performance of pNGAL was significantly better than that of SCr (P=0.039). PNGAL at H+2 better predicted the RRT requirement [0.96 (0.90-1.0)] compared to SCr [0.86 (0.73-0.98)], but this difference was not statistically significant. CONCLUSIONS A 2-hour postoperative determination of pNGAL outperformed SCr level in predicting postoperative AKI after major aortic surgery.