Search

Purpose of review This review presents important pathophysiological alterations associated with impaired liver function and discusses protective perioperative strategies and the various anaesthetic agents recommended. Recent findings Perioperative liver impairment is a serious complication of anaesthesia and surgery. Unfortunately, clinicians are provided with only crude macrohaemodynamic monitoring devices to optimize their therapy. Technical improvements have revealed some complex mysteries of perioperative microcirculatory alterations and have disclosed a large heterogeneity between different vascular beds. The present review will critically discuss current clinical concepts of optimizing global haemodynamic variables and the often contrasting effects of vasoactive agents on the microcirculatory nutritional blood flow. Finally, promising protective experimental interventions of pharmacological or ischaemic preconditioning are presented and their often disillusioning transition into recent clinical trials is highlighted. Summary Targeted perioperative liver protection still lacks adequate monitoring tools and is currently based on optimization of global haemodynamic variables. While there is currently no evidence suggesting a positive effect of ischaemic preconditioning, promising experimental results of pharmacological preconditioning and therapeutic hypothermia require further evaluation in larger randomized clinical trials.

Background:Adequate gastrointestinal mucosal oxygenation is regarded to be crucial in the prevention and therapy of critical illness. Epinephrine and norepinephrine are used for perioperative haemodynamic support. However, their per se effects on gastromucosal haemoglobin oxygenation (mu HbO(2)) remain unclear. Moreover, respective effects of epinephrine and norepinephrine may be affected by the type of underlying anaesthesia. Thus, we studied the effects of epinephrine and norepinephrine during anaesthesia with sevoflurane or propofol on regional gastromucosal mu HbO(2) and systemic O(2)-derived variables.Methods:In a double-randomized cross-over study, chronically instrumented dogs (n=6 per group) were anaesthetized randomly with sevoflurane or propofol, ventilated, and then randomly received either epinephrine or norepinephrine (0, 0.05, 0.1, and 0.2 mu g kg(-1) min(-1)). We measured gastromucosal mu HbO(2), systemic haemodynamics, and O(2)-derived variables.Results:During sevoflurane anaesthesia, norepinephrine markedly increased mu HbO(2) (P Conclusions:The effects of epinephrine and norepinephrine depended on the type of anaesthesia. In addition, regional effects (i.e. mu HbO(2)) were not predictable from systemic effects (i.e. DO(2)).

Microcirculatory dysfunction, reduced oxygen supply and thus impaired hepatosplanchnic organ function, play a pivotal role in the clinical manifestation of sepsis and septic shock. Early correction of the perfusion mismatch is essential to maintain end-organ and gut mu-cosa barrier function and thus to prevent bacterial translocation, bacteraemia and possible multiple organ failure. This review will outline the clinical tightrope-walk optimizing hepa-tosplanchnic oxygen supply while maintaining adequate end-organ perfusion pressure and highlight the diagnostic challenges and limitations faced in clinical practice.

Objective: To evaluate the impact of clonidine on mucosal red cell flux during baseline sedation with propofol or sevoflurane, respectively. Materials and Methods:Six healthy, chronically instrumented dogs for the measurement of cardiac output (CO) were repeatedly studied. During baseline sedation with either propofol (15 mg kg -1 h -1) or sevoflurane (1.5 MAC), local tissue cell flux was assessed using laser Doppler flowmetry at the enoral mucosa. After baseline measurements, a bolus of clonidine (2.0 μ g/kg) was infused within 1 min. Data are presented as mean ± SEM; Statistics: ANOVA, Scheffé's post hoc test, p < 0.05. Results:Clonidine significantly reduced CO from 75 ± 4 and 75 ± 6 ml kg -1 min -1 (sedation with propofol or sevoflurane, respectively) to 40 ± 3 and 49 ± 5 ml kg -1 min -1 , however, with almost complete recovery to baseline after 30 min (70 ± 4 and 71 ± 6 ml kg -1 min -1 , NS from baseline). Similarly, clonidine decreased mucosal red cell flux by 44 ± 8% and 54 ± 4%. However, mucosal perfusion did not return to baseline (-25 ± 5% and -27 ± 3%). Conclusions:In spite of the rapid return to baseline in systemic perfusion, the mucosal red cell flux of the enoral mucosa remained markedly reduced after a single bolus of clonidine. Given the crucial role of preserved microcirculatory perfusion for an intact mucosal barrier function, our data suggest that clonidine might impair this important mechanism to prevent the translocation of bacteria and endotoxins into the systemic circulation.

To clarify the effects of hypercapnia (increased PaCO2) on gastric mucosal oxygenation during anaesthesia in dogs.Prospective, controlled animal study.Experimental research laboratory of an university hospital.Six chronically instrumented dogs.Dogs were anaesthetized (sevoflurane 1.5 MAC), mechanically ventilated (etCO2 = 35 mmHg) and randomly assigned to the following protocols: in a first series, ventilation was adjusted to increase etCO(2) to 45, 55, 65 and 70 mmHg. In a second series, animals were ventilated to achieve 70 mmHg of etCO2, which was maintained for 120 min to test if effects are transient or prolonged and to achieve a similar time course in both protocols.Gastric mucosal oxygenation (microHbO2) was assessed continuously by tissue reflectance spectrophotometry. Mean arterial blood pressure (MAP) and cardiac output (CO) were continuously measured. Blood was sampled for blood gas analysis and lactate concentration. Hypercapnia increased gastric mucosal oxygenation concentration dependently from 48 +/- 6% (35 mmHg etCO2) to 51 +/- 4, 54 +/- 5, 56 +/- 5 and 59 +/- 3% (etCO2 45, 55, 65 and 70 mmHg, respectively). This reflected changes in CO (68 +/- 16, 74 +/- 16, 82 +/- 12, 91 +/- 11 and 97 +/- 16 ml kg(-1) min(-1), respectively) and systemic oxygen delivery (10 +/- 2, 11 +/- 3, 13 +/- 2, 14 +/- 2 and 14 +/- 2 ml kg(-1) min(-1), respectively). These effects persisted for 2 h (microHbO2 53 +/- 6 vs. 64 +/- 4%, CO 73 +/- 16 vs. 92 +/- 15 ml kg(-1) min(-1), DO2 12 +/- 4 vs. 14 +/- 3 ml kg(-1) min(-1), etCO2 35 and 70 mmHg, respectively).Hypercapnia increased systemic and regional oxygenation. If this experimental finding may be transferred to the clinical setting, permissive hypercapnia might be used to augment the oxygenation of the splanchnic region, e.g. gastrointestinal mucosa.

To evaluate the impact of clonidine on mucosal red cell flux during baseline sedation with propofol or sevoflurane, respectively.Six healthy, chronically instrumented dogs for the measurement of cardiac output (CO) were repeatedly studied. During baseline sedation with either propofol (15 mg x kg(-1) x h(-1)) or sevoflurane (1.5 MAC), local tissue cell flux was assessed using laser Doppler flowmetry at the enoral mucosa. After baseline measurements, a bolus of clonidine (2.0 microg/kg) was infused within 1 min. Data are presented as mean +/- SEM;ANOVA, Scheffé's post hoc test, p0.05.Clonidine significantly reduced CO from 75 +/- 4 and 75 +/- 6 ml x kg(-1) x min(-1) (sedation with propofol or sevoflurane, respectively) to 40 +/- 3 and 49 +/- 5 ml x kg(-1) x min(-1), however, with almost complete recovery to baseline after 30 min (70 +/- 4 and 71 +/- 6 ml x kg(-1) x min(-1), NS from baseline). Similarly, clonidine decreased mucosal red cell flux by 44 +/- 8% and 54 +/- 4%. However, mucosal perfusion did not return to baseline (-25 +/- 5% and -27 +/- 3%).In spite of the rapid return to baseline in systemic perfusion, the mucosal red cell flux of the enoral mucosa remained markedly reduced after a single bolus of clonidine. Given the crucial role of preserved microcirculatory perfusion for an intact mucosal barrier function, our data suggest that clonidine might impair this important mechanism to prevent the translocation of bacteria and endotoxins into the systemic circulation.

Background and objective: During high epidural anaesthesia, endothelin only contributes minimally to blood pressure stabilization. This phenomenon could result from the inhibitory action of nitric oxide on the endothelin system. To clarify this, we studied the interaction between nitric oxide and endothelin during high epidural anaesthesia in conscious dogs, in comparison to the interaction of nitric oxide and vasopressin. Methods: Six animals were used in 45 individual experiments randomly arranged as follows: N-ω-nitro- arginine-methylester 0.3-10 mg kg-1 under physiological conditions or during high epidural anaesthesia (lidocaine 1%) and N-ω-nitro-arginine- methylester (L-NAME) 0.3-10 mg kg-1 after preceding endothelin (Tezosentan®) or vasopressin (β-mercapto-β,β-cyclo-penta- methylene-propionyl-O-Me-Tyr-Arg-vasopressin) receptor blockade under physiological conditions or during high epidural anaesthesia. During control experiments normal saline was injected either intravenously (n = 5) or into the epidural space (n = 4). Results: N-ω-nitro-arginine-methylester increased mean arterial pressure dose-dependently in all groups. However, this effect was substantially reduced in the presence of the endothelin receptor antagonist compared to N-ω-nitro-arginine-methylester alone, both under control conditions (7 ± 3 vs. 21 ± 3 mmHg; P < 0.05) and during high epidural anaesthesia (17 ± 3 vs. 30 ± 1 mmHg; P < 0.05). Blockade of vasopressin showed no similar relationship with N-ω-nitro- arginine-methylester. Conclusions: The diminished increase in mean arterial pressure after injection of N-ω-nitro-arginine-methylester only during endothelin receptor blockade indicates that endogenous nitric oxide inhibits the action of endothelin during high epidural anaesthesia and might thus explain the reduced efficacy of endothelin in maintaining blood pressure during high epidural anaesthesia.

Objective: The effect of levosimendan, a novel inotropic vasodilator (inodilator), on the microvascular gastric mucosal hemoglobin oxygenation (μHbO2) is unknown. A possible effect could thereby be selective for the splanchnic region or could primarily reflect changes in systemic oxygen transport (ḊO2) and/or oxygen consumption (V̇O2). We compared systemic and regional effects of levosimendan with those of established inotropes, milrinone and dobutamine. Design: Laboratory experiment. Setting: University animal research laboratory of experimental anesthesiology. Subjects: Chronically instrumented dogs with flow probes for cardiac output measurement. Interventions: Anesthetized, mechanically ventilated dogs (each group n = 6) on different days randomly received levosimendan (10 μg·kg-1, followed by four infusion steps: 0.125-1.0 μg·kg-1·min-1), milrinone (5.0 μg·kg-1, followed by 1.25-10 μg·kg -1·min-1), or dobutamine (2.5-10.0 μg·kg-1·min-1). Since these drugs may modify regional or systemic responses to fluid load, an additional predefined volume challenge was subsequently performed with hydroxyethyl starch 6% (10 mL·kg-1). Measurements and Main Results: We measured μHbO2 (reflectance spectrophotometry), ḊO2, V̇O2, and systemic hemodynamics. Levosimendan significantly increased μHbO2 from baseline (∼55% for all groups) to 64 ± 4% and further to 69 ± 2% with volume challenge (mean ± SEM). At the systemic level, levosimendan alone only slightly increased ḊO2 at a stable V̇O2. Milrinone elicited similar systemic effects (ḊO2, V̇O2, hemodynamics) but failed to increase μHbO2. Dobutamine, conversely, increased μHbO2 to a similar extent as levosimendan; however, this was accompanied by marked increases in ḊO2 and V̇O2. The gastric mucosa selectivity of these interventions, expressed as slope of the μHbO2/ḊO2 relation, was highest for levosimendan (+1.89 and +1.14, without and with volume challenge), compared with milrinone (+0.45 and + 0.47) and dobutamine (+0.48 and + 0.33). Conclusions: Levosimendan is superior to milrinone (no significant regional effects) and dobutamine (marked systemic effects) in increasing gastric mucosal oxygenation selectiveiy (i.e., at only moderately increased ḊO2 and stable V̇O 2). If our experimental data apply to the clinical setting, levosimendan may serve as an option to selectively increase gastrointestinal mucosa oxygenation in patients at risk to develop splanchnic ischemia.

Background. The effects of thoracic epidural anaesthesia (TEA) on gastric mucosal microvascular haemoglobin oxygenation (μHbO2) are unclear. At the splanchnic level, reduction of sympathetic tone may promote vasodilation and increase μHbO2. However, these splanchnic effects are counteracted by systemic effects of TEA (e.g., decreased cardiac output (CO) and mean arterial pressure (MAP)), thus making the net effect on μHbO2 difficult to predict. In this respect, effects of TEA on μHbO2 may differ between physiological and compromised circulatory conditions, and additionally may depend on adequate fluid resuscitation. Furthermore, TEA may alter the relationship between regional μHbO2 and systemic oxygen-transport (DO2). Methods. Chronically instrumented dogs (flow probes for CO measurement) were anaesthetized, their lungs ventilated and randomly received TEA with lidocaine (n=6) or epidural saline (controls, n=6). Animals were studied under physiological and compromised circulatory conditions (PEEP 10 cm H2O), both with and without fluid resuscitation. We measured gastric mucosal μHbO2 by reflectance spectrophotometry, systemic DO2, and systemic haemodynamics (CO, MAP). Results. Under physiological conditions, TEA preserved μHbO2 (47 (3)% and 49 (5)%, mean (SEM)) despite significantly decreasing DO2 (11.3 (0.8) to 10.0 (0.7) ml kg-1 min-1) and MAP (66 (2) to 59 (3) mm Hg). However, during compromised circulatory conditions, TEA aggravated the reduction in μHbO2 (to 32 (1)%), DO2 (to 6.7 (0.8) ml kg-1 min-1) and MAP (to 52 (4) mm Hg), compared with controls. During TEA, fluid resuscitation completely restored these variables. TEA preserved the correlation between μHbO2 and DO2, compared with controls. Conclusions. TEA maintains μHbO2 under physiological conditions, but aggravates the reduction of μHbO2 induced by cardiocirculatory depression, thereby preserving the relationship between gastric mucosal and systemic oxygenation.

Background. In this study we aimed to clarify the role of endothelin in arterial pressure regulation during anaesthesia with increasing concentrations of sevoflurane (1-3 MAC) and compare it with those of vasopressin and angiotensin. Methods. After an awake control period, on different days, six dogs underwent each of the following four interventions: sevoflurane anaesthesia alone (1-3 MAC), sevoflurane after block of either endothelin receptors using tezosentan (3 mg kg-1 followed by 3 mg kg-1 h-1), vasopressin V1a receptors using [d(CH2)5Tyr(Me2)]AVP (40 μg kg-1) or angiotensin receptors using losartan (6 mg kg-1 h-1). Plasma concentrations of endothelin, big endothelin, vasopressin and renin were measured. Effects of sevoflurane in the presence and absence of the respective receptor block were analysed and compared using analysis of variance for repeated measures (ANOVA followed by Fisher's PLSD (protected least significant difference) (P-1 at 3 MAC as arterial pressure decreased in all groups. Conclusions. At 1 MAC, angiotensin attenuated the decrease in arterial pressure during sevoflurane anaesthesia more than endothelin and vasopressin. However, at higher MAC only vasopressin was specifically activated to partly compensate for the arterial pressure decrease.

Background and objective: Desflurane has been shown to increase sympathetic activity and heart rate (HR) in a concentration-dependent manner. Nevertheless, desflurane, like all other volatile anaesthetics, increased HR in parallel to vagal inhibition in a previous study. Therefore, our hypothesis is that desflurane elicits tachycardia by vagal inhibition rather than by activation of the sympathetic nervous system. Methods: Six dogs were studied awake and during desflurane anaesthesia (1 and 2 MAC) alone, after pretreatment with propranolol (2 mg kg-1 followed by 1 mg kg-1 h -1), or after pre-treatment with atropine (0.1 mg kg-1 followed by 0.05 mg kg-1 h-1). The effects on HR and HR variability were compared by an analysis of variance (P < 0.05). HR variability was analysed in the frequency domain as power in the high-(0.15-0.5 Hz, vagal activity) and low-frequency range (0.04-0.15 Hz, sympathetic and vagal activity). Results: HR increased during 2 MAC of desflurane from about 60 (awake) to 118 ± 2 beats min-1 (mean ± SEM) in controls and to 106 ± 3 beats min-1 in dogs pre-treated with propranolol. In contrast, pretreatment with atropine increased HR from 64 ± 2 to 147 ± 5 beats min-1 (awake) and HR decreased to 120 ± 5 beats min-1 after adding desflurane. High-frequency power correlated inversely with HR (r2 = 0.95/0.93) during desflurane alone and in the presence of β-adrenoceptor blockade, with no significant difference between regression lines. There was no correlation between these variables during atropine/desflurane. Conclusions: The increase in HR elicited by desflurane mainly results from vagal inhibition and not from sympathetic activation.

Background Oxygen consumption ( V ˙ o 2) is rarely measured during anaesthesia, probably because of technical difficulties. Theoretically, oxygen delivery into a closed anaesthesia circuit ( V ˙ o 2-PF; PhysioFlex™ Draeger Medical Company, Germany) should measure V ˙ o 2. We aimed to measure V ˙ o 2-PF in vitro and in vivo. Methods Three sets of experiments were performed. V ˙ o 2-PF was assessed with five values of V ˙ o 2 (0–300 ml min−1) simulated by a calibrated lung model ( V ˙ o 2-Model) at five values of F I O2 (0.25–0.85). The time taken for V ˙ o 2-PF to respond to changes in V ˙ o 2-Model gave a measure of dynamic performance. In six healthy anaesthetized dogs we compared V ˙ o 2-PF with V ˙ o 2 measured by the Fick method ( V ˙ o 2-Fick) during ventilation with nine values of F I O2 (0.21–1.00). V ˙ o 2-PF and V ˙ o 2-Fick were also compared in three dogs when V ˙ o 2 was changed pharmacologically [102 ( sd 14), 121 (17) and 200 (57) ml min−1]. In patients during surgery, we measured V ˙ o 2-PF and V ˙ o 2-Fick simultaneously after induction of anaesthesia (n=21) and during surgery (n=17) (F I O2 0.3–0.5). Results Compared with V ˙ o 2-Model, V ˙ o 2-PF values varied from time to time so that averaging over 10 min is recommended. Furthermore, at an F I O2 >0.8, V ˙ o 2-PF always overestimated V ˙ o 2. With F I O2

Inhalation anaesthetics decrease heart rate in isolated hearts but mostly increase heart rate in the intact organism, although most inhibit sympathetic drive. Differences in the degree of increase in heart rate between agents may be related to differences in their vagolytic action. To test this hypothesis, we studied the effects of halothane (H), isoflurane (I), enflurane (E), sevoflurane (S) and desflurane (D) [1-3 MAC (minimum alveolar concentration)] on heart rate and heart rate variability (HRV) as a measure of cardiac vagal activity in seven dogs. HRV was analysed in the time domain as the standard deviation of the RR interval (SDNN) and in the frequency domain as power in the high-frequency (HF, 0.15-0.5 Hz) and low-frequency (LF, 0.04-0.15 Hz) ranges. Heart rate increased with anaesthetic concentration and there were corresponding decreases in SDNN, HF power and LF power. Heart rate increased most with D (+40 beats min(-1)), least with H (+8 beats min(-1)) and to an intermediate extent with S, I and E. SDNN and HF power, as measures of vagal activity, changed in the opposite direction and decreased in the same order as heart rate increased. However, SDNN and HF power correlated significantly with heart rate [r=-0.81 (0.04) and -0.81 (0.03) respectively] and were independent of the anaesthetic and its concentration (P0.05). Consistent with our hypothesis, these results suggest that differences between agents in the degree of increase in heart rate are explained by differences in their vagolytic action.

UNLABELLED: We studied whether endogenous endothelin, like endogenous vasopressin, helps to maintain blood pressure during high epidural anesthesia when efferent sympathetic drive is diminished. On different days, six awake dogs underwent each of the following five interventions: blockade of vasopressin V(1a) receptors using [d(CH(2))(5)Tyr(Me(2))]AVP, (40 microg/kg) or endothelin receptors using tezosentan (3 mg/kg followed by 3 mg. kg(-1). h(-1)) with or without epidural anesthesia (1% lidocaine, intraindividual dose did not differ between experiments), and epidural saline (n = 5). The effects of endothelin- or vasopressin-receptor blockade were analyzed (means +/- SEM) and compared by an analysis of variance for repeated measures (paired Student's t-test, alpha-adjusted, P < 0.05). Vasopressin-receptor blockade decreased blood pressure (10 +/- 2 mm Hg) only in the presence of epidural anesthesia, whereas endothelin-receptor blockade reduced blood pressure both in the presence and absence of epidural anesthesia (12 +/- 3 versus 10 +/- 1 mm Hg). During baseline and each intervention, plasma concentrations of vasopressin and big-endothelin were measured and compared by a Wilcoxon's rank sum test; P < 0.05. Vasopressin concentrations increased during epidural anesthesia and after additional endothelin receptor blockade, but big-endothelin concentrations remained unchanged during each intervention. We conclude that vasopressin acts as a reserve system, as it stabilizes blood pressure specifically during epidural anesthesia, whereas the unchanged concentrations of big-endothelin indicate that the endothelin system is not specifically activated to support blood pressure during epidural anesthesia. IMPLICATIONS: We studied in awake dogs whether endogenous endothelin, like endogenous vasopressin, helps to maintain blood pressure during resting conditions and epidural anesthesia. Only vasopressin was specifically activated to support blood pressure during epidural anesthesia, whereas endothelin supported blood pressure to the same extent during epidural anesthesia and during resting conditions.

Objective: To assess the accuracy and reproducibility of long-term implanted ultrasound transit-time flow probes for measuring cardiac output. Design: Prospective animal study. Settings: Animal research laboratory in a university department. Animals: Eleven anaesthetised dogs, 24–34 kg. Measurements and results: Flow probes (16–24 mm S-series, Transonic) were implanted around the pulmonary artery for a mean duration of 22 months (range 6–47 months). Comparisons (n = 147) were made between cardiac output thus obtained and that measured by the direct Fick principle using oxygen uptake (Deltatrac II Metabolic Monitor) and the arterial to mixed venous oxygen content difference measured by a galvanic cell (Lex-O2-Con-TL). Measurements were made either during baseline conditions or during pharmacologically altered cardiac output (range 22–180 ml · kg–1· min–1). Regardless of the intervention, the two methods yielded the same results in half of the dogs. In the others, however, cardiac output was underestimated by the flow probes by up to 38 % (probably because of non-perpendicular position of the probe towards the vessel). This difference was constant for the whole range of cardiac output studied and remained constant over the entire observation period for each individual dog, so that a correction factor was used. Thereafter, the mean difference between the two methods was –1.1 ml · kg–1· min–1 with a precision (SD) of 14.2 ml · kg–1· min–1 for all experiments. Conclusions: After in vivo calibration, ultrasound transit-time flow probes measure cardiac output precisely for several years, regardless of the intervention.

Background & Goal: To study the response to acute hypoxia in micro- and macrocirculatory variables during propofol- (PROPO) vs. sevoflurane- (SEVO) anesthesia. Materials and methods: In randomized cross over design, chronically instrumented dogs (24-32 kg, n=6 per group) were anesthetized with PROPO or SEVO (equi-anesthetic dosages), mechanically ventilated (FiO2 ~0.3) and underwent a hypoxemic episode (FiO2 ~0.1 for 15 min [1] ), followed by normoxic recovery (FiO2 ~0.3; 90 min). Microcirculatory perfusion of oral mucosa (µ-mucosa) and hindlimb skin (µ-skin) were measured by laser Doppler fluxmetry. Regional gastric intraluminal PCO2 (PgiCO2) was measured by tonometry. Macro-hemodynamics (cardiac output (CO), blood pressures), metabolism (O2-consumption (VO2); indirect alorimetry) and blood-derived variables (art. O2-content (CaO2), acid/base variables) were also measured. Statistics: Data are mean±sem for n=6 experiments per group, and compared by two-way-ANOVA, corrected for multiple comparisons (Bonferroni). A p< 0.05 was significant. Results and Discussion: During normoxia, both CO and CaO2 were similar (both N.S.) under PROPO and SEVO. In addition, the hypoxia model achieved comparable drops in CaO2 (N.S.) under both anesthesia types. However, hypoxia elicited markedly different responses between PROPO and SEVO: Hypoxia during PROPO significantly increased µ-mucosa and µ-skin, but decreased both during SEVO. After hypoxia under PROPO, PgiCO2 was preserved, whereas under SEVO, PgiCO2 significantly accumulated. During PROPO, hypoxia generally stimulated macro-hemodynamics (e.g., CO increased from 86±3 to 123±8 ml/kg/min), whereas in contrast during SEVO-hypoxia hemodynamics were unresponsive (CO, 72±8 vs. 71±6 ml/kg/min) or even depressed. On the metabolic level, VO2 was maintained during PROPO/hypoxia, but significantly decreased with SEVO/hypoxia. Conclusion: In this dog model, only PROPO permitted hemodynamic stimulation by acute hypoxia, resulting in (compensatory) increased micro- and macrocirculatory perfusion. In line, PROPO-hypoxia maintained VO2 , without aggravating anaerobiosis markers (lactate), and with preservation of PgiCO2. In contrast, SEVO failed to permit hemodynamic stimulation by hypoxia and failed to preserve PgiCO2. If our data can be transferred to the clinical setting, PROPO appears beneficial (vs. SEVO) to support the systemic and regional circulation during acute hypoxia. Reference: 1. Schwarte LA et al, Int Care Med 2011;37:701-10