Effect of patient-controlled epidural analgesia with and without automatic intermittent bolus on levobupivacaine consumption during labour: A single-centre prospective double-blinded randomised controlled study.

BACKGROUND: During labour, the effects of adding a programmed intermittent epidural bolus (PIEB) baseline analgesic regimen to patient-controlled epidural analgesia (PCEA) remain uncertain. METHODS: This single centre prospective double-blinded controlled study randomised nulliparous women over 35 weeks of gestational age in a PCEA + PIEB or PCEA only group. After an epidural analgesia catheter was inserted, a specific pump administered a solution of levobupivacaine 0.625 mg mL-1, sufentanil 0.25 µg mL-1, and clonidine 0.375 µg mL-1. In both groups the PCEA mode delivered an 8 mL bolus with a lockout period of 8 min. In the PCEA + PIEB group, women also received a programmed 8 mL bolus every 60 min. Additional bolus were allowed if required. The primary outcome was the hourly consumption of levobupivacaine from epidural catheter placement to new-born delivery. Secondary outcome were motor block, oxytocin use, sufentanil consumption, additional bolus required, instrumental vaginal delivery, unplanned caesarean section, pain during labour and women's satisfaction. RESULTS: Analysis included 162 and 155 women in the PCEA and PCEA + PIEB groups, respectively. The median [IQR] hourly consumption of levobupivacaine was significantly lower in the PCEA group (9.9 (7.8-12.4] mg h-1) as compared to the PCEA + PIEB group (11.2 [7.9-14.3] mg h-1; p = 0.046). The difference between medians was 1.3 mg h-1 95 % CI (0.1-2.9). There was no difference between groups for secondary outcomes. CONCLUSIONS: PCEA only modestly decreased the hourly consumption of local anaesthetic as compared to PCEA + PIEB but the difference was not clinically relevant.

Preventing the Long-term Effects of General Anesthesia on the Developing Brain: How Translational Research can Contribute.

In 2017, the Food and Drug Administration published a safety recommendation to limit the exposure to general anesthesia as much as possible below the age of three. Indeed, several preclinical and clinical studies have questioned the possible toxicity of general anesthesia on the developing brain. Since then, recent clinical studies tried to mitigate this alarming issue. What is true, what is false? Contrary to some perceptions, the debate is not over yet. Only stronger translational research will allow scientists to provide concrete answers to this public health issue. In this review, we will provide and discuss the more recent data in this field, including the point of view of preclinical researchers, neuropsychologists and pediatric anesthesiologists. Through translational research, preclinical researchers have more than ever a role to play to better understand and identify long-term effects of general anesthesia for pediatric surgery on brain development in order to minimize it.

Perioperative non-invasive haemodynamic optimisation using photoplethysmography: A randomised controlled trial and meta-analysis.

BACKGROUND: The present study aimed at evaluating the effect of a haemodynamic algorithm using SV maximisation by non-invasive photoplethysmography (interventional group) on the incidence of postoperative complications compared with a control group using intermittent mean arterial pressure. METHODS: The non-blinded parallel-group trial randomised low-risk patients undergoing colorectal surgery into either interventional group or control group. The primary outcome was the incidence of patients with at least one complication during the 30 days following surgery. The secondary outcomes were the total number of complications, the length of hospital stay and postoperative mortality. A meta-analysis of randomised trials comparing perioperative haemodynamic optimisation (interventional group) using photoplethysmography with control group was performed to assess the external validity. RESULTS: Among 160 randomised patients, 159 were analysed (80 and 79 in interventional and control groups, respectively). Demographic characteristics were similar in both groups. Postoperative complications occurred in 40 (50%) and 34 (43%) patients in the interventional and control groups, respectively (P=0.471). There were no significant differences between the two groups regarding the total number of complications (P=0.078), the hospital length of stay (P=0.960), or postoperative mortality (P=1.000). In the meta-analysis including 1089 patients in 7 randomised controlled studies, 203 (38%) and 221 (40%) patients suffered from at least one complication following surgery [risk ratio 0.89 (95% CI 0.68-1.17), P=0.407] in interventional and control groups, respectively. CONCLUSIONS: Based on the findings of the present study and meta-analysis, a haemodynamic perioperative algorithm using SV maximisation by non-invasive photoplethysmography cannot reduce postoperative morbidity.

Effect of Oral Carbohydrate Intake During Labor on the Rate of Instrumental Vaginal Delivery: A Multicenter, Randomized Controlled Trial.

BACKGROUND: Carbohydrate intake during physical exercise improves muscle performance and decreases fatigue. We hypothesized that carbohydrate intake during labor, which is a period of significant physical activity, can decrease the instrumental vaginal delivery rate. METHODS: In a multicenter, prospective, randomized, controlled trial, healthy adult pregnant women presenting with spontaneous labor were assigned to a "Carbohydrate" group (advised to drink 200 mL of apple or grape juice without pulp every 3 hours) or a "Fasting" group (water only). The primary outcome was the instrumental vaginal delivery rate. Secondary outcomes included duration of labor, rate of cesarean delivery, evaluation of maternal hunger, thirst, stress, fatigue, and overall feeling during labor by numeric rating scale (0 worst rating to 10 best rating), rate of vomiting, and hospital length of stay. Statistical analysis was performed on an intention-to-treat basis. The primary outcome was tested with the "Fasting" group as the reference group. The P values for secondary outcomes were adjusted for multiple comparisons. The differences between groups are reported with 99% confidence interval (CI). RESULTS: A total of 3984 women were analyzed (2014 in the Carbohydrate group and 1970 in the Fasting group). There was no difference in the rate of instrumental delivery between the Carbohydrate (21.0%) and the Fasting (22.4%) groups (difference, -1.4%; 99% CI, -4.9 to 2.2). No differences were found between the Carbohydrate and the Fasting groups for the duration of labor (difference, -7 minutes; 99% CI, -25 to 11), the rate of cesarean delivery (difference, -0.3%; 99% CI, -2.4 to 3.0), the rate of vomiting (difference, 2.8%; 99% CI, 0.2-5.7), the degree of self-reported fatigue (difference, 1; 99% CI, 0-2), self-reported hunger (difference, 0; 99% CI, -1 to 1), thirst (difference, 0; 99% CI, -1 to 1), stress (difference, 0; 99% CI, -1 to 1), overall feeling (difference, 0; 99% CI, 0-0), and the length of hospitalization (difference, 0; 99% CI, -1 to 0). CONCLUSIONS: Carbohydrate intake during labor did not modify the rate of instrumental vaginal delivery.

Comparison of pre-oxygenation using spontaneous breathing through face mask and high-flow nasal oxygen: A randomised controlled crossover study in healthy volunteers.

BACKGROUND: High-flow nasal oxygen (HFNO) therapy has been proposed for pre-oxygenation before intubation, but the end-tidal fraction of oxygen (ETO2) obtained remains unknown. OBJECTIVE(S): To compare the ETO2 following a 3 min pre-oxygenation with HFNO and face mask. SETTING: Operating room in a primary university hospital. DESIGN: A randomised crossover study. PARTICIPANTS: Fifty healthy volunteers. INTERVENTIONS: Participants were randomly pre-oxygenated through spontaneous breathing 100% oxygen in a face mask and with HFNO (mouth closed, heated and humidified gas flow at 60 l min). In the face mask group, the ETO2 was measured continuously. In the HFNO group, the nasal cannula was quickly exchanged with a face mask while the subject held their breath at end inspiration and the ETO2 was measured after a deep expiration. The protocol ended when ETO2 reached 90% or otherwise at 6 min. MAIN OUTCOME MEASURES: The primary endpoint was the ETO2 after 3 min of pre-oxygenation. Secondary endpoints were the proportion of participants with an ETO2 at least 90% and the time until the ETO2 at least 90%. RESULTS: The ETO2 after 3 min of pre-oxygenation was 89 (2) % and 77 (12) % in the face mask and HFNO groups [difference 12% (95% confidence interval, 95% CI: 8 to 15]; P < 0.001), respectively. After 3 min of pre-oxygenation, 54 and 4% (P < 0.001) of volunteers had an ETO2 at least 90% in the face-mask and HFNO groups, respectively. After 6 min of pre-oxygenation, 96 and 46% (P < 0.001) of volunteers had an ETO2 at least 90% in the face-mask and HFNO groups, respectively. In the face mask group, the hazard ratio to achieve an ETO2 of 90% was 5.3 (95% CI: 3.2 to 8.9; P < 0.001). CONCLUSION: Our study demonstrates that pre-oxygenation with HFNO is not a reliable method of pre-oxygenation before the induction of anaesthesia. TRIAL REGISTRATION: clinical trial NCT03399695.

Changes in stroke volume during an alveolar recruitment maneuvers through a stepwise increase in positive end expiratory pressure and transient continuous positive airway pressure in anesthetized patients. A prospective observational pilot study.

BACKGROUND AND AIMS: Recruitment maneuvers may be used during anesthesia as part of perioperative protective ventilation strategy. However, the hemodynamic effect of recruitment maneuvers remain poorly documented in this setting. MATERIAL AND METHODS: This was a prospective observational study performed in operating theatre including patients scheduled for major vascular surgery. Patients were monitored with invasive arterial pressure and esophageal doppler. After induction of general anesthesia, before surgery began, preload optimization based on stroke volume (SV) variation following fluid challenge was performed. Then, an alveolar recruitment maneuver (ARM) through stepwise increase in positive end expiratory pressure (PEEP) or continuous positive airway pressure (CPAP) was performed. Hemodynamic data were noted before, during, and after the alveolar recruitment maneuver. RESULTS: ARM through stepwise increase in PEEP and CPAP were applied in 22 and 14 preload independent patients, respectively. Relative changes in SV during ARMs were significantly greater in the ARMCPAP group (-39 ± 20%) as compared to the ARMPEEP group (-15 ± 22%; P = 0.002). The difference (95% CI) in relative decrease in SV between ARMCPAP and ARMPEEP groups was -24% (-38 to -9; P = 0.001). Changes in arterial pressure, cardiac index, pulse pressure variation, peak velocity, and corrected flow time measures were not different between groups. CONCLUSION: During anesthesia, in preload independent patients, ARMs through CPAP resulted in a significantly greater decrease in SV than stepwise increase in PEEP. During anesthesia, ARM should be used cautiously.

Assessment of changes in blood glucose concentration with intravascular microdialysis.

Blood glucose and its variability of is a major prognostic factor associated with morbidity. We hypothesized that intravenous microdialysis incorporated in a central venous catheter (CVC) would be interchangeable with changes in blood glucose measured by the reference method using a blood gas analyzer. Microdialysis and central venous blood glucose measurements were simultaneously recorded in high-risk cardiac surgical patients. The correlation between absolute values was determined by linear regression and the Bland-Altman test for repeated measurements was used to compare bias, precision, and limits of agreement. Changes in blood glucose measurement were evaluated by four-quadrant plot and trend interchangeability methods (TIM). In the 23 patients analyzed, the CVC was used as part of standard care with no complications. The correlation coefficient for absolute values (N = 99) was R = 0.91 (P < 0.001). The bias, precision and limits of agreement were - 9.1, 17.4 and - 43.2 to 24.9 mg/dL, respectively. The concordance rate for changes in blood glucose measurements (N = 77) was 85% with the four-quadrant plot. The TIM showed that 14 (18%) changes of blood glucose measurements were uninterpretable. Among the remaining 63 (82%) interpretable changes, 23 (37%) were interchangeable, 13 (20%) were in the gray zone, and 27 (43%) were not interchangeable. Microdialysis using a CVC appears to provide imprecise absolute blood glucose values with risk of insulin misuse. Moreover, only one third of changes in blood glucose measurements were interchangeable with the reference method using the TIM.

Comparison of the Mallampati Classification in Sitting and Supine Position to Predict Difficult Tracheal Intubation: A Prospective Observational Cohort Study.

BACKGROUND: The Mallampati classification (MLPT) is normally evaluated in the sitting position. However, many patients cannot be evaluated in the sitting position for medical reasons. Thus, we compared the MLPT in sitting and supine positions in predicting difficult tracheal intubation (DTI). We hypothesized that the diagnostic accuracy of the MLPT performed in sitting and supine positions would differ. METHODS: We performed a single-center prospective observational study in adult patients who received general anesthesia and orotracheal intubation for noncardiac surgery. During the preanesthesia consultation, the MLPT in the sitting position was recorded. The day of surgery, the MLPT in the supine position and the difficulty of intubation (DTI) were recorded by an independent observer. The diagnostic performance of the MLPT for the prediction of DTI was evaluated in the sitting and supine positions through the area under the receiver operating characteristic (ROC) curve. The performance of the Naguib score in predicting DTI was calculated with the MLPT in sitting and supine positions. RESULTS: Among the 3036 patients, 157 (5.1%) had DTI. The area under the ROC curve for the MLPT in supine position (0.82 [0.78-0.84]) was greater than that for the MLPT in the sitting position (0.70 [0.66-0.75]; P < .001). The relationship between the sitting and supine MLPTs was moderate (Spearman rank correlation coefficient: 0.50; P < .001). The area under ROC curve for predicting DTI by the Naguib score calculated with the supine MLPT (0.78 [95% confidence interval, 0.74-0.82]) was greater than that for the Naguib score calculated with MLPT in the sitting position (0.69 [95% confidence interval, 0.63-0.74)]; P < .001). CONCLUSIONS: The MLPT performed in the supine position is possibly superior to that performed in the sitting position for predicting difficult intubation in adults.

Sevoflurane-induced cardioprotection in coronary artery bypass graft surgery: Randomised trial with clinical and ex-vivo endpoints.

BACKGROUND: Myocardial ischaemia reperfusion injury following cardiac surgery with cardiopulmonary bypass (CPB) increases postoperative mortality. Setting techniques to protect the heart during this critical period therefore represents a considerable challenge. METHOD: A randomised controlled study in Caen University Hospital Centre, investigated whether the clinical cardio protective effects of administration sevoflurane before cardiopulmonary bypass during coronary artery bypass graft surgery (CABG) could translate into protected atrial trabeculae contractility against hypoxia-reoxygenation in vitro. Patients undergoing elective on-pump CABG surgery were allocated to receive either sevoflurane (n=24) or no halogenated volatile anaesthetic (n=21). MAIN OUTCOME MEASURES: the relationship between sevoflurane exposure before CPB and the incidence of major adverse cardiac events, with primary endpoint: the postoperative troponin I peak level, and secondary endpoints: the inotropic support, and the duration of stay in intensive unit and in-hospital stay were chosen as study endpoints. The right atrial was collected at the beginning of bypass surgery for the in vitro experimentation. Isometrically contracting isolated human right atrial trabeculae obtained from the two groups were exposed to 30-min hypoxia followed by 60-min reoxygenation. RESULTS: The patients receiving sevoflurane prior to aortic clamping significantly exhibited less cardiac Troponin I (1.39 [0.34-2.97] vs. 2.80 [2.54-3.64] ng·mL-1 in Control; P=0.03) and required a reduced inotropic drug support (P<0.001). Isolated trabeculae from patients receiving sevoflurane enhanced the recovery of force after reoxygenation compared to the Control group (79±5% vs. 53±8% of baseline in Control; P<0.001). CONCLUSIONS: Administration of sevoflurane before CPB induced cardioprotection in patients undergoing CABG and preconditioned human myocardium against hypoxia-reoxygenation in vitro.

Effectiveness of a blended learning course and flipped classroom in first year anaesthesia training.

BACKGROUND: Blended learning, which combines internet-based platform and lecturing, is used in anaesthesiology and critical care teaching. However, the benefits of this method remain unclear. METHODS: We conducted a prospective, multicentre, non-randomised work between 2007 and 2014 to study the effect of blended learning on the results of first year anaesthesia and critical care residents in comparison with traditional teaching. Blended learning was implemented in Rouen University Hospital in 2011 and residents affiliated to this university corresponded as the blended learning group. The primary outcome was the resident's results as measured with multiple-choice questions between blended learning and control groups after beginning blended learning (post-interventional stage). The secondary outcomes included residents' results between pre and post-interventional stages and homework's time. Moreover, comparison between control and blended learning group before beginning blended learning (pre-interventional stage) was performed. RESULTS: From 2007 to 2014, 308 residents were included. For the pre-interventional period, the mean score in the blended learning group (n=53) was 176 (CI 95% 163 to 188) whereas the mean score in the control group (n=106) was 167 (CI 95% 160 to 174) (no difference). For the post-interventional period, the mean score in blended learning group (n=54) was 232 on 300 (CI95% 227-237) whereas the mean score in the control group (n=95) is 215 (CI95% 209-220) (P<0.001). In the two groups, comparison between pre and post-interventional stages showed the increase of mean score, stronger for blended learning group (32% and 28% in blended learning and control group, P<0.05). The average time of homework in the blended learning group was 27h (CI 95% 18.2-35.8) and 10h in the control group (CI 95% 2-18) (P<0.05). CONCLUSIONS: This work suggests the positive effect of blended learning (associating internet-based learning and flipped classroom) on the anaesthesia and critical care residents' knowledge by increasing their homework's time.

Effect of sleep deprivation after a night shift duty on simulated crisis management by residents in anaesthesia. A randomised crossover study.

BACKGROUND: Sleep deprivation has been associated with an increased incidence of medical errors and can jeopardise patients' safety during medical crisis management. The aim of the study was to assess the effect of sleep deprivation on the management of simulated anaesthesia crisis by residents in anaesthesiology. METHODS: A randomised, comparative, monocentric crossover study involving 48 residents in anaesthesia was performed on a high fidelity patient simulator. Each resident was evaluated in a sleep-deprived state (deprived group, after a night shift duty) and control state (control group, after a night of sleep). Performance was assessed through points obtained during crisis scenario 1 (oesophageal intubation followed by anaphylactic shock) and scenario 2 (anaesthesia-related bronchospasm followed by ventricular tachycardia). Sleep periods were recorded by actigraphy. Two independent observers assessed the performances. The primary endpoint of the study was the score obtained for each scenario. RESULTS: Resident's crisis management performance is associated with sleep deprivation (scenario 1: control=39 [33-42] points vs. deprived=26 [19-40] points, P=0.02; scenario 2: control=21 [17-24] vs. deprived=14 [12-19], P=0.01). The main errors observed were: error in drug administration and dose, delay in identification of hypotension, and missing communication with the surgical team about situation. CONCLUSIONS: The present study showed that sleep deprivation is associated with impairment of performance to manage crisis situations by residents in anaesthesia.

Assessment of macro- and micro-oxygenation parameters during fractional fluid infusion: A pilot study.

PURPOSE: The main goal of this study was to assess whether maximal fluid infusion improves both oxygen delivery (DO2) and micro-circulatory parameters during hemodilution. The secondary objective was to assess the ability of baseline micro-circulatory parameters to predict oxygen consumption (VO2) response following fluid infusion. MATERIALS AND METHODS: In a postoperative cardiac ICU, patients received repeated fluid infusion until stroke volume (SV) was maximized. Before and after each fluid expansion, macro- (DO2, VO2) and micro-circulatory oxygenation parameters were recorded [central venous oxygen saturation (ScVO2), blood lactate, difference in veno-arterial carbon dioxide tension (P(v-a)CO2), somatic and cerebral oxygen saturation (rSO2)]. Patients were classified as VO2-Responders or VO2-Non-Responders according to an increase in VO2 above or below 15%, respectively. RESULTS: After maximal fluid infusion, all patients showed improved macro- and micro-circulatory oxygenation parameters, but VO2-Responders had lower values (especially for ScVO2 and cerebral rSO2). Only baseline ScVO2 and cerebral rSO2 were useful to predict the VO2 response to maximal fluid infusion (ROCAUC 0.80 (95% CI: 0.54-0.95, P=0.012) and 0.83 (95% CI: 0.57-0.96, P=0.001). CONCLUSIONS: Maximal fluid infusion improves macro- and micro-circulatory oxygenation parameters. For VO2-Responders, only ScVO2 and cerebral rSO2 could serve as markers of tissue hypoxia.

Middle cerebral artery blood flow velocity during beach chair position for shoulder surgery under general anesthesia.

OBJECTIVES: The goal of the present study was to examine changes of middle cerebral artery (VMCA) blood flow velocity in patients scheduled for shoulder surgery in beach chair position. DESIGN: Prospective observational study. SETTING: Operating room, shoulder surgery. PATIENTS: Fifty-three consecutive patients scheduled for shoulder surgery in beach chair position. INTERVENTIONS: Transcranial Doppler performed after induction of general anesthesia (baseline), after beach chair positioning (BC1), during surgery 20minutes (BC2), and after back to supine position before stopping anesthesia (supine). MEASUREMENTS: Mean arterial pressure (MAP), end-tidal CO2, and volatile anesthetic concentration and VMCA were recorded at baseline, BC1, BC2, and supine. Postoperative neurologic complications were searched. MAIN RESULTS: Beach chair position induced decrease in MAP (baseline: 73±10mm Hg vs lower MAP recorded: 61±10mm Hg; P<.0001) requiring vasopressors and fluid challenge in 44 patients (83%). There was a significant decrease in VMCA after beach chair positioning (BC1: 33±10cm/s vs baseline: 39±14cm/s; P=.001). The VMCA at baseline (39±2cm/s), BC2 (35±14cm/s), and supine (39±14cm/s) were not different. The minimal alveolar concentration of volatile anesthetics, end-tidal CO2, SpO2, and MAP were not different at baseline, BC1, BC2, and supine. CONCLUSION: Beach chair position resulted in transient decrease in MAP requiring fluid challenge and vasopressors and a moderate decrease in VMCA.

The role of age and comorbidities in postoperative outcome of mitral valve repair: A propensity-matched study.

The average age of patients undergoing mitral valve repair is increasing each year. This retrospective study aimed to compare postoperative complications of mitral valve repair (known to be especially high-risk) between 2 age groups: under and over the age of 80.Patients who underwent mitral valve repair were divided into 2 groups: group 1 (<80 years old) and group 2 (≥80 years old). Baseline characteristics, pre- and postoperative hemodynamic data, surgical characteristics, and postoperative follow-up data until hospital discharge were collected.A total of 308 patients were included: 264 in group 1 (age 63 ±â€Š13 years) and 44 in group 2 (age 83 ±â€Š2 years). Older patients had more comorbidities (atrial fibrillation, history of cardiac decompensation, systemic hypertension, pulmonary hypertension, and chronic kidney disease) and they presented more postoperative complications (50.0% vs 33.7%; P = 0.043), with a longer hospital stay (8.9 ±â€Š6.9 vs 6.6 ±â€Š4.6 days; P = 0.005). To assess the burden of age, a propensity score was awarded to postoperative complications. Active smoking, chronic pulmonary disease, chronic kidney disease, associated ischemic heart disease, obesity, and cardio pulmonary by-pass duration were described as independent risk factors. When matched on this propensity score, there was no difference in morbidity or mortality between group 1 and group 2.Older patients suffered more postoperative complications, which were related to their comorbidities and not only to their age.

Impact of a feedback device on chest compression quality during extended manikin CPR: a randomized crossover study.

PURPOSES: Chest compressions require physical effort leading to increased fatigue and rapid degradation in the quality of cardiopulmonary resuscitation overtime. Despite harmful effect of interrupting chest compressions, current guidelines recommend that rescuers switch every 2 minutes. The impact on the quality of chest compressions during extended cardiopulmonary resuscitation has yet to be assessed. BASIC PROCEDURES: We conducted randomized crossover study on manikin (ResusciAnne; Laerdal). After randomization, 60 professional emergency rescuers performed 2 × 10 minutes of continuous chest compressions with and without a feedback device (CPRmeter). Efficient compression rate (primary outcome) was defined as the frequency target reached along with depth and leaning at the same time (recorded continuously). MAIN FINDINGS: The 10-minute mean efficient compression rate was significantly better in the feedback group: 42% vs 21% (P< .001). There was no significant difference between the first (43%) and the tenth minute (36%; P= .068) with feedback. Conversely, a significant difference was evident from the second minute without feedback (35% initially vs 27%; P< .001). The efficient compression rate difference with and without feedback was significant every minute, from the second minute onwards. CPRmeter feedback significantly improved chest compression depth from the first minute, leaning from the second minute and rate from the third minute. PRINCIPAL CONCLUSIONS: A real-time feedback device delivers longer effective, steadier chest compressions over time. An extrapolation of these results from simulation may allow rescuer switches to be carried out beyond the currently recommended 2 minutes when a feedback device is used.

Preload dependency determines the effects of phenylephrine on cardiac output in anaesthetised patients: A prospective observational study.

BACKGROUND: Although phenylephrine is widely used in the operating room to control arterial pressure, its haemodynamic effects remain controversial. OBJECTIVE: We hypothesised that the effect of phenylephrine on cardiac output is affected by preload dependency. DESIGN: A prospective observational study. SETTING: Single-centre, University Hospital of Caen, France. PATIENTS: Fifty ventilated patients undergoing surgery were studied during hypotension before and after administration of phenylephrine. MAIN OUTCOME MEASURES: Cardiac index (CI), stroke volume (SV), corrected flow time, mean arterial pressure, pulse pressure variations (PPV) and systemic vascular resistance index were used to assess effects of changes in preload dependency. RESULTS: Twenty seven (54%) patients were included in the preload-dependent group (PPV ≥ 13%) and 23 (46%) in the preload-independent group (PPV < 13%) before administration of phenylephrine. For the whole cohort, phenylephrine increased mean arterial pressure [58 (±8) mmHg vs. 79 (±13) mmHg; P < 0.0001] and calculated systemic vascular resistance index [2010 (1338; 2481) dyn s cm m vs. 2989 (2155; 3870) dyn s cm m; P < 0.0001]. However, CI and SV decreased in the preload-independent group [2.3 (1.9; 3.7) l min m vs. 1.8 (1.5; 2.7) l min m; P < 0.0001 and 65 (44; 81) ml vs. 56 (39; 66) ml; P < 0.0001 for both] but not in the preload-dependent group [respectively 2.1 (1.8; 3.5) l min m vs. 2.1 (1.8; 3.3) l min m; P = 0.168 and 49 (41; 67) ml vs. 53 (41; 69) ml; P = 0.191]. Corrected flow time increased [294 (47) ms vs. 306 (56) ms; P = 0.031], and PPV decreased [17 (15; 19) % vs.12 (14; 16) %; P < 0.0003] only in the PPV at least 13% group. CONCLUSION: The effects of phenylephrine on CI and SV depend on preload. CI and SV decreased in preload-independent patients through increase in afterload, but were unchanged in those preload-dependent through increased venous return.