Awake prone position in COVID-19 acute respiratory failure: a randomised crossover study using electrical impedance tomography.

Background: The goal of this study was to determine whether an awake prone position (aPP) reduces the global inhomogeneity (GI) index of ventilation measured by electrical impedance tomography (EIT) in COVID-19 patients with acute respiratory failure (ARF). Methods: This prospective crossover study included COVID-19 patients with COVID-19 and ARF defined by arterial oxygen tension:inspiratory oxygen fraction (P aO2 :F IO2 ) of 100-300 mmHg. After baseline evaluation and 30-min EIT recording in the supine position (SP), patients were randomised into one of two sequences: SP-aPP or aPP-SP. At the end of each 2-h step, oxygenation, respiratory rate, Borg scale and 30-min EIT were recorded. Results: 10 patients were randomised in each group. The GI index did not change in the SP-aPP group (baseline 74±20%, end of SP 78±23% and end of aPP 72±20%, p=0.85) or in the aPP-SP group (baseline 59±14%, end of aPP 59±15% and end of SP 54±13%, p=0.67). In the whole cohort, P aO2 :F IO2 increased from 133±44 mmHg at baseline to 183±66 mmHg in aPP (p=0.003) and decreased to 129±49 mmHg in SP (p=0.03). Conclusion: In spontaneously breathing nonintubated COVID-19 patients with ARF, aPP was not associated with a decrease of lung ventilation inhomogeneity assessed by EIT, despite an improvement in oxygenation.

Validation and Critical Evaluation of the Effective Arterial Elastance in Critically Ill Patients.

OBJECTIVES: First, to validate bedside estimates of effective arterial elastance = end-systolic pressure/stroke volume in critically ill patients. Second, to document the added value of effective arterial elastance, which is increasingly used as an index of left ventricular afterload. DESIGN: Prospective study. SETTING: Medical ICU. PATIENTS: Fifty hemodynamically stable and spontaneously breathing patients equipped with a femoral (n = 21) or radial (n = 29) catheter were entered in a "comparison" study. Thirty ventilated patients with invasive hemodynamic monitoring (PiCCO-2; Pulsion Medical Systems, Feldkirchen, Germany), in whom fluid administration was planned were entered in a " dynamic" study. INTERVENTIONS: In the "dynamic" study, data were obtained before/after a 500 mL saline administration. MEASUREMENTS AND MAIN RESULTS: According to the "cardiocentric" view, end-systolic pressure was considered the classic index of left ventricular afterload. End-systolic pressure was calculated as 0.9 × systolic arterial pressure at the carotid, femoral, and radial artery level. In the "comparison" study, carotid tonometry allowed the calculation of the reference effective arterial elastance value (1.73 ± 0.62 mm Hg/mL). The femoral estimate of effective arterial elastance was more accurate and precise than the radial estimate. In the "dynamic" study, fluid administration increased stroke volume and end-systolic pressure, whereas effective arterial elastance (femoral estimate) and systemic vascular resistance did not change. Effective arterial elastance was related to systemic vascular resistance at baseline (r = 0.89) and fluid-induced changes in effective arterial elastance and systemic vascular resistance were correlated (r = 0.88). In the 15 fluid responders (cardiac index increases ≥ 15%), fluid administration increased end-systolic pressure and decreased effective arterial elastance and systemic vascular resistance (each p < 0.05). In the 15 fluid nonresponders, end-systolic pressure increased (p < 0.05), whereas effective arterial elastance and systemic vascular resistance remained unchanged. CONCLUSIONS: In critically ill patients, effective arterial elastance may be reliably estimated at bedside (0.9 × systolic femoral pressure/stroke volume). We support the use of this validated estimate of effective arterial elastance when coupled with an index of left ventricular contractility for studying the ventricular-arterial coupling. Conversely, effective arterial elastance should not be used in isolation as an index of left ventricular afterload.

Esophageal Doppler Can Predict Fluid Responsiveness Through End-Expiratory and End-Inspiratory Occlusion Tests.

OBJECTIVES: To assess whether, in patients under mechanical ventilation, fluid responsiveness is predicted by the effects of short respiratory holds on cardiac index estimated by esophageal Doppler. DESIGN: Prospective, monocentric study. SETTING: Medical ICU. PATIENTS: Twenty-eight adult patients with acute circulatory failure and a decision of the clinicians in charge to administer fluids. INTERVENTIONS: Before and after infusing 500 mL of saline, we measured cardiac index estimated by esophageal Doppler before and during the last 5 seconds of successive 15-second end-inspiratory occlusion and end-expiratory occlusion, separated by 1 minute. Patients in whom volume expansion increased cardiac index measured by transpulmonary thermodilution greater than or equal to 15% were defined as "fluid responders." Cardiac index measured by the Pulse Contour Cardiac Output device (from pulse contour analysis or transpulmonary thermodilution) was used as the reference. MEASUREMENTS AND MAIN RESULTS: End-expiratory occlusion increased cardiac index estimated by esophageal Doppler more in responders than in nonresponders (8% ± 2% vs 3% ± 1%, respectively; p < 0.0001) and end-inspiratory occlusion decreased cardiac index estimated by esophageal Doppler more in responders than in nonresponders (-8% ± 5% vs -4% ± 2%, respectively; p = 0.0002). Fluid responsiveness was predicted by the end-expiratory occlusion induced percent change in cardiac index estimated by esophageal Doppler with an area under the receiver operating characteristic curve of 1.00 (95% CI, 0.88-1.00) and a threshold value of 4% increase in cardiac index estimated by esophageal Doppler. It was predicted by the sum of absolute values of percent changes in cardiac index estimated by esophageal Doppler during both occlusions with a similar area under the receiver operating characteristic curve (0.99 [0.86-1.00]) and with a threshold of 9% change in cardiac index estimated by esophageal Doppler, which is compatible with the esophageal Doppler precision. CONCLUSIONS: If the absolute sum of the percent change in cardiac index estimated by esophageal Doppler induced by two successive end-inspiratory occlusion and end-expiratory occlusion maneuvers is greater than 9%, it is likely that a 500 mL fluid infusion will increase cardiac output. This diagnostic threshold is higher than if only end-expiratory occlusion induced percent changes in cardiac index estimated by esophageal Doppler are taken into account.

Does metformin exposure before ICU stay have any impact on patients' outcome? A retrospective cohort study of diabetic patients.

BACKGROUND: Impact of metformin exposure before ICU stay remains controversial. Metformin is thought to induce lactic acidosis and haemodynamic instability but may reduce ICU mortality. We evaluated its influence on outcome in diabetic patients admitted in the ICU and then compared two different populations based on the presence of septic shock. METHODS: We conducted a retrospective cohort study in a 24-bed French ICU between October 2010 and December 2013, including all ICU-admitted diabetic patients. RESULTS: Among 635 diabetic patients admitted during the study period, 131 (21%) were admitted with septic shock. Multivariate analysis showed no difference in hospital mortality in all metformin users (OR 0.75 [95% CI 0.44-1.28]; p = 0.29), except in the septic shock subgroup (OR 0.61; 95% CI [0.37-0.99]; p = 0.04) despite higher vasopressor dosages in the first hours after shock onset. Blood lactate level was higher in metformin users than in non-metformin users in all patients (p < 0.001), in septic shock patients (p < 0.001) and in patients without kidney injury (p < 0.001). Metformin users did not have more septic shock from unknown aetiology (p = 0.65) or unknown pathogen (p = 0.99). CONCLUSIONS: Metformin use before admission to ICU did not affect in-hospital mortality. However, for patients with septic shock, mortality was lower, despite worse clinical presentation on admission. Blood lactate levels were always higher with or without septic shock and indifferent of kidney function.

Predicting Fluid Responsiveness in Critically Ill Patients by Using Combined End-Expiratory and End-Inspiratory Occlusions With Echocardiography.

OBJECTIVES: First, we aimed at assessing whether fluid responsiveness is predicted by the effects of an end-expiratory occlusion on the velocity-time integral of the left ventricular outflow tract. Second, we investigated whether adding the effects of an end-inspiratory occlusion and of an end-expiratory occlusion on velocity-time integral can predict fluid responsiveness with similar reliability than end-expiratory occlusion alone but with a higher threshold, which might be more compatible with the precision of echocardiography. DESIGN: Diagnostic study. SETTING: Medical ICU. PATIENTS: Thirty mechanically ventilated patients in whom fluid administration was planned. INTERVENTIONS: A 15-second end-expiratory occlusion and end-inspiratory occlusion, separated by 1 minute, followed by a 500-mL saline administration. MEASUREMENTS AND MAIN RESULTS: Pulse contour analysis-derived cardiac index and velocity-time integral were measured during the last 5 seconds of 15-second end-inspiratory occlusion and end-expiratory occlusion and after fluid administration. End-expiratory occlusion increased velocity-time integral more in responders than in nonresponders to fluid administration (11% ± 5% vs 3% ± 1%, respectively; p < 0.0001), and end-inspiratory occlusion decreased velocity-time integral more in responders than in nonresponders (12% ± 5% vs 5% ± 2%, respectively; p = 0.0002). When adding the absolute values of changes in velocity-time integral observed during both occlusions, velocity-time integral changed by 23% ± 9% in responders and by 8% ± 3% in nonresponders. Fluid responsiveness was predicted by the end-expiratory occlusion-induced change in velocity-time integral with an area under the receiver operating characteristic curve of 0.938 (0.785-0.989) and a threshold value of 5%. Fluid responsiveness was predicted by the sum of absolute values of changes in velocity-time integral during both occlusions with a similar reliability (area under the receiver operating characteristic curve = 0.973 [0.838-1.000]) but with a threshold of 13%. Both sensitivity and specificity were 93% (68-100%). CONCLUSIONS: If consecutive end-inspiratory occlusion and end-expiratory occlusion change velocity-time integral is greater than or equal to 13% in total, fluid responsiveness is accurately predicted. This threshold is more compatible with the precision of echocardiography than that obtained by end-expiratory occlusion alone.

Efficiency of goal-directed oxygen delivery in ICU patients.

BACKGROUND: Current clinical practice guidelines promote a goal-directed approach for oxygen delivery with respect to SpO2 objectives. We evaluated the efficiency of a strategy based on goal-directed O2 delivery in the ICU. METHODS: A group of 30 patients (Group 1) with a proven history of chronic obstructive pulmonary disease suffering from acute hypercarbic exacerbation was compared to 2 other groups of patients admitted for acute respiratory failure with no history of pulmonary disease: 30 patients requiring oxygen supply and/or non-invasive ventilation (Group 2) and 30 requiring invasive ventilation (Group 3). The delivery of oxygen was based on SpO2 measurement: 88-94% for Group 1 and 90-96% for others. The time spent with an SpO2 below, within and above the prescribed limits was collected. RESULTS: The mean time spent within the prescribed range was for Groups 1, 2 and 3, respectively as follows: 61.9% [60.5-63.2], 63.7% [62.3-65] and 56.4% [55.3-57.6] (P < 0.001 for each group). A history of chronic obstructive pulmonary disease was not correlated with better results (P = 0.11), while invasive ventilation was related to the time spent out of the prescribed range (P < 0.001; OR 1.3 [1.22-1.28]) especially in hyperoxaemia (40.7% [39.6-41.8] P < 0.001). Efficiency seems unrelated to nursing workload or night team exhaustion (r = -0.09, P = 0.77). CONCLUSIONS: Goal-directed oxygen delivery based on SpO2 objectives in ICU patients ensures that in only approximately 64% of the time, SpO2 stays within the prescribed range.