IMPACT OF DIFFERENT VENTILATION STRATEGIES ON GAS EXCHANGES AND CIRCULATION DURING PROLONGED MECHANICAL CARDIO-PULMONARY RESUSCITATION IN A PORCINE MODEL.

ABSTRACT: Background: Optimal ventilation during cardio-pulmonary resuscitation (CPR) is still controversial. Ventilation is expected to provide sufficient arterial oxygen content and adequate carbon dioxide removal, while minimizing the risk of circulatory impairment. The objective of the present study was to compare three ventilation strategies in a porcine model during mechanical continuous chest compressions (CCC) according to arterial oxygenation and hemodynamic impact. Method: Ventricular fibrillation was induced and followed by five no-flow minutes and thirty low-flow minutes resuscitation with mechanical-CCC without vasopressive drugs administration. Three groups of eight Landras pig were randomized according to the ventilation strategy: 1. Standard nonsynchronized volume-control mode (SD-group); 2. synchronized bilevel pressure-controlled ventilation (CPV-group); 3. continuous insufflation with Boussignac Cardiac-Arrest Device (BC-group). We assessed 1. arterial blood gases, 2. macro hemodynamics, 3. tissular cerebral macro and micro-circulation and 4. airway pressure, minute ventilation at baseline and every 5 minutes during the protocol. Results: Arterial PaO2 level was higher at each measurement time in SD-group (>200 mm Hg) compare to CPV-group and BC-group ( P < 0.01). In BC-group, arterial PaCO2 level was significantly higher (>90mm Hg) than in SD and CPV groups ( P < 0.01). There was no difference between groups concerning hemodynamic parameters, cerebral perfusion and microcirculation. Conclusion: Ventilation modalities in this porcine model of prolonged CPR influence oxygenation and decarboxylation without impairing circulation and cerebral perfusion. Synchronized bi-level pressure-controlled ventilation' use avoid hyperoxia and was as efficient as asynchronized volume ventilation to maintain alveolar ventilation and systemic perfusion during prolonged CPR.

Potential for Lung Recruitment Estimated by the Recruitment-to-Inflation Ratio in Acute Respiratory Distress Syndrome. A Clinical Trial.

Rationale: Response to positive end-expiratory pressure (PEEP) in acute respiratory distress syndrome depends on recruitability. We propose a bedside approach to estimate recruitability accounting for the presence of complete airway closure.Objectives: To validate a single-breath method for measuring recruited volume and test whether it differentiates patients with different responses to PEEP.Methods: Patients with acute respiratory distress syndrome were ventilated at 15 and 5 cm H2O of PEEP. Multiple pressure-volume curves were compared with a single-breath technique. Abruptly releasing PEEP (from 15 to 5 cm H2O) increases expired volume: the difference between this volume and the volume predicted by compliance at low PEEP (or above airway opening pressure) estimated the recruited volume by PEEP. This recruited volume divided by the effective pressure change gave the compliance of the recruited lung; the ratio of this compliance to the compliance at low PEEP gave the recruitment-to-inflation ratio. Response to PEEP was compared between high and low recruiters based on this ratio.Measurements and Main Results: Forty-five patients were enrolled. Four patients had airway closure higher than high PEEP, and thus recruitment could not be assessed. In others, recruited volume measured by the experimental and the reference methods were strongly correlated (R2 = 0.798; P < 0.0001) with small bias (-21 ml). The recruitment-to-inflation ratio (median, 0.5; range, 0-2.0) correlated with both oxygenation at low PEEP and the oxygenation response; at PEEP 15, high recruiters had better oxygenation (P = 0.004), whereas low recruiters experienced lower systolic arterial pressure (P = 0.008).Conclusions: A single-breath method quantifies recruited volume. The recruitment-to-inflation ratio might help to characterize lung recruitability at the bedside.Clinical trial registered with www.clinicaltrials.gov (NCT02457741).

Information conveyed by electrical diaphragmatic activity during unstressed, stressed and assisted spontaneous breathing: a physiological study.

BACKGROUND: The electrical activity of the crural diaphragm (Eadi), a surrogate of respiratory drive, can now be measured at the bedside in mechanically ventilated patients with a specific catheter. The expected range of Eadi values under stressed or assisted spontaneous breathing is unknown. This study explored Eadi values in healthy subjects during unstressed (baseline), stressed (with a resistance) and assisted spontaneous breathing. The relation between Eadi and inspiratory effort was analyzed. METHODS: Thirteen healthy male volunteers were included in this randomized crossover study. Eadi and esophageal pressure (Peso) were recorded during unstressed and stressed spontaneous breathing and under assisted ventilation delivered in pressure support (PS) at low and high assist levels and in neurally adjusted ventilatory assist (NAVA). Overall eight different situations were assessed in each participant (randomized order). Peak, mean and integral of Eadi, breathing pattern, esophageal pressure-time product (PTPeso) and work of breathing (WOB) were calculated offline. RESULTS: Median [interquartile range] peak Eadi at baseline was 17 [13-22] µV and was above 10 µV in 92% of the cases. Eadimax defined as Eadi measured at maximal inspiratory capacity reached 90 [63 to 99] µV. Median peak Eadi/Eadimax ratio was 16.8 [15.6-27.9]%. Compared to baseline, respiratory rate and minute ventilation were decreased during stressed non-assisted breathing, whereas peak Eadi and PTPeso were increased. During unstressed assisted breathing, peak Eadi decreased during high-level PS compared to unstressed non-assisted breathing and to NAVA (p = 0.047). During stressed breathing, peak Eadi was lower during all assisted ventilation modalities compared to stressed non-assisted breathing. During assisted ventilation, across the different conditions, peak Eadi changed significantly, whereas PTPeso and WOB/min were not significantly modified. Finally, Eadi signal was still present even when Peso signal was suppressed due to high assist levels. CONCLUSION: Eadi analysis provides complementary information compared to respiratory pattern and to Peso monitoring, particularly in the presence of high assist levels. Trial registration The study was registered as NCT01818219 in clinicaltrial.gov. Registered 28 February 2013.

Ventilation During Cardiopulmonary Resuscitation: What Have We Learned From Models?

The optimization of ventilation during cardiopulmonary resuscitation (CPR) is a broad field of research. Recent physiological observations in this field challenge the current understanding of respiratory and circulatory interactions. Thanks to different models available (bench, animal, human), the understanding of physiological phenomena occurring during CPR has progressed. In this review, we describe the clinical observations that have led to the emerging concept of lung volume reduction and associated thoracic airway closure. We summarize the clinical and animal observations supporting these concepts. We then discuss the different contributions of bench, animal, and human models to the understanding of airway closure and their impact on intrathoracic pressure, airway closure, and hemodynamics generated by chest compression. The limitation of airway pressure and ventilation, resulting from airway closure reproducible in models, may play a major role in ventilation and gas exchange impairment observed during prolonged resuscitation.

Intrathoracic Airway Closure Impacts CO2 Signal and Delivered Ventilation during Cardiopulmonary Resuscitation.

RATIONALE: End-tidal CO2 (EtCO2) is used to monitor cardiopulmonary resuscitation (CPR), but it can be affected by intrathoracic airway closure. Chest compressions induce oscillations in expired CO2, and this could reflect variable degrees of airway patency. OBJECTIVES: To understand the impact of airway closure during CPR, and the relationship between the capnogram shape, airway closure, and delivered ventilation. METHODS: This study had three parts: 1) a clinical study analyzing capnograms after intubation in patients with out-of-hospital cardiac arrest receiving continuous chest compressions, 2) a bench model, and 3) experiments with human cadavers. For 2 and 3, a constant CO2 flow was added in the lung to simulate CO2 production. Capnograms similar to clinical recordings were obtained and different ventilator settings tested. EtCO2 was compared with alveolar CO2 (bench). An airway opening index was used to quantify chest compression-induced expired CO2 oscillations in all three clinical and experimental settings. MEASUREMENTS AND MAIN RESULTS: A total of 89 patients were analyzed (mean age, 69 ± 15 yr; 23% female; 12% of hospital admission survival): capnograms exhibited various degrees of oscillations, quantified by the opening index. CO2 value varied considerably across oscillations related to consecutive chest compressions. In bench and cadavers, similar capnograms were reproduced with different degrees of airway closure. Differences in airway patency were associated with huge changes in delivered ventilation. The opening index and delivered ventilation increased with positive end-expiratory pressure, without affecting intrathoracic pressure. Maximal EtCO2 recorded between ventilator breaths reflected alveolar CO2 (bench). CONCLUSIONS: During chest compressions, intrathoracic airway patency greatly affects the delivered ventilation. The expired CO2 signal can reflect CPR effectiveness but is also dependent on airway patency. The maximal EtCO2 recorded between consecutive ventilator breaths best reflects alveolar CO2.

A new physiological model for studying the effect of chest compression and ventilation during cardiopulmonary resuscitation: The Thiel cadaver.

BACKGROUND: Studying ventilation and intrathoracic pressure (ITP) induced by chest compressions (CC) during Cardio Pulmonary Resuscitation is challenging and important aspects such as airway closure have been mostly ignored. We hypothesized that Thiel Embalmed Cadavers could constitute an appropriate model. METHODS: We assessed respiratory mechanics and ITP during CC in 11 cadavers, and we compared it to measurements obtained in 9 out-of-hospital cardiac arrest patients and to predicted values from a bench model. An oesophageal catheter was inserted to assess chest wall compliance, and ITP variation (ΔITP). Airway pressure variation (ΔPaw) at airway opening and ΔITP generated by CC were measured at decremental positive end expiratory pressure (PEEP) to test its impact on flow and ΔPaw. The patient's data were derived from flow and airway pressure captured via the ventilator during resuscitation. RESULTS: Resistance and Compliance of the respiratory system were comparable to those of the out-of-hospital cardiac arrest patients (CRSTEC 42 ±â€¯12 vs CRSPAT 37.3 ±â€¯10.9 mL/cmH2O and ResTEC 17.5 ±â€¯7.5 vs ResPAT 20.2 ±â€¯5.3 cmH2O/L/sec), and remained stable over time. During CC, ΔITP varied from 32 ±â€¯12 cmH2O to 69 ±â€¯14 cmH2O with manual and automatic CC respectively. Transmission of ΔITP at the airway opening was significantly affected by PEEP, suggesting dynamic small airway closure at low lung volumes. This phenomenon was similarly observed in patients. CONCLUSION: Respiratory mechanics and dynamic pressures during CC of cadavers behave as predicted by a theoretical model and similarly to patients. The Thiel model is a suitable to assess ITP variations induced by ventilation during CC.

Esophageal Manometry and Regional Transpulmonary Pressure in Lung Injury.

RATIONALE: Esophageal manometry is the clinically available method to estimate pleural pressure, thus enabling calculation of transpulmonary pressure (Pl). However, many concerns make it uncertain in which lung region esophageal manometry reflects local Pl. OBJECTIVES: To determine the accuracy of esophageal pressure (Pes) and in which regions esophageal manometry reflects pleural pressure (Ppl) and Pl; to assess whether lung stress in nondependent regions can be estimated at end-inspiration from Pl. METHODS: In lung-injured pigs (n = 6) and human cadavers (n = 3), Pes was measured across a range of positive end-expiratory pressure, together with directly measured Ppl in nondependent and dependent pleural regions. All measurements were obtained with minimal nonstressed volumes in the pleural sensors and esophageal balloons. Expiratory and inspiratory Pl was calculated by subtracting local Ppl or Pes from airway pressure; inspiratory Pl was also estimated by subtracting Ppl (calculated from chest wall and respiratory system elastance) from the airway plateau pressure. MEASUREMENTS AND MAIN RESULTS: In pigs and human cadavers, expiratory and inspiratory Pl using Pes closely reflected values in dependent to middle lung (adjacent to the esophagus). Inspiratory Pl estimated from elastance ratio reflected the directly measured nondependent values. CONCLUSIONS: These data support the use of esophageal manometry in acute respiratory distress syndrome. Assuming correct calibration, expiratory Pl derived from Pes reflects Pl in dependent to middle lung, where atelectasis usually predominates; inspiratory Pl estimated from elastance ratio may indicate the highest level of lung stress in nondependent "baby" lung, where it is vulnerable to ventilator-induced lung injury.

Effect of inspiratory synchronization during pressure-controlled ventilation on lung distension and inspiratory effort.

BACKGROUND: In pressure-controlled (PC) ventilation, tidal volume (V T) and transpulmonary pressure (P L ) result from the addition of ventilator pressure and the patient's inspiratory effort. PC modes can be classified into fully, partially, and non-synchronized modes, and the degree of synchronization may result in different V T and P L despite identical ventilator settings. This study assessed the effects of three PC modes on V T, P L , inspiratory effort (esophageal pressure-time product, PTPes), and airway occlusion pressure, P 0.1. We also assessed whether P 0.1 can be used for evaluating patient effort. METHODS: Prospective, randomized, crossover physiologic study performed in 14 spontaneously breathing mechanically ventilated patients recovering from acute respiratory failure (1 subsequently withdrew). PC modes were fully (PC-CMV), partially (PC-SIMV), and non-synchronized (PC-IMV using airway pressure release ventilation) and were applied randomly; driving pressure, inspiratory time, and set respiratory rate being similar for all modes. Airway, esophageal pressure, P 0.1, airflow, gas exchange, and hemodynamics were recorded. RESULTS: V T was significantly lower during PC-IMV as compared with PC-SIMV and PC-CMV (387 ± 105 vs 458 ± 134 vs 482 ± 108 mL, respectively; p < 0.05). Maximal P L was also significantly lower (13.3 ± 4.9 vs 15.3 ± 5.7 vs 15.5 ± 5.2 cmH2O, respectively; p < 0.05), but PTPes was significantly higher in PC-IMV (215.6 ± 154.3 vs 150.0 ± 102.4 vs 130.9 ± 101.8 cmH2O × s × min-1, respectively; p < 0.05), with no differences in gas exchange and hemodynamic variables. PTPes increased by more than 15% in 10 patients and by more than 50% in 5 patients. An increased P 0.1 could identify high levels of PTPes. CONCLUSIONS: Non-synchronized PC mode lowers V T and P L in comparison with more synchronized modes in spontaneously breathing patients but can increase patient effort and may need specific adjustments. Clinical Trial Registration Clinicaltrial.gov # NCT02071277.

Epidemiology of Weaning Outcome according to a New Definition. The WIND Study.

RATIONALE: The weaning process concerns all patients receiving mechanical ventilation. A previous classification into simple, prolonged, and difficult weaning ignored weaning failure and presupposed the use of spontaneous breathing trials. OBJECTIVES: To describe the weaning process, defined as starting with any attempt at separation from mechanical ventilation and its prognosis, according to a new operational classification working for all patients under ventilation. METHODS: This was a multinational prospective multicenter observational study over 3 months of all patients receiving mechanical ventilation in 36 intensive care units, with daily collection of ventilation and weaning modalities. Pragmatic definitions of separation attempt and weaning success allowed us to allocate patients in four groups. MEASUREMENTS AND MAIN RESULTS: A total of 2,729 patients were enrolled. Although half of them could not be classified using the previous definition, 99% entered the groups on the basis of our new definition as follows: 24% never started a weaning process, 57% had a weaning process of less than 24 hours (group 1), 10% had a difficult weaning of more than 1 day and less than 1 week (group 2), and 9% had a prolonged weaning duration of 1 week or more (group 3). Duration of ventilation, intensive care unit stay, and mortality (6, 17, and 29% for the three groups, respectively) all significantly increased from one group to the next. The unadjusted risk of dying was 19% after the first separation attempt and increased to 37% after 10 days. CONCLUSIONS: A new classification allows us to categorize all weaning situations. Every additional day without a weaning success after the first separation attempt increases the risk of dying.

Multivariable fractional polynomial interaction to investigate continuous effect modifiers in a meta-analysis on higher versus lower PEEP for patients with ARDS.

OBJECTIVES: A recent individual patient data (IPD) meta-analysis suggested that patients with moderate or severe acute respiratory distress syndrome (ARDS) benefit from higher positive end-expiratory pressure (PEEP) ventilation strategies. However, thresholds for continuous variables (eg, hypoxaemia) are often arbitrary and linearity assumptions in regression approaches may not hold; the multivariable fractional polynomial interaction (MFPI) approach can address both problems. The objective of this study was to apply the MFPI approach to investigate interactions between four continuous patient baseline variables and higher versus lower PEEP on clinical outcomes. SETTING: Pooled data from three randomised trials in intensive care identified by a systematic review. PARTICIPANTS: 2299 patients with acute lung injury requiring mechanical ventilation. INTERVENTIONS: Higher (N=1136) versus lower PEEP (N=1163) ventilation strategy. OUTCOME MEASURES: Prespecified outcomes included mortality, time to death and time-to-unassisted breathing. We examined the following continuous baseline characteristics as potential effect modifiers using MFPI: PaO2/FiO2 (arterial partial oxygen pressure/ fraction of inspired oxygen), oxygenation index, respiratory system compliance (tidal volume/(inspiratory plateau pressure-PEEP)) and body mass index (BMI). RESULTS: We found that for patients with PaO2/FiO2 below 150 mm Hg, but above 100 mm Hg or an oxygenation index above 12 (moderate ARDS), higher PEEP reduces hospital mortality, but the beneficial effect appears to level off for patients with very severe ARDS. Patients with mild ARDS (PaO2/FiO2 above 200 mm Hg or an oxygenation index below 10) do not seem to benefit from higher PEEP and might even be harmed. For patients with a respiratory system compliance above 40 mL/cm H2O or patients with a BMI above 35 kg/m(2), we found a trend towards reduced mortality with higher PEEP, but there is very weak statistical confidence in these findings. CONCLUSIONS: MFPI analyses suggest a nonlinear effect modification of higher PEEP ventilation by PaO2/FiO2 and oxygenation index with reduced mortality for some patients suffering from moderate ARDS. STUDY REGISTRATION NUMBER: CRD42012003129.

Impact of ventilation strategies during chest compression. An experimental study with clinical observations.

The optimal ventilation strategy during cardiopulmonary resuscitation (CPR) is unknown. Chest compression (CC) generates circulation, while during decompression, thoracic recoil generates negative pressure and venous return. Continuous flow insufflation of oxygen (CFI) allows noninterrupted CC and generates positive airway pressure (Paw). The main objective of this study was to assess the effects of positive Paw compared with the current recommended ventilation strategy on intrathoracic pressure (P(IT)) variations, ventilation, and lung volume. In a mechanical model, allowing compression of the thorax below an equilibrium volume mimicking functional residual capacity (FRC), CC alone or with manual bag ventilation were compared with two levels of Paw with CFI. Lung volume change below FRC at the end of decompression and P(IT), as well as estimated alveolar ventilation, were measured during the bench study. Recordings were obtained in five cardiac arrest patients to confirm the bench findings. Lung volume was continuously below FRC, and as a consequence P(IT) remained negative during decompression in all situations, including with positive Paw. Compared with manual bag or CC alone, CFI with positive Paw limited the fall in lung volume and resulted in larger positive and negative P(IT) variations. Positive Paw with CFI significantly augmented ventilation induced by CC. Recordings in patients confirmed a major loss of lung volume below FRC during CPR, even with positive Paw. Compared with manual bag ventilation, positive Paw associated with CFI limits the loss in lung volume, enhances CC-induced positive P(IT), maintains negative P(IT) during decompression, and generates more alveolar ventilation.

High-flow oxygen through nasal cannula in acute hypoxemic respiratory failure.

BACKGROUND: Whether noninvasive ventilation should be administered in patients with acute hypoxemic respiratory failure is debated. Therapy with high-flow oxygen through a nasal cannula may offer an alternative in patients with hypoxemia. METHODS: We performed a multicenter, open-label trial in which we randomly assigned patients without hypercapnia who had acute hypoxemic respiratory failure and a ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen of 300 mm Hg or less to high-flow oxygen therapy, standard oxygen therapy delivered through a face mask, or noninvasive positive-pressure ventilation. The primary outcome was the proportion of patients intubated at day 28; secondary outcomes included all-cause mortality in the intensive care unit and at 90 days and the number of ventilator-free days at day 28. RESULTS: A total of 310 patients were included in the analyses. The intubation rate (primary outcome) was 38% (40 of 106 patients) in the high-flow-oxygen group, 47% (44 of 94) in the standard group, and 50% (55 of 110) in the noninvasive-ventilation group (P=0.18 for all comparisons). The number of ventilator-free days at day 28 was significantly higher in the high-flow-oxygen group (24±8 days, vs. 22±10 in the standard-oxygen group and 19±12 in the noninvasive-ventilation group; P=0.02 for all comparisons). The hazard ratio for death at 90 days was 2.01 (95% confidence interval [CI], 1.01 to 3.99) with standard oxygen versus high-flow oxygen (P=0.046) and 2.50 (95% CI, 1.31 to 4.78) with noninvasive ventilation versus high-flow oxygen (P=0.006). CONCLUSIONS: In patients with nonhypercapnic acute hypoxemic respiratory failure, treatment with high-flow oxygen, standard oxygen, or noninvasive ventilation did not result in significantly different intubation rates. There was a significant difference in favor of high-flow oxygen in 90-day mortality. (Funded by the Programme Hospitalier de Recherche Clinique Interrégional 2010 of the French Ministry of Health; FLORALI ClinicalTrials.gov number, NCT01320384.).

Driving pressure and survival in the acute respiratory distress syndrome.

BACKGROUND: Mechanical-ventilation strategies that use lower end-inspiratory (plateau) airway pressures, lower tidal volumes (VT), and higher positive end-expiratory pressures (PEEPs) can improve survival in patients with the acute respiratory distress syndrome (ARDS), but the relative importance of each of these components is uncertain. Because respiratory-system compliance (CRS) is strongly related to the volume of aerated remaining functional lung during disease (termed functional lung size), we hypothesized that driving pressure (ΔP=VT/CRS), in which VT is intrinsically normalized to functional lung size (instead of predicted lung size in healthy persons), would be an index more strongly associated with survival than VT or PEEP in patients who are not actively breathing. METHODS: Using a statistical tool known as multilevel mediation analysis to analyze individual data from 3562 patients with ARDS enrolled in nine previously reported randomized trials, we examined ΔP as an independent variable associated with survival. In the mediation analysis, we estimated the isolated effects of changes in ΔP resulting from randomized ventilator settings while minimizing confounding due to the baseline severity of lung disease. RESULTS: Among ventilation variables, ΔP was most strongly associated with survival. A 1-SD increment in ΔP (approximately 7 cm of water) was associated with increased mortality (relative risk, 1.41; 95% confidence interval [CI], 1.31 to 1.51; P<0.001), even in patients receiving "protective" plateau pressures and VT (relative risk, 1.36; 95% CI, 1.17 to 1.58; P<0.001). Individual changes in VT or PEEP after randomization were not independently associated with survival; they were associated only if they were among the changes that led to reductions in ΔP (mediation effects of ΔP, P=0.004 and P=0.001, respectively). CONCLUSIONS: We found that ΔP was the ventilation variable that best stratified risk. Decreases in ΔP owing to changes in ventilator settings were strongly associated with increased survival. (Funded by Fundação de Amparo e Pesquisa do Estado de São Paulo and others.).

Noninvasive ventilation for patients with hypoxemic acute respiratory failure.

Noninvasive ventilation (NIV) has an established efficacy to improve gas exchange and reduce the work of breathing in patients with hypoxemic acute respiratory failure. The clinical efficacy in terms of meaningful outcome is less clear and depends very much on patient selection and assessment of the risks of the technique. The potential risks include an insufficient reduction of the oxygen consumption of the respiratory muscles in case of shock, an excessive increase in tidal volume in case of lung injury, and a risk of delayed or emergent intubation. With a careful selection of patients and a rapid decision regarding the need for intubation in case of failure, great benefits can be offered to patients. Emerging indications include its use in patients with treatment limitations, in the postoperative period, and in patients with immunosuppression. This last indication will necessitate reappraisal because the prognosis of the conditions associated with immunosuppression has improved over the years. In all cases, there is both a time window and a severity window for NIV to work, after which delaying endotracheal intubation may worsen outcome. The preventive use of NIV seems promising in this setting but needs more research. An emerging interesting new option is the use of high flow humidified oxygen, which seems to be intermediate between oxygen alone and NIV.

Feasibility and reliability of an automated controller of inspired oxygen concentration during mechanical ventilation.

INTRODUCTION: Hypoxemia and high fractions of inspired oxygen (FiO2) are concerns in critically ill patients. An automated FiO2 controller based on continuous oxygen saturation (SpO2) measurement was tested. Two different SpO2-FiO2 feedback open loops, designed to react differently based on the level of hypoxemia, were compared. The results of the FiO2 controller were also compared with a historical control group. METHODS: The system measures SpO2, compares with a target range (92% to 96%), and proposes in real time FiO2 settings to maintain SpO2 within target. In 20 patients under mechanical ventilation, two different FiO2-SpO2 open loops were applied by a dedicated research nurse during 3 hours, each in random order. The times spent in and outside the target SpO2 values were measured. The results of the automatic controller were then compared with a retrospective control group of 30 ICU patients. SpO2-FiO2 values of the control group were collected over three different periods of 6 hours. RESULTS: Time in the target range was higher than 95% with the controller. When the 20 patients were separated according to the median PaO2/FiO2 (160(133-176) mm Hg versus 239(201-285)), the loop with the highest slope was slightly better (P = 0.047) for the more-hypoxemic patients. Hyperoxemia and hypoxemia durations were significantly shorter with the controller compared with usual care: SpO2 target range was reached 90% versus 24%, 27% and 32% (P < .001) with the controller, compared with three historical control-group periods. CONCLUSION: A specific FiO2 controller is able to maintain SpO2 reliably within a predefined target range. Two different feedback loops can be used, depending on the initial PaO2/FiO2; with both, the automatic controller showed excellent performance when compared with usual care.