key: cord-0004472-any3ztrm authors: Diniz-Silva, Fabia; Moriya, Henrique T.; Alencar, Adriano M.; Amato, Marcelo B. P.; Carvalho, Carlos R. R.; Ferreira, Juliana C. title: Neurally adjusted ventilatory assist vs. pressure support to deliver protective mechanical ventilation in patients with acute respiratory distress syndrome: a randomized crossover trial date: 2020-02-10 journal: Ann Intensive Care DOI: 10.1186/s13613-020-0638-0 sha: 8b637fe79618c0664d68352accf9eec80f7fe18e doc_id: 4472 cord_uid: any3ztrm BACKGROUND: Protective mechanical ventilation is recommended for patients with acute respiratory distress syndrome (ARDS), but it usually requires controlled ventilation and sedation. Using neurally adjusted ventilatory assist (NAVA) or pressure support ventilation (PSV) could have additional benefits, including the use of lower sedative doses, improved patient–ventilator interaction and shortened duration of mechanical ventilation. We designed a pilot study to assess the feasibility of keeping tidal volume (V(T)) at protective levels with NAVA and PSV in patients with ARDS. METHODS: We conducted a prospective randomized crossover trial in five ICUs from a university hospital in Brazil and included patients with ARDS transitioning from controlled ventilation to partial ventilatory support. NAVA and PSV were applied in random order, for 15 min each, followed by 3 h in NAVA. Flow, peak airway pressure (Paw) and electrical activity of the diaphragm (EAdi) were captured from the ventilator, and a software (Matlab, Mathworks, USA), automatically detected inspiratory efforts and calculated respiratory rate (RR) and V(T). Asynchrony events detection was based on waveform analysis. RESULTS: We randomized 20 patients, but the protocol was interrupted for five (25%) patients for whom we were unable to maintain V(T) below 6.5 mL/kg in PSV due to strong inspiratory efforts and for one patient for whom we could not detect EAdi signal. For the 14 patients who completed the protocol, V(T) was 5.8 ± 1.1 mL/kg for NAVA and 5.6 ± 1.0 mL/kg for PSV (p = 0.455) and there were no differences in RR (24 ± 7 for NAVA and 23 ± 7 for PSV, p = 0.661). Paw was greater in NAVA (21 ± 3 cmH(2)O) than in PSV (19 ± 3 cmH(2)O, p = 0.001). Most patients were under continuous sedation during the study. NAVA reduced triggering delay compared to PSV (p = 0.020) and the median asynchrony Index was 0.7% (0–2.7) in PSV and 0% (0–2.2) in NAVA (p = 0.6835). CONCLUSIONS: It was feasible to keep V(T) in protective levels with NAVA and PSV for 75% of the patients. NAVA resulted in similar V(T), RR and Paw compared to PSV. Our findings suggest that partial ventilatory assistance with NAVA and PSV is feasible as a protective ventilation strategy in selected ARDS patients under continuous sedation. Trial registration ClinicalTrials.gov (NCT01519258). Registered 26 January 2012, https://clinicaltrials.gov/ct2/show/NCT01519258 Acute respiratory distress syndrome (ARDS) has a high mortality burden [1] , especially in low and middleincome countries [2, 3] . Protective mechanical ventilation (MV)-consisting of the use of tidal volume (V T ) equal or less than 6 mL/kg of predicted body weight (PBW) and plateau pressure (Pplat) limited to 30 cmH 2 O-reduces mortality and is recommended for ARDS [4] [5] [6] [7] [8] . In the initial phase of ARDS, patients are often ventilated with controlled modes for rigorous control of V T and Pplat, requiring sedation and sometimes, neuromuscular blockade [6, [9] [10] [11] [12] , which are associated with diaphragmatic weakness [13] [14] [15] . On the other hand, overload of the respiratory muscles during acute respiratory failure causes muscle fatigue and is also associated with adverse events [16] . Using partial ventilatory support could be one alternative to prevent respiratory muscles weakness and complications associated with controlled mechanical ventilation [13, 14] while also preventing muscle fatigue [16] . Partial ventilatory support in ARDS could have additional benefits, including the use of lower sedative doses, improved patient-ventilator interaction and shortened duration of MV [12, 17, 18] . Neurally adjusted ventilatory assist (NAVA) is a proportional ventilatory mode that uses the electrical activity of the diaphragm (EAdi) to trigger, cycle and provide inspiratory assistance in proportion to patient's effort [19] [20] [21] . Studies have shown that NAVA prevents excessive lung distension, reduces the work of breathing and improves patient-ventilator synchrony when compared with pressure support ventilation (PSV) [22] [23] [24] [25] [26] [27] [28] . We designed a pilot study to assess the feasibility of using NAVA and PSV for ARDS patients transitioning from controlled ventilation to partial ventilatory support. We hypothesized that it would be feasible to keep V T at protective levels with NAVA and PSV. We conducted a crossover study in five intensive care units of a university hospital in São Paulo, Brazil, from November 2012 to November 2015. The Institution's Ethics committee (CaPPesq 02874612.6.0000.0068) approved the study and informed consent was obtained from the families of patients and attending physicians. The study was registered on ClinicalTrials.gov (NCT01519258). We screened intubated and mechanically ventilated patients with ARDS according to the Berlin definition [29] and included a pilot sample of 20 patients. We could not calculate a sample size since data on the performance of NAVA during protective ventilation for ARDS was not available in the literature. Inclusion criteria were MV for more than 24 h; diagnosis of ARDS; indication of protective MV, by the ICU team; presence of inspiratory efforts triggering the ventilator for more than 6 h. Exclusion criteria were age < 18 years, pregnancy, severe hemodynamic instability, contraindications to the placement of the esophageal catheter and participation in other clinical trials. We randomized patients who fulfilled all inclusion criteria and no exclusion criteria to the order of ventilation in NAVA and PSV using a computer generated randomization list (http://www.R-proje ct.org/, Vienna, Austria), and numbered, opaque and sealed envelopes. Patients were ventilated with the Servoi Ventilator (Maquet Critical Care, Solna, Sweden), and ventilator settings before initiation of the protocol were adjusted by ICU team (Baseline). PEEP (positive end-expiratory pressure) and FIO 2 (fraction of inspired oxygen) were kept constant during the study. We registered baseline ventilatory parameters and demographic data. Sedation was adjusted by the ICU team according to the ICU sedation protocol, targeting a RASS of − 5 for patients receiving neuromuscular blocker and RASS − 2 to 0 for patients transitioning to assisted ventilation. Patients were switched from Baseline to PSV and we titrated PSV to generate a V T ≤ 6 mL/kg PBW. We started with a PS level of 10 cmH 2 O and increased or decreased it to reach the target V T . If V T was ≥ 6.5 mL/ kg despite using PS ≤ 3 cmH 2 O, the protocol was interrupted, and Baseline was resumed (Fig. 1) . If we were able to provide protective MV with PSV, we placed the NAVA catheter and titrated NAVA. Triggering sensitivity and cycling criteria in PSV were adjusted by the ICU Conclusions: It was feasible to keep V T in protective levels with NAVA and PSV for 75% of the patients. NAVA resulted in similar V T , RR and Paw compared to PSV. Our findings suggest that partial ventilatory assistance with NAVA and PSV is feasible as a protective ventilation strategy in selected ARDS patients under continuous sedation. Trial registration ClinicalTrials.gov (NCT01519258). Registered 26 January 2012, https ://clini caltr ials.gov/ct2/show/ NCT01 51925 8 Keywords: Respiration, artificial, Respiratory distress syndrome, adult, Interactive ventilatory support, Positivepressure respiration, Neurally adjusted ventilatory assist team. The ICU team was instructed to use flow triggering adjusted to be as sensitive as possible without generating auto-triggering. As previously described [30] , we measured EAdi using a dedicated NAVA catheter (Maquet, Sweden) and titrated NAVA level to generate the same peak pressure generated with PSV. Triggering sensitivity in NAVA was fixed at 0.5 µV, and cycling criteria was fixed at 70% of peak EAdi. Pneumatic triggering sensitivity was adjusted by the ICU team and kept constant. Then, patients were ventilated in NAVA and PSV for 15 min each, in the order determined by randomization, followed by ventilation with NAVA for 3 h (NAVA3h). We collected hemodynamic and respiratory parameters at the end of NAVA and PSV periods and every 15 min in the NAVA3h and obtained arterial blood gases at the end of NAVA and PSV periods and NAVA3h. We recorded peak airway pressure (Paw), flow, and EAdi continuously from the ventilator at a sampling rate of 100 Hz, using dedicated software (ServoTracker V.4.2; Maquet, Solna, Sweden). We analyzed 3-min recordings of the end of the period of Baseline, NAVA and PSV ventilation. For analysis of NAVA3h, data from several consecutive recordings were processed in order to analyze at least 10 min of recording for every 30 min. We processed and analyzed the data using a custom computer routine (Matlab, Mathworks, MA, USA), which automatically detected the moment of initiation and termination of inspiratory efforts and ventilator cycles, and calculated V T , Paw, respiratory rate (RR), minute ventilation, mean airway pressure (MAP), EAdi peak, and neural inspiratory time (TIn): as the difference of time (in seconds) between the initiation of an inspiratory effort and EAdi Peak. We averaged all cycles in each situation to generate a mean value for the above variables. We calculated EAdi/TIn for each respiratory cycle to have an index of respiratory drive. Asynchrony event detection was based on a custom computer routine (Matlab, Mathworks, MA, USA) which detected the beginning and termination of each ventilator cycle, the beginning and termination of each respiratory effort and calculated cycle times and mean inspiratory times. This analysis was followed by visual inspection of Paw, flow and EAdi waveforms to confirm the presence of asynchrony. We defined the types of asynchrony in accordance with previous publications [31] [32] [33] [34] as: autotriggering, a ventilator cycle not preceded by an inspiratory effort; triggering delay, delay between start of patient effort and triggering > 25% mean inspiratory time; ineffective effort, an inspiratory effort not accompanied by a ventilator cycle; double triggering, two cycles separated by an expiratory time less than half of mean inspiratory time; prolonged cycle, inspiratory time greater than twice the mean inspiratory time; short cycle, inspiratory time less than half the mean inspiratory time. The asynchrony index (AI) was calculated as the number of cycles with major asynchronies (auto-triggering, ineffective efforts, double triggering and short cycle) divided by the number of monitored neural cycles, expressed as a percentage. The primary endpoint was V T in mL/kg of predictive body weight, and secondary endpoints were Paw, EAdi, RR and asynchrony index. Continuous variables are expressed by mean and standard deviation or median and 25-75% interquartile range. We used paired t tests to compare continuous variables. For non-normally distributed variables, we used paired Wilcoxon signed-rank tests. Statistical analysis was done with R (http://www.R-proje ct.org/, Vienna, Austria). A p value less than 0.05 was considered significant. We assessed 104 patients with P/F ratio less than 300, and excluded 84, mainly because they did not meet the inclusion criteria, refused to participate or had severe hemodynamic instability (Fig. 1) . Twenty patients were randomized and included in the trial, and 14 completed the protocol. The protocol was interrupted for five patients (25%) for whom we were unable to maintain V T below 6.5 mL/kg PBW in PSV due to strong inspiratory efforts and for one patient for whom we could not detect the EAdi signal. Patient's demographic characteristics are shown in Table 1 . The baseline mode was volume-controlled for six patients and pressure-controlled for 14 patients. Ventilatory settings during the study protocol are shown in Table 2 . Most patients received neuromuscular blockage on the first 48 h of the diagnosis, but none were still receiving it at the time of the study. Sedation was used for most patients at ICU team discretion, targeting a RASS of − 2 to 0, but since we studied patients a few hours after NMB interruption, some patients were still deeply sedated. The most commonly sedatives used were fentanyl, propofol and midazolam ( Table 2) . The V T stayed within protective levels for the 14 patients who completed the protocol, and there was no difference between NAVA and PSV (Table 3 ). There was also no statistical difference in RR, minute ventilation, MAP, EAdi and EAdi/TIn comparing the two modes. Paw was greater in NAVA than in PSV, but it remained at protective levels. Median PaO 2 and median P/F were greater in NAVA than in PSV. There were no differences for other blood gas variables between NAVA and PSV (Table 4 ) or in the blood gases comparing baseline and NAVA3h (Additional file 1: Table S1 ). Comparisons of the incidence of asynchronies between NAVA and PSV for each patient are shown in Fig. 2 . Auto-triggering (Fig. 2a) was similar for both modes; Double triggering ( Fig. 2b) was observed in four patients during NAVA and in one patient during PSV, but the difference was not significant. In NAVA, the second cycle in a double triggering event had very low or zero flow (Fig. 3) . Prolonged cycle (Fig. 2c ) was similar in NAVA and PSV; ineffective effort was observed in two patients during PSV, and absent during NAVA (Fig. 2d) . Triggering delay (Fig. 2e) was observed in 13 patients during PSV and in 10 patients during NAVA. Short cycle was observed in only one patient in PSV. AI was greater than 10% in two patients in both PSV and NAVA and was not significantly different between the two modes of ventilation (Fig. 2f ) , with a median of 0.7% (0-2.7) in PSV and 0% (0-2.2) in NAVA (p = 0.6835). To evaluate if protective MV could be sustained with NAVA, we compared the V T , RR and Paw for the 13 patients who completed the period of NAVA3h with controlled ventilation at Baseline. There was no statistical difference in V T (5.8 ± 1.2 mL/kg in NAVA and 5.5±0.8 mL/ kg in baseline; p = 0.364), RR (24 ± 7 in NAVA and 25 ± 7 in baseline; p = 0.946), or Paw (20 ± 4 cmH 2 O in NAVA and 23 ± 4 cmH 2 O in baseline; p = 0.051). The median P/F ratio was similar for both modes, 216 (172-281) vs. 217 (167-266), p = 0.765. One patient who presented apneas on the first hour of NAVA3h was excluded from this assessment. Mean V T was stable and remained at protective levels over NAVA3h (Fig. 4 ). In this study, we compared V T , RR, Paw, gas exchange and patient-ventilator asynchronies in NAVA and PSV in patients with ARDS on the first day after neuromuscular blockage were interrupted. The main findings were: (1) most patients could be ventilated in PSV and NAVA within protective levels, but one-quarter had strong respiratory efforts that resulted in high tidal volumes that prevented the application of protective ventilation with partial ventilatory support. (2) NAVA and PSV resulted in similar V T and RR; NAVA resulted in greater Paw than PSV, but Paw in both modes remained within protective levels. (3) There was no difference in the asynchrony index between NAVA and PSV. (4) NAVA was well tolerated during a 3-h period and had similar V T , Paw and RR compared to controlled ventilation. To our knowledge, this is the first study using NAVA while providing protective ventilation in ARDS patients. Our protocol began a few hours after the deep sedation and/or neuromuscular blockage was discontinued and patients had respiratory drive. Other studies using NAVA in patients with ARDS were performed in the weaning phase [25, 34] , in patients on ECMO [33] or did not specify timing [24] . The largest trial using NAVA randomized 128 patients to NAVA or PSV in the beginning of the weaning of MV, but the study population was a mix of several causes of respiratory failure [35] . Our intention was to apply partial ventilatory support-NAVA and PSV-while providing protective MV with a potential to reduce the need for sedation and the occurrence of asynchrony, as an improved strategy to avoid ventilatorinduced lung injury (VILI) [36] . We found that this approach was feasible for most ARDS patients under continuous sedation, and a short trial on PSV was used to detect patients with strong inspiratory efforts who could not be kept on protective ventilation in partial ventilatory support. Patients who could be ventilated with approximately 6 mL/kg in PSV could also be safely ventilated in NAVA. We included patients on the three categories of ARDS, with a median of 3 days of MV. On the day of the study, the median P/F was still low, 190, and the two patients with P/F above 300 had received recruitment maneuvers and were on high PEEP (13 and 15 cmH 2 O). Tidal volume was similar in PSV and NAVA and within protective values for the 14 patients who completed the protocol. However, we had to interrupt the protocol and resume deep sedation for five patients (25%) who had strong inspiratory efforts resulting in V T ≥ 8 mL/kg, even with low levels of PS. Our concern was that strong inspiratory efforts and high tidal volumes could contribute to VILI [37] . In such cases, patients may benefit from the use of controlled ventilation and sedation to avoid further injury caused by MV [6] . The breathing pattern was similar in NAVA and PSV, there were no differences in RR, in accordance with previous studies [24, 33] . The respiratory drive was similar in both modes for the 14 patients enrolled, as indicated by the mean values of EAdi/TIn and EAdi. Paw was greater in NAVA than in PSV, despite titration of NAVA level to generate the same Paw in PSV. That happened because Paw in NAVA varies with EAdi, in contrast to PSV, where it is set. However, Paw remained within protective levels in both modes and this difference did not translate into more assistance in NAVA compared to PSV, since RR, V T and EAdi were similar in both modes [21, 38, 39] . The P/F ratio was higher in NAVA when compared with PSV, but the difference was not clinically relevant. The incidence of the asynchrony was relatively low in our study. AI was greater than 10% in only two patients in both PSV and NAVA, which is considered clinically important and associated with prolonged mechanical ventilation [31, 32, 40] . This finding may be related to the fact that most patients were still sedated during the study and that they were ventilated with low tidal volumes, which may have prevented the occurrence of ineffective triggering [41] the most common type of asynchrony [40, 42] . Triggering delay was the most common type of asynchrony in our study, and it was significantly reduced in NAVA compared to PSV. Since triggering delay greater than 0.15 s may cause considerable discomfort [26, 43] and is easily measured with the NAVA catheter [44] [45] [46] , we report its value but do not include it in the computation of AI to allow comparison of our results with other studies. The occurrence of ineffective effort was rare and observed in only two patients in PSV. This is a contrast with previous studies comparing NAVA and PSV [22, 23, 25, 26, 45] . One possible explanation for this difference is that the most important risk factor for ineffective effort is over assistance and air trapping. Since ARDS patients have low respiratory system compliance and we titrated NAVA and PSV to deliver low V T , risk factors for air trapping were reduced. Double triggering was more common in NAVA than in PSV. This result is consistent with results obtained in other studies [26, 47] . The presence of double triggering in NAVA is related to EAdi signal showing a biphasic curve. In our study, the second cycle in NAVA usually resulted in zero flow and did not result in breath stacking and high V T , which is a concern related to double triggering in assisted-controlled modes [24, [48] [49] [50] . Auto-triggering was observed in four patients in PSV and in only one patient during NAVA. The low incidence of prolonged cycle compared to previous studies [22, 23, 25, 26] is related to patients' mechanics and influenced by the cycling criterion used. In NAVA, the cycling criterion is fixed at 70% of peak EAdi, and in PSV it was adjusted by the ICU team. NAVA marginally improved PaO 2 and P/F compared with PSV, but the difference was not clinically relevant. Previous studies showed that spontaneous breathing efforts are associated with improved gas exchange compared with controlled ventilation [24, 51, 52] , but we did not observe differences in gas exchange comparing NAVA3h and controlled ventilation. We found that NAVA was well tolerated for 3 h, and had similar RR, V T , Paw and hemodynamics compared to controlled ventilation. Our study has several limitations: first, it was a pilot, single-center study, we recruited a small sample size and patients included in the study showed a great variability regarding the number of days of MV, PEEP levels and P/F ratio. In addition, each ventilatory mode was studied for a short period. Therefore, more studies are necessary to test the feasibility and safety of using NAVA continuously in patients with ARDS to deliver protective MV. Second, we opted to interrupt the protocol for 25% of patients due to strong inspiratory efforts that prevented us from maintaining low V T in PSV. This choice was made to address a safety issue, since there is uncertainty about deleterious effects of strong inspiratory efforts at the early phase of ARDS. Since the patients had already been randomized, our power to detect differences between NAVA and PSV was reduced. Therefore, our conclusions about the safety of NAVA and PSV to deliver protective MV cannot be extrapolated to all patients with ARDS. And third, we used the EAdi to identify patients' inspiratory efforts and Filled diamonds represent the mean tidal volume for predicted body weight of 13 patients who completed the 3 h of NAVA and bars represent standard deviation. V T mL/kg: tidal volume for predicted body weight asynchrony events; therefore, inspiratory efforts initiated by accessory muscles were not detected, which might impact asynchrony detection. In addition, EAdi is a processed signal, which may have impacted precision. In this pilot study, we found that it was feasible to keep tidal volume within protective levels with NAVA and PSV for 75% of the patients with ARDS under continuous sedation. Tidal volume was similar in PSV and NAVA and remained within protective levels for 3 h with NAVA. These findings suggest that using partial ventilatory assistance in ARDS can be used as part of a protective ventilation strategy, with potential benefits of less sedation and less muscle paralysis. Supplementary information accompanies this paper at https ://doi. org/10.1186/s1361 3-020-0638-0. Additional file 1: Table S1 . Comparison of blood gases between controlled ventilation and NAVA at the end of 3 h. Author's contributions FDS and JCF designed the study, participated in data collection and interpretation, and drafted the initial manuscript. HTM and AMA participated in data collection and interpretation and critically revised the manuscript. MBPA and CRRC participated in study design, data interpretation, and critically revised the manuscript. All authors read and approved the final version of the manuscript. This study was funded by FAPESP (Project Number 2011/20225-1), a government funding agency in Brazil. All data are available in the manuscript. The Institution's Ethics committee (CaPPesq 02874612.6.0000.0068) approved the study and informed consent was obtained from the families of patients and attending physician. The study was registered on ClinicalTrials.gov (NCT01519258). Figure 4 shows a double triggering of one patient of the study, but the image is entirely unidentifiable. Dr. Juliana Ferreira received fees for lecturing from Medtronic from 2016 to 2019. Dr. Marcelo B.P. Amato, MD, reports that his research laboratory has received grants in the last 5 years from the Covidien/Medtronic (mechanical ventilation), Orange Med/Nihon Kohden (mechanical ventilation) and Timpel S.A (Electrical Impedance Tomography). Dr. Marcelo B.P. Amato is also a minority shareholder in Timpel. The authors, FDS, HTM, AML, and CRRC have no conflicts of interest to disclose related to the contents of the manuscript. Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries Clinical outcomes of patients requiring ventilatory support in Brazilian intensive care units: a multicenter, prospective, cohort study Global critical care: moving forward in resource-limited settings Brazilian recommendations of mechanical ventilation 2013. Parte I. Rev Bras Ter Intensiva An official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine Clinical Practice Guideline: mechanical ventilation in adult patients with acute respiratory distress syndrome Respiratory support in patients with acute respiratory distress syndrome: an expert opinion Low tidal volume versus non-volume-limited strategies for patients with acute respiratory distress syndrome-a systematic review and meta-analysis Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome Neuromuscular blockers in early acute respiratory distress syndrome Early neuromuscular blockade in the acute respiratory distress syndrome Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome Daily interruption of sedative infusions in critically ill patients undergoing mechanical ventilation Ventilator-associated diaphragmatic dysfunction Ventilator-induced diaphragmatic dysfunction Rapidly progressive diaphragmatic weakness and injury during mechanical ventilation in humans Patient-ventilator interaction during acute lung injury, and the role of spontaneous breathing: part 1: respiratory muscle function during critical illness Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial Impact of ventilator adjustment and sedation-analgesia practices on severe asynchrony in patients ventilated in assist-control mode Proportional assist ventilation and neurally adjusted ventilatory assist-better approaches to patient ventilator synchrony? Neural control of mechanical ventilation in respiratory failure Inspiratory muscle unloading by neurally adjusted ventilatory assist during maximal inspiratory efforts in healthy subjects Physiologic response to varying levels of pressure support and neurally adjusted ventilatory assist in patients with acute respiratory failure Patient-ventilator interaction during pressure support ventilation and neurally adjusted ventilatory assist Assisted ventilation in patients with acute respiratory distress syndrome-lung-distending pressure and patient-ventilator interaction Neurally adjusted ventilatory assist in patients recovering spontaneous breathing after acute respiratory distress syndrome: physiological evaluation Neurally adjusted ventilatory assist improves patient-ventilator interaction Double cycling during mechanical ventilation frequency, mechanisms and physiologic implications Comparison between neurally adjusted ventilatory assist and pressure support ventilation levels in terms of respiratory effort Acute respiratory distress syndrome: the Berlin Definition Evaluation of the catheter positioning for neurally adjusted ventilatory assist Patient-ventilator trigger asynchrony in prolonged mechanical ventilation Patient-ventilator asynchrony during assisted mechanical ventilation Patient-ventilator interaction in ARDS patients with extremely low compliance undergoing ECMO: a novel approach based on diaphragm electrical activity Neurally adjusted ventilatory assist (NAVA) or pressure support ventilation (PSV) during spontaneous breathing trials in critically ill patients: a crossover trial Neurally adjusted ventilatory assist as an alternative to pressure support ventilation in adults: a French multicenter randomized trial Acute respiratory distress syndrome: advances in diagnosis and treatment Spontaneous effort causes occult pendelluft during mechanical ventilation Physiological response to increasing levels of neurally adjusted ventilatory assist (NAVA) Titration and implementation of neurally adjusted ventilatory assist in critically ill patients Asynchronies during mechanical ventilation are associated with mortality Reduction of patientventilator asynchrony by reducing tidal volume during pressure-support ventilation Predictors of asynchronies during assisted ventilation and its impact on clinical outcomes: the EPISYNC cohort study Asynchrony, neural drive, ventilatory variability and COMFORT: NAVA versus pressure support in pediatric patients. A non-randomized cross-over trial Patient-ventilator asynchrony during conventional mechanical ventilation in children Efficacy of ventilator waveforms observation in detecting patientventilator asynchrony Clinical assessment of auto-positive end-expiratory pressure by diaphragmatic electrical activity during pressure support and neurally adjusted ventilatory assist Patientventilator synchrony in Neurally Adjusted Ventilatory Assist (NAVA) and Pressure Support Ventilation (PSV): a prospective observational study Quantifying unintended exposure to high tidal volumes from breath stacking dyssynchrony in ARDS: the BREATHE criteria The association between ventilator dyssynchrony, delivered tidal volume, and sedation using a novel automated ventilator dyssynchrony detection algorithm Patient-ventilator asynchrony Long-term effects of spontaneous breathing during ventilatory support in patients with acute lung injury Spontaneous breathing activity in acute lung injury and acute respiratory distress syndrome Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations We acknowledge and thank Dr. Ho Yeh Li and Dr. Leandro Taniguchi for their support in inclusion of patients. We also acknowledge the MECOR program, ATS and ALAT, for their support for research in Latin America. 1