Weaning Outcomes in Patients with Brain Injury.

BACKGROUND: Despite the need for specific weaning strategies in neurological patients, evidence is generally insufficient or lacking. We aimed to describe the evolution over time of weaning and extubation practices in patients with acute brain injury compared with patients who are mechanically ventilated (MV) due to other reasons. METHODS: We performed a secondary analysis of three prospective, observational, multicenter international studies conducted in 2004, 2010, and 2016 in adults who had need of invasive MV for more than 12 h. We collected data on baseline characteristics, variables related to management ventilator settings, and complications while patients were ventilated or until day 28. RESULTS: Among the 20,929 patients enrolled, we included 12,618 (60%) who started the weaning from MV, of whom 1722 (14%) were patients with acute brain injury. In the acutely brain-injured cohort, 538 patients (31%) did not undergo planned extubation, defined as the need for a tracheostomy without an attempt of extubation, accidental extubation, and death. Among the 1184 planned extubated patients with acute brain injury, 202 required reintubation (17%). Patients with acute brain injury had a higher odds for unplanned extubation (odds ratio [OR] 1.35, confidence interval for 95% [CI 95%] 1.19-1.54; p < 0.001), a higher odds of failure after the first attempt of weaning (spontaneous breathing trial or gradual reduction of ventilatory support; OR 1.14 [CI 95% 1.01-1.30; p = 0.03]), and a higher odds for reintubation (OR 1.41 [CI 95% 1.20-1.66; p < 0.001]) than patients without brain injury. Patients with hemorrhagic stroke had the highest odds for unplanned extubation (OR 1.47 [CI 95% 1.22-1.77; p < 0.001]), of failed extubation after the first attempt of weaning (OR 1.28 [CI 95% 1.06-1.55; p = 0.009]), and for reintubation (OR 1.49 [CI 95% 1.17-1.88; p < 0.001]). In relation to weaning evolution over time in patients with acute brain injury, the risk for unplanned extubation showed a downward trend; the risk for reintubation was not associated to time; and there was a significant increase in the percentage of patients who underwent extubation after the first attempt of weaning from MV. CONCLUSIONS: Patients with acute brain injury, compared with patients without brain injury, present higher odds of undergoing unplanned extubated after weaning was started, lower odds of being extubated after the first attempt, and a higher risk of reintubation.

Evolution Over Time of Ventilatory Management and Outcome of Patients With Neurologic Disease.

OBJECTIVES: To describe the changes in ventilator management over time in patients with neurologic disease at ICU admission and to estimate factors associated with 28-day hospital mortality. DESIGN: Secondary analysis of three prospective, observational, multicenter studies. SETTING: Cohort studies conducted in 2004, 2010, and 2016. PATIENTS: Adult patients who received mechanical ventilation for more than 12 hours. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among the 20,929 patients enrolled, we included 4,152 (20%) mechanically ventilated patients due to different neurologic diseases. Hemorrhagic stroke and brain trauma were the most common pathologies associated with the need for mechanical ventilation. Although volume-cycled ventilation remained the preferred ventilation mode, there was a significant (p < 0.001) increment in the use of pressure support ventilation. The proportion of patients receiving a protective lung ventilation strategy was increased over time: 47% in 2004, 63% in 2010, and 65% in 2016 (p < 0.001), as well as the duration of protective ventilation strategies: 406 days per 1,000 mechanical ventilation days in 2004, 523 days per 1,000 mechanical ventilation days in 2010, and 585 days per 1,000 mechanical ventilation days in 2016 (p < 0.001). There were no differences in the length of stay in the ICU, mortality in the ICU, and mortality in hospital from 2004 to 2016. Independent risk factors for 28-day mortality were age greater than 75 years, Simplified Acute Physiology Score II greater than 50, the occurrence of organ dysfunction within first 48 hours after brain injury, and specific neurologic diseases such as hemorrhagic stroke, ischemic stroke, and brain trauma. CONCLUSIONS: More lung-protective ventilatory strategies have been implemented over years in neurologic patients with no effect on pulmonary complications or on survival. We found several prognostic factors on mortality such as advanced age, the severity of the disease, organ dysfunctions, and the etiology of neurologic disease.

Machine learning predicts mortality based on analysis of ventilation parameters of critically ill patients: multi-centre validation.

BACKGROUND: Mechanical Ventilation (MV) is a complex and central treatment process in the care of critically ill patients. It influences acid-base balance and can also cause prognostically relevant biotrauma by generating forces and liberating reactive oxygen species, negatively affecting outcomes. In this work we evaluate the use of a Recurrent Neural Network (RNN) modelling to predict outcomes of mechanically ventilated patients, using standard mechanical ventilation parameters. METHODS: We performed our analysis on VENTILA dataset, an observational, prospective, international, multi-centre study, performed to investigate the effect of baseline characteristics and management changes over time on the all-cause mortality rate in mechanically ventilated patients in ICU. Our cohort includes 12,596 adult patients older than 18, associated with 12,755 distinct admissions in ICUs across 37 countries and receiving invasive and non-invasive mechanical ventilation. We carry out four different analysis. Initially we select typical mechanical ventilation parameters and evaluate the machine learning model on both, the overall cohort and a subgroup of patients admitted with respiratory disorders. Furthermore, we carry out sensitivity analysis to evaluate whether inclusion of variables related to the function of other organs, improve the predictive performance of the model for both the overall cohort as well as the subgroup of patients with respiratory disorders. RESULTS: Predictive performance of RNN-based model was higher with Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) of 0.72 (± 0.01) and Average Precision (AP) of 0.57 (± 0.01) in comparison to RF and LR for the overall patient dataset. Higher predictive performance was recorded in the subgroup of patients admitted with respiratory disorders with AUC of 0.75 (± 0.02) and AP of 0.65 (± 0.03). Inclusion of function of other organs further improved the performance to AUC of 0.79 (± 0.01) and AP 0.68 (± 0.02) for the overall patient dataset and AUC of 0.79 (± 0.01) and AP 0.72 (± 0.02) for the subgroup with respiratory disorders. CONCLUSION: The RNN-based model demonstrated better performance than RF and LR in patients in mechanical ventilation and its subgroup admitted with respiratory disorders. Clinical studies are needed to evaluate whether it impacts decision-making and patient outcomes. TRIAL REGISTRATION: NCT02731898 ( https://clinicaltrials.gov/ct2/show/NCT02731898 ), prospectively registered on April 8, 2016.

Characteristics and outcomes of ventilated patients according to time to liberation from mechanical ventilation.

RATIONALE: A new classification of patients based on the duration of liberation of mechanical ventilation has been proposed. OBJECTIVES: To analyze outcomes based on the new weaning classification in a cohort of mechanically ventilated patients. METHODS: Secondary analysis included 2,714 patients who were weaned and underwent scheduled extubation from a cohort of 4,968 adult patients mechanically ventilated for more than 12 hours. MEASUREMENTS AND MAIN RESULTS: Patients were classified according to a new weaning classification: 1,502 patients (55%) as simple weaning,1,058 patients (39%) as difficult weaning, and 154 (6%) as prolonged weaning.Variables associated with prolonged weaning(.7d)were: severity at admission (odds ratio [OR] per unit of Simplified Acute Physiology Score II, 1.01; 95% confidence interval [CI], 1.001­1.02), duration of mechanical ventilation before first attempt of weaning (OR per day, 1.10; 95% CI, 1.06­1.13), chronic pulmonary disease other than chronic obstructive pulmonary disease (OR,13.23; 95% CI, 3.44­51.05), pneumonia as the reason to start mechanical ventilation (OR, 1.82; 95% CI, 1.07­3.08), and level of positive end-expiratory pressure applied before weaning (OR per unit,1.09; 95% CI, 1.04­1.14). The prolonged weaning group had a nonsignificant trend toward a higher rate of reintubation (P » 0.08),tracheostomy (P » 0.15), and significantly longer length of stay and higher mortality in the intensive care unit (OR for death, 1.97;95%CI, 1.17­3.31). The adjusted probability of death remained constant until Day 7, at which point it increased to 12.1%.

Propensity-Adjusted Comparison of Mortality of Elderly Versus Very Elderly Ventilated Patients.

BACKGROUND: The growing proportion of elderly intensive care patients constitutes a public health challenge. The benefit of critical care in these patients remains unclear. We compared outcomes in elderly versus very elderly subjects receiving mechanical ventilation. METHODS: In total, 5,557 mechanically ventilated subjects were included in our post hoc retrospective analysis, a subgroup of the VENTILA study. We divided the cohort into 2 subgroups on the basis of age: very elderly subjects (age ≥ 80 y; n = 1,430), and elderly subjects (age 65-79 y; n = 4,127). A propensity score on being very elderly was calculated. Evaluation of associations with 28-d mortality was done with logistic regression analysis. RESULTS: Very elderly subjects were clinically sicker as expressed by higher SAPS II scores (53 ± 18 vs 50 ± 18, P < .001), and their rates of plateau pressure < 30 cm H2O were higher, whereas other parameters did not differ. The 28-d mortality was higher in very elderly subjects (42% vs 34%, P < .001) and remained unchanged after propensity score adjustment (adjusted odds ratio 1.31 [95% CI 1.16-1.49], P < .001). CONCLUSIONS: Age was an independent and unchangeable risk factor for death in mechanically ventilated subjects. However, survival rates of very elderly subjects were > 50%. Denial of critical care based solely on age is not justified. (ClinicalTrials.gov registration NCT02731898.).

Driving Pressure Is a Risk Factor for ARDS in Mechanically Ventilated Subjects Without ARDS.

BACKGROUND: Driving pressure (ΔP) has been described as a risk factor for mortality in patients with ARDS. However, the role of ΔP in the outcome of patients without ARDS and on mechanical ventilation has received less attention. Our objective was to evaluate the association between ΔP on the first day of mechanical ventilation with the development of ARDS. METHODS: This was a post hoc analysis of a multicenter, prospective, observational, international study that included subjects who were on mechanical ventilation for > 12 h. Our objective was to evaluate the association between ΔP on the first day of mechanical ventilation with the development of ARDS. To assess the effect of ΔP, a logistic regression analysis was performed when adjusting for other potential risk factors. Validation of the results obtained was performed by using a bootstrap method and by repeating the same analyses at day 2. RESULTS: A total of 1,575 subjects were included, of whom 65 (4.1%) developed ARDS. The ΔP was independently associated with ARDS (odds ratio [OR] 1.12, 95% CI 1.07-1.18 for each cm H2O of ΔP increase, P < .001). The same results were observed at day 2 (OR 1.14, 95% CI 1.07-1.21; P < .001) and after bootstrap validation (OR 1.13, 95% CI 1.04-1.22; P < .001). When taking the prevalence of ARDS in the lowest quartile of ΔP (≤9 cm H2O) as a reference, the subjects with ΔP > 12-15 cm H2O and those with ΔP > 15 cm H2O presented a higher probability of ARDS (OR 3.65, 95% CI 1.32-10.04 [P = .01] and OR 7.31, 95% CI, 2.89-18.50 [P < .001], respectively). CONCLUSIONS: In the subjects without ARDS, a higher level of ΔP on the first day of mechanical ventilation was associated with later development of ARDS. (ClinicalTrials.gov registration NCT02731898.).

Inter-country variability over time in the mortality of mechanically ventilated patients.

PURPOSE: Variations in clinical characteristics and management and in the mortality of mechanically ventilated patients have not been sufficiently evaluated. We hypothesized that mortality shows a variability associated with country after adjustment for clinical characteristics and management. METHODS: Analysis of four studies carried out at 6-year intervals over an 18-year period. The studies included 26,024 patients (5183 in 1998, 4968 in 2004, 8108 in 2010, and 7765 in 2016) admitted to 1253 units from 38 countries. The primary outcome was 28-day mortality. We performed analyses using multilevel logistic modeling with mixed-random effects, including country as a random variable. To evaluate the effect of management strategies on mortality, a mediation analysis was performed. RESULTS: Adjusted 28-day mortality decreased significantly over time (first study as reference): 2004: odds ratio 0.82 (95% confidence interval [CI] 0.72-0.93); 2010: 0.63 (95% CI 0.53-0.75); 2016: 0.49 (95% CI 0.39-0.61). A protective ventilatory strategy and the use of continuous sedation mediated a moderate fraction of the effect of time on mortality in patients with moderate hypoxemia and without hypoxemia, respectively. Logistic multilevel modeling showed a significant effect of country on mortality: median odds ratio (MOR) in 1998: 2.02 (95% CI 1.57-2.48); in 2004: 1.76 (95% CI 1.47-2.06); in 2010: 1.55 (95% CI 1.37-1.74), and in 2016: 1.39 (95% CI 1.25-1.54). CONCLUSIONS: These findings suggest that country could contribute, independently of confounder variables, to outcome. The magnitude of the effect of country decreased over time. Clinical trials registered with http://www.clinicaltrials.gov (NCT02731898).

Evolution of mechanical ventilation in response to clinical research.

RATIONALE: Recent literature in mechanical ventilation includes strong evidence from randomized trials. Little information is available regarding the influence of these trials on usual clinical practice. OBJECTIVES: To describe current mechanical ventilation practices and to assess the influence of interval randomized trials when compared with findings from a 1998 cohort. METHODS: A prospective international observational cohort study, with a nested comparative study performed in 349 intensive care units in 23 countries. We enrolled 4,968 consecutive patients receiving mechanical ventilation over a 1-month period. We recorded demographics and daily data related to mechanical ventilation for the duration of ventilation. We systematically reviewed the literature and developed 11 practice-change hypotheses for the comparative cohort study before seeing these results. In assessing practice changes, we only compared data from the 107 intensive care units (1,675 patients) that also participated in the 1998 cohort (1,383 patients). MEASUREMENTS AND MAIN RESULTS: In 2004 compared with 1998, the use of noninvasive ventilation increased (11.1 vs. 4.4%, P < 0.001). Among patients with acute respiratory distress syndrome, tidal volumes decreased (7.4 vs. 9.1 ml/kg, P < 0.001) and positive end-expiratory pressure levels increased slightly (8.7 vs. 7.7 cm H(2)O, P = 0.02). More patients were successfully extubated after their first attempt of spontaneous breathing (77 vs. 62%, P < 0.001). Use of synchronized intermittent mandatory ventilation fell dramatically (1.6 vs. 11%, P < 0.001). Observations confirmed 10 of our 11 practice-change hypotheses. CONCLUSIONS: The strong concordance of predicted and observed practice changes suggests that randomized trial results have advanced mechanical ventilation practices internationally.

Easy prognostic assessment of concomitant organ failure in critically ill patients undergoing mechanical ventilation.

BACKGROUND: Acute respiratory distress syndrome (ARDS) is a life-threatening disease. We evaluated the prognostic utility of Model for End-stage Liver Disease excluding INR (MELD-XI) score for predicting mortality in a cohort of critically ill patients on mechanical ventilation. METHODS: In total, 11,091 mechanically ventilated patients were included in our post-hoc retrospective analysis, a subgroup of the VENTILA study (NCT02731898). Evaluation of associations with mortality was done by logistic and Cox regression analysis, an optimal cut-off was calculated using the Youden Index. We divided the cohort in two sub-groups based on their MELD-XI score at the optimal cut-off (12 score points). RESULTS: Peak-, plateau- and positive end-expiratory pressure were higher in patients with MELD-XI>12. Patients with MELD-XI>12 had higher driving pressures (14 ±â€¯6 cmH2O versus 13 ±â€¯6; p < 0.001). MELD-XI was associated with 28-day mortality after correction for relevant cofounders including SAPS II and ventilation pressures (HR 1.04 95%CI 1.03-1.05; p < 0.001. Patients with MELD-XI>12 evidenced both increased hospital (46% versus 27%; p < 0.001) and 28-day mortality (39% versus 22%). CONCLUSIONS: MELD-XI is independently associated with mortality and constitutes a useful and easily applicable tool for risk stratification in critically ill patients receiving mechanical ventilation. TRIAL REGISTRATION: NCT02731898, registered 4 April 2016.

Acute Kidney Injury in Mechanically Ventilated Patients: The Risk Factor Profile Depends on the Timing of Aki Onset.

BACKGROUND: Acute kidney injury (AKI) is a frequent complication in patients under mechanical ventilation (MV). We aimed to assess the risk factors for AKI with particular emphasis on those potentially preventable. STUDY DESIGN, SETTING, AND PARTICIPANTS: Retrospective analysis of a large, multinational database of MV patients with >24 h of MV and normal renal function at admission. AKI was defined according to creatinine-based KDIGO criteria. Risk factors were analyzed according to the time point at which AKI occurred: early (≤48 h after ICU admission, AKIE) and late (day 3 to day 7 of ICU stay, AKIL). A conditional logistic regression model was used to identify variables independently associated with AKI. RESULTS: Three thousand two hundred six patients were included. Seven hundred patients had AKI (22%), the majority of them AKIE (547/704). The risk factor profile was highly dependent upon the timing of AKI onset. In AKIE risk factors were older age; SAPS II score; postoperative and cardiac arrest as the reasons for MV; worse cardiovascular SOFA, pH, serum creatinine, and platelet count; higher level of peak pressure and Vt/kg; and fluid overload at admission. In contrast, AKIL was linked mostly to events that occurred after admission (lower platelet count and pH; ICU-acquired sepsis; and fluid overload). None ventilation-associated parameters were identify as risk factors for AKIL. CONCLUSIONS: In the first 48 h, risk factors are associated with the primary disease and the patient's condition at admission. Subsequently, emergent events like sepsis and organ dysfunction appear to be predictive factors making prevention a challenge.

Prognosis factors and outcome of community-acquired pneumonia needing mechanical ventilation.

PURPOSE: To evaluate the variables associated with mortality of patients with community-acquired pneumonia who require mechanical ventilation and to determine the attributable morbidity and intensive care unit (ICU) mortality of community-acquired pneumonia. MATERIAL AND METHODS: Retrospective cohort study carried out in 361 ICUs from 20 countries including 124 patients who required mechanical ventilation on the first day of admission to the hospital due to acute respiratory failure secondary to severe community-acquired pneumonia. To assess the factors associated with outcome, a forward stepwise logistic regression analysis was performed, and to determine the attributable mortality of community-acquired pneumonia, a matched study design was used. RESULTS: We found 3 independent variables significantly associated with death in patients with community-acquired pneumonia requiring mechanical ventilation: simplified acute physiological score greater than 45 (odds ratio, 5.5 [95% confidence interval, 1.7-12.3]), shock (odds ratio, 5.7 [95% confidence interval, 1.7-10.1]), and acute renal failure (odds ratio, 3.0 [95% confidence interval, 1.1-4.0]). There was no statistically significant difference in ICU mortality among patients with or without community-acquired pneumonia (32% vs 35%; P=.59). CONCLUSIONS: Community-acquired pneumonia needing mechanical ventilation is not a disease associated with higher mortality. The main determinants of patient outcome were initial severity of illness and the development of shock and/or acute renal failure.

An assessment of the Acute Kidney Injury Network creatinine-based criteria in patients submitted to mechanical ventilation.

BACKGROUND AND OBJECTIVES: The aim of our study was to assess the new diagnostic criteria of acute kidney injury (AKI) proposed by the Acute Kidney Injury Network (AKIN) in a large cohort of mechanically ventilated patients. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This is a prospective observational cohort study enrolling 2783 adult intensive care unit patients under mechanical ventilation (MV) with data on serum creatinine concentration (SCr) in the first 48 hours. The absolute and the relative AKIN diagnostic criteria (changes in SCr ≥ 0.3 mg/dl or ≥ 50% over the first 48 hours of MV, respectively) were analyzed separately. In addition, patients were classified into three groups according to their change in SCr (ΔSCr) over the first day on MV (ΔSCr): group 1, ΔSCr ≤ -0.3 mg/dl; group 2, ΔSCr between -0.3 and +0.29 mg/dl; and group 3, ΔSCr ≥ +0.3 mg/dl). The primary end point was in-hospital mortality, and secondary end points were intensive care unit and hospital length of stay, and duration of MV. RESULTS: Of 2783 patients, 803 (28.8%) had AKI according to both criteria: 431 only absolute (AKI(A)), 362 both relative and absolute (AKI(R+A)), and 10 only relative. The relative criterion identified more patients when baseline SCr (SCr0) was <0.9 mg/dl and the absolute when SCr0 was >1.5 mg/dl. The diagnosis of AKI was associated with mortality. CONCLUSIONS: Our study confirms the validity of the AKIN criteria in a population of mechanically patients and the criteria's relationship with the baseline SCr.

Outcome of mechanically ventilated patients who require a tracheostomy.

OBJECTIVE: To estimate the prevalence of, the risk factors associated with, and the outcome of tracheostomy in a heterogeneous population of mechanically ventilated patients. DESIGN: Prospective, observational cohort study. SETTING: A total of 361 intensive care units from 12 countries. PATIENTS: A cohort of 5,081 patients mechanically ventilated for >12 hrs. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A total of 546 patients (10.7%) had a tracheostomy during their stay in the intensive care unit. Tracheostomy was performed at a median time of 12 days (interquartile range, 7-17) from the beginning of mechanical ventilation. Variables associated with the performance of tracheostomy were duration of mechanical ventilation, need for reintubation, and neurologic disease as the primary reason of mechanical ventilation. The intensive care unit stay of patients with or without tracheostomy was a median of 21 days (interquartile range, 12-32) vs. 7 days (interquartile range, 4-12; p < .001), respectively, and the hospital stay was a median 36 days (interquartile range, 23-53) vs. 15 days (interquartile range, 8-26; p < .001), respectively. Adjusting by other variables, tracheostomy was independently related with survival in the intensive care unit (odds ratio, 2.22; 95% confidence interval, 1.72-2.86). Mortality in the hospital was similar in both groups (39% vs. 40%, p = .65). CONCLUSIONS: Tracheostomy is a common surgical procedure in the intensive care unit that is associated with a lower mortality in the unit but with a longer stay and a similar mortality in the hospital than in patients without tracheostomy.