Exploring the Impact of Mechanical Power on Mortality and Phenotypes in ARDS Patients: A Retrospective Analysis

Exploring the Impact of Mechanical Power on Mortality and Phenotypes in ARDS Patients: A Retrospective Analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Exploring the Impact of Mechanical Power on Mortality and Phenotypes in ARDS Patients: A Retrospective Analysis Qi Zhang, Na Liu, Fan Wang, Huiyong Wang, Renshuang Ding, Yan Li, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4441850/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In this study, we investigated the effect of mechanical power (MP) on mortality in acute respiratory distress syndrome (ARDS) patients. Patients diagnosed with ARDS were identified from the MIMIC-IV database. Kaplan-Meier curves and Cox proportional hazards models were utilized for survival analysis. The optimal cut-off value for MP was determined by using 'survminer' package. Causal mediation analysis (CMA) further investigated the effect of MP on 28-day mortality. Key predictive indicators were used to cluster and identify characteristics of different phenotypes. A total of 1333 patients were included. MP lower than 18.7J/min was associated with reduced mortality. Arterial pH and P/F ratio separately accounted for 29.2% and 20% of the mediating effect of high MP on increased 28-day mortality. Clustering analysis showed that phenotype-I had the worst respiratory mechanical parameters and the highest 28-day mortality. Phenotype-II was correlated with less organ dysfunction, the best oxygenation index and lower mechanical ventilation hours. Phenotype-III had the most laboratory abnormalities, the worse P/F ratio and longer ICU staytime. MP is strongly associated with mortality of ARDS patients belong to phenotype-III. High MP is independently associated with increased mortality in patients with ARDS. MP of less than 18.7 J/min is safer for ARDS patients. Health sciences/Medical research Health sciences/Risk factors Acute Respiratory Distress Syndrome Mechanical ventilation Mechanical power mortality consensus clustering analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Acute respiratory distress syndrome (ARDS) is an acute diffuse inflammatory injury of the lung caused by acute inducing factors, which leads to damage to the alveolar epithelium and pulmonary capillary endothelium, increases capillary permeability, and then causes pulmonary edema, gravity-dependent atelectasis, and reduced effective lung volume. In addition, it also leads to increased shunt and dead space, and presents a variety of pathological manifestations, including inflammatory response, hyaline membrane formation, and alveolar hemorrhage[1,2]. It is a common clinical problem in critically ill patients worldwide, and its mortality rate is still as high as 30%-45%[3]. Mechanical ventilation has certain advantages in improving respiratory function and correcting hypoxemia, which is considered as the main life support strategy for ARDS patients[4]. Improper ventilator settings can increase the risk of ventilator-induced lung injury (VILI). In recent years, there has been more interest in quantifying and monitoring the mechanical energy exerted on the lungs by ventilators. One approach is to calculate mechanical power, a parameter that the mechanical energy transmitted from the ventilator to the respiratory system during ventilation [5]. It is currently recognized as the determinants of VILI, taking into account tidal volume, respiratory rate, driving pressure and other indicators. The prognosis of patients with ARDS is closely related to the size of mechanical energy during mechanical ventilation. Serpa et al. [6] found that mechanical energy > 17.0 J/min was independently associated with high ICU mortality and 30-day mortality through analysis of 8207 critically ill patients included in MIMIC-III and eICU databases. Zhang et al. [7] found that using ideal weight standardized mechanical energy to predict mortality in patients with ARDS was better than absolute mechanical energy. In the treatment of mechanical ventilation for ARDS, reasonable control of the size of mechanical energy is of great significance for improving the prognosis of patients. Despite the pivotal role of mechanical power in ARDS management, the literature reveals a gap in comprehensive understanding and utilization of mechanical ventilation strategies to optimize patient outcomes. Currently, there are few studies on mechanical power in patients with ARDS. Therefore, this study took patients with invasive mechanical ventilation in the MIMIC-IV database as the research object to explore the effect of mechanical power on ARDS patients and the prognosis of patients with different ARDS phenotypes after clustering. Methods Study design This study was a retrospective analysis based on the MIMIC-IV database, which included data from patients with acute respiratory distress syndrome treated in the intensive care unit at Beth Israel Deaconess Medical Center from 2008 to 2019. One author has completed the courses and tests of the Collaborating Institution Training Program for "Data or Specimen Studies" and has obtained a license to use the MIMIC database (Number: 9535772). Because the patient privacy information in the database has been anonymized, this study does not require ethical approval. Patients selection Patients in the MIMIC-IV database who were older than 18 years, received invasive mechanical ventilation for more than 24 hours, and met the 2012 Berlin definition of ARDS were included[ 8 ]. The Berlin defnition including bilateral opacities on chest imaging, oxygenation index less than 300 mmHg with positive end-expiratory pressure (PEEP) ≥ 5cmH 2 O, and absence of heart failure. Exclusion criteria were as follows: (1) patients younger than 18 years; (2) received less than 24 hours of invasive mechanical ventilation; (3) patients who died within the first 24 hours after ARDS diagnosis; (4) patients that did not sufficiently capture the ventilatory variables needed to calculate MP were excluded. Only the first admission data were used when patients were admitted multiply. Variable extraction The collected data included:(1) Demographic characteristics: age, height, weight and body mass index (BMI). (2) Severity Scores: sequential organ failure assessment (SOFA score), acute physiology score-III (APS-III). (3) Vital signs: systolic blood pressure (ABPs), mean arterial pressure (ABPm) and diastolic blood pressure (ABPd). (4) Respiratory characteristics: tidal volume, spontaneous respiratory rate, total respiratory rate, set respiratory rate, PEEP, the plateau pressure, peak airway pressure, and mechanical ventilation time. (5) Lab events: white blood cell (WBC), platelet (PLT), blood urea nitrogen (BUN), blood glucose, creatinine, and lactate level (Lac). (6) Blood gas analysis parameters: oxygenation index on the first day (PF ratio), arterial pH value, arterial partial pressure of oxygen (PaO2), arterial partial pressure of carbon dioxide (PaCO2), and HCO3-. (7) Outcomes: 28-day mortality, ICU mortality, in-hospital mortality, and length of ICU stay. ll data were extracted using Google Cloud BigQuery and Structured Query Language (SQL) in PostgreSQL (version 12.0). Indicators with missing values exceeding 35% of the total were excluded. Outliers were defined as values outside the 1.5-fold interquartile range. After excluding outliers, the mean value of the data within 24 hours after admission was calculated. The Miss Forest model was then employed to impute missing values (Additional file : Figure S1 ). Mechanical power Mechanical power (MP) was calculated by tidal volume (Vt), peak airway pressure (Ppeak), total respiratory rate (RR), and driving pressure (ΔP) as proposed previously [MP = 0.098×RR×Vt×(Ppeak-1/2ΔP)][ 9 ]. And driving pressure was calculated by the plateau pressure (Pplat) and PEEP (ΔP = Pplat-PEEP)[ 10 ]. Statistical analysis R version 4.3.1 software was performed for statistical analyses. The influence of mechanical power on the prognosis of patients with ARDS was investigated using univariate Logistic regression and univariate Cox proportional hazards regression models. Survival curves were plotted using the Kaplan-Meier method, and differences in survival status among various groups were compared using the Log-rank test. A p-value of less than 0.05 was considered statistically significant. The optimal cutoff value for continuous variables in survival analysis was determined using the surv_cutpoint() function from the survminer package, based on Maximally selected rank statistics calculated using the maxstat.test function. The patients were divided into the high MP group and the low MP group according to the optimal cutoff value. Comparisons between groups were made using the t-test, Wilcoxon’s test, and χ 2 test. Data that followed the normal distribution were presented as mean ± standard deviation, and the comparison of means between the two groups was performed using a t-test. For Non-normal continuous data, median (interquartile range) was used for representation, and the comparison between the two groups was carried out using Wilcoxon’s test. Data that fitted categorical variables were expressed as number of cases (rate), and the comparison between the two groups was assessed using χ 2 test. Comparisons between groups were made using the t-test or Wilcoxon’s test. The propensity score matching was conducted with a 1:1 matching and 0.05 caliper width in our study to balance covariates, including Age, APSIII, SOFA score, BMI, WBC, Creatinine, Glucose, BUN, PlateletCount, ABPd, ABPm, ABPs, Lac, and Heart rate. Logistic regression was then used to assess the impact of MP on mortality in the PSM cohort. Causal mediation analysis (CMA) is a statistical method that can distinguish the total effect of treatment into direct and indirect effects. It aims to identify and quantify the mechanisms that underlie the relationship between the independent and dependent variables. If the independent variable X can influence Y through a variable M, then M is called an intermediary variable. After M is controlled, the effect of X on Y is the average direct effect (ADE). The effect of X on M and the effect of M on Y are collectively called the average causal mediation effect (ACME)[ 11 ]. In our study, we hypothesized a particular indicator was a potential mediator. The different levels of MP might lead to changes in the indicator and such changes were associated with ARDS patients' mortality. Finally, arterial PH, partial pressure of arterial oxygen, partial pressure of arterial carbon dioxide, PF ratio, tidal volume, spontaneous respiratory rate, total respiratory rate, set respiratory rate, PEEP, peak airway pressure, and peak pressure were further investigated for the presence of mediating effects because of their differences between the high and low MP groups. ACME, ADE, and the total effect obtained through CMA to confirm our hypothesis that MP affects mortality through particular mediators. The variables with right-skewed distribution were logarithmically transformed to stabilize the variance and were standardized. The optimal number of clusters and distribution of K-means clustering under resampling conditions were evaluated by consensus clustering (CC), with a subsampling ratio set at 80%, 1000 iterations, and exploring cluster numbers (k) from 2 to 6. The determination of the best cluster number was based on the analysis of the heatmap of the Consistency Matrix (CM), the Cumulative Distribution Function (CDF), and the intra-cluster consistency scores. The intra-cluster consistency score, representing the average consistency value of all sample pairs within the same cluster, approaching 1 indicates higher cluster stability. Simultaneously, the the proportion of ambiguously clustered pairs (PAC) was employed to automatically determine the optimal number of clusters, with a PAC value close to 0 indicating better clustering stability. After the optimal cluster was determined, the results were visualized using t-distribution random neighbor embedding (t-SNE) and line graphs[ 12 , 13 ]. We investigated the association between clinical features of different clustering phenotypes and mortality, with clinical features represented as percentages, mean ± standard deviation, or median and inter quartile range. The comparisons between groups were performed using the chi-square test, variance analysis, and Wilcoxon’s test. All analyses were performed using R version 4.3.1. Results Patients The MIMIC-IV database contained 53,569 patients admitted to the ICU. Following the exclusion of individuals who underwent invasive mechanical ventilation for a duration of less than 24 hours, a cohort of 1,517 patients was identified as meeting the criteria outlined in the Berlin definition for ARDS in 2012. An additional 184 patients were disqualified due to the absence of essential data required for the calculation of mechanical power, culminating in a final sample size of 1333 patients. The flow diagram is presented in Fig. 1 . Higher mechanical power in ARDS patients is associated with increased ICU, in-hospital, and 28-day mortality. In the first 24 hours of ventilation, mechanical power was significantly associated with higher ICU mortality, in-hospital mortality, and 28-day mortality (Fig. 2 a). The optimal cut-off value for MP was determined to be 18.7 J/min using the Survminer package (Fig. 3 ). The best cutoff found in the ROC analyses was 18.8 J/min, but this had a poor predictive power [AUC was 0.553 (0.516–0.590), sensitivity was 47%, and specificity was 65%]( Additional File Figure S2 ). ARDS patients were categorized into high and low MP groups using 18.7 J/min as the threshold. Before matching, the groups showed differences in age, BMI, disease severity score, vital signs, laboratory parameters, and respiratory mechanics indicators, as presented in Table 1 . After 1:1 propensity score matching, a total of 778 patients (389 in the high MP group and 389 in the low MP group) were included. Post-matching, there were no statistically significant differences between the groups in terms of disease severity score, vital signs, and laboratory parameters ( P > 0.05) (Table 1 ). Additional File Figure S3 shows the changes in SMD before and after matching in the two groups of patients. In the matched cohort, MP is still significantly correlated with high ICU mortality, in-hospital mortality, and 28-day mortality (Fig. 2 b). Table 1 Comparison of the covariates between the low MP group and high MP group in original cohort and matched cohort. Original Cohort Matched Cohort MP ≤18.7J/min MP > 18.7J/min p MP ≤18.7J/min MP > 18.7J/min p N 832 501 389 389 Demographic characteristics Age (median [IQR]) 65.70 [55.00, 76.62] 59.10 [48.20, 68.70] < 0.001 59.40 [49.70, 69.30] 61.30 [50.90, 70.40] 0.284 BMI (median [IQR]) 27.80 [24.10, 31.52] 30.90 [26.30, 35.30] < 0.001 28.80 [24.90, 33.70] 30.10 [25.70, 33.90] 0.047 The severity of illness APSIII (median [IQR]) 67.00 [49.00, 86.00] 81.00 [59.00, 105.00] < 0.001 76.00 [56.00, 94.00] 73.00 [56.00, 98.00] 0.959 SOFA score (median [IQR]) 6.90 [5.00, 9.10] 8.20 [6.20, 11.00] < 0.001 8.00 [5.70, 10.20] 7.70 [5.80, 10.40] 0.962 Vital signs ABPd (median [IQR]) 57.45 [53.98, 61.00] 58.40 [54.60, 62.90] 0.009 58.40 [56.00, 62.50] 58.10 [54.00, 62.00] 0.163 ABPm (median [IQR]) 75.30 [73.00, 79.00] 75.00 [71.90, 79.00] 0.055 75.30 [73.00, 79.00] 75.00 [72.00, 79.00] 0.169 ABPs (median [IQR]) 112.40 [107.00, 119.00] 108.70 [103.20, 115.00] < 0.001 110.00 [105.00, 115.40] 109.00 [104.00, 116.00] 0.445 Heart Rate (median [IQR]) 84.90 [73.88, 96.50] 92.60 [78.30, 105.00] < 0.001 89.10 [77.70, 101.70] 89.10 [76.10, 102.00] 0.892 Laboratory test WBC (median [IQR]) 10.60 [8.40, 13.70] 11.20 [8.20, 14.60] 0.138 10.90 [8.30, 14.00] 11.10 [8.40, 14.50] 0.449 Glucose (median [IQR]) 125.50 [110.57, 148.57] 131.00 [115.00, 155.50] 0.003 129.00 [113.00, 153.00] 129.00 [114.40, 154.60] 0.705 Creatinine (median [IQR]) 1.00 [0.70, 1.60] 1.30 [0.90, 2.10] < 0.001 1.10 [0.70, 2.00] 1.20 [0.80, 2.00] 0.196 BUN (median [IQR]) 21.00 [14.88, 34.00] 25.40 [17.30, 38.00] < 0.001 22.00 [15.00, 37.80] 24.40 [17.00, 34.50] 0.351 Lac (median [IQR]) 1.80 [1.40, 2.40] 2.20 [1.50, 3.10] < 0.001 2.10 [1.60, 2.90] 2.10 [1.50, 2.90] 0.612 Platelet Count (median [IQR]) 173.00 [113.92, 225.52] 172.00 [107.50, 226.30] 0.708 166.70 [105.00, 221.70] 171.70 [113.00, 224.30] 0.466 Blood gas analysis parameters Arterial PH (median [IQR]) 7.40 [7.30, 7.40] 7.30 [7.30, 7.40] < 0.001 7.40 [7.30, 7.40] 7.30 [7.30, 7.40] < 0.001 PaO2 (median [IQR]) 113.85 [98.22, 133.83] 105.30 [89.00, 121.70] < 0.001 114.50 [97.70, 134.50] 106.00 [88.10, 121.00] < 0.001 PaCO2 (median [IQR]) 39.30 [35.00, 43.00] 40.80 [36.50, 47.00] < 0.001 38.50 [35.00, 42.30] 41.00 [36.60, 47.00] < 0.001 HCO3 (median [IQR]) 22.00 [19.50, 24.72] 21.00 [17.80, 24.50] < 0.001 21.00 [18.80, 24.00] 21.60 [18.00, 24.80] 0.382 PF ratio (median [IQR]) 213.90 [160.75, 270.00] 170.30 [127.80, 221.20] < 0.001 210.00 [158.00, 274.00] 172.10 [129.20, 224.60] < 0.001 Respiratory characteristics Tidal Volume (mean (SD)) 457.81 (70.97) 478.49 (71.86) < 0.001 461.95 (69.57) 482.77 (68.97) < 0.001 Spontaneous Respiratory Rate (median [IQR]) 1.10 [0.00, 4.23] 0.00 [0.00, 1.20] < 0.001 0.70 [0.00, 3.50] 0.00 [0.00, 1.30] < 0.001 Total Respiratory Rate (median [IQR]) 19.40 [16.88, 21.80] 25.50 [22.40, 28.30] < 0.001 20.20 [17.60, 22.30] 24.90 [22.20, 27.70] < 0.001 Set Respiratory Rate (median [IQR]) 17.60 [15.78, 20.00] 24.00 [20.20, 26.60] < 0.001 18.20 [16.00, 20.80] 23.00 [20.00, 26.00] < 0.001 PEEP (median [IQR]) 5.00 [5.00, 8.00] 9.70 [7.90, 10.70] < 0.001 5.00 [5.00, 8.30] 9.60 [8.00, 10.60] < 0.001 Plateau Pressure (median [IQR]) 18.70 [16.20, 21.50] 23.70 [21.00, 27.00] < 0.001 19.00 [16.30, 22.00] 23.20 [20.80, 26.20] < 0.001 Peak Pressure (median [IQR]) 21.70 [19.00, 24.40] 28.40 [25.60, 32.00] < 0.001 22.00 [19.40, 24.60] 28.10 [25.40, 31.40] < 0.001 Driving pressure (median [IQR]) 12.00 [10.00, 14.50] 14.20 [12.00, 16.60] < 0.001 12.30 [10.30, 14.60] 14.00 [11.90, 16.20] < 0.001 Mechanical power (median [IQR]) 13.70 [10.90, 16.10] 23.80 [21.10, 27.20] < 0.001 14.10 [11.40, 16.60] 23.60 [21.00, 26.70] < 0.001 Outcomes Mechanical ventilation hour (median [IQR]) 54.35 [35.00, 94.55] 80.00 [47.00, 133.70] < 0.001 60.00 [35.00, 110.60] 74.20 [43.40, 115.00] 0.003 ICU staydays (median [IQR]) 7.98 [4.74, 13.68] 9.85 [5.65, 15.83] 0.001 8.23 [4.98, 13.17] 8.37 [4.99, 13.99] 0.927 ICU mortality (%) 137 (16.47) 139 (27.74) < 0.001 68 (17.48) 116 (29.82) < 0.001 28-day mortality (%) 167 (20.07) 148 (29.54) < 0.001 80 (20.57) 123 (31.62) 0.001 Hospital mortality(%) 177 (21.27) 157 (31.34) < 0.001 84 (21.59) 127 (32.65) 0.001 Causal Mediation Analysis In our study, the mediating role of arterial PH and PF ratio in the association between mechanical power and mortality was found to be statistically significant. Figure 4 showed that in terms of 28-day mortality in ARDS patients, the arterial PH mediated 29.2% (95% CI 0.135–0.660; P 18.7 J/min (ACME: P 18.7 J/min (ACME: P = 0.010). Additional File Figure S4 showed that in terms of hospital mortality in ARDS patients, the PF ratio mediated 18.2% (95% CI 0.033–0.460; P = 0.018) of the effect of MP > 18.7 J/min (ACME: P = 0.020) and the arterial PH mediated 29.2% (95% CI 0.134–0.660; P 18.7 J/min (ACME: P = 0.032). Additional File Figure S5 showed that in terms of ICU mortality in ARDS patients, the arterial PH mediated 24.6% (95% CI 0.115- 0.500; P 18.7 J/min (ACME: P = 0.030). Cluster Analysis In this study, we identified three major sample clusters through consensus clustering analysis. The cumulative distribution function (CDF) graph showed the distribution of CDF across different numbers of clusters, with the Δarea graph revealing the largest changes in the area under the CDF curve occurring between k = 2 and k = 4 (Fig. 5 ). The heat map of the consensus matrix indicated clear boundaries at k = 2 and k = 3 ( Additional file Figure S6 ). The stability evaluation of the clustering results showed that the average cluster consistency scores were similar and both higher than 0.8 when k = 2 and k = 3 ( Additional file Figure S7 ), while the PAC value was the lowest at k = 3 (PAC = 0.051) ( Additional file Figure S8 , leading to the determination of three as the optimal number of clusters. In this study, three distinct clinical phenotypes were identified in patients with ARDS using consensus clustering analysis. The phenotypes were visualized using line and t-SNE plots ( Additional file Figure S9 and Fig. 6 ). The characteristics of the three phenotypes are presented in Additional file Table S2 . Phenotype-I (n = 483) exhibited the highest 28-day mortality rate at 29%. Patients in this phenotype had lower tidal volumes (mean: 404.69 ml), higher total respiratory rates (median: 23.00 bpm), higher plateau pressure (median: 22.10 cmH2O), higher peak pressure (median: 25.60 cmH2O), and the worst lung compliance (median: 29.40 ml/cmH2O). Phenotype-II (n = 434) had the lowest duration of mechanical ventilation (median: 52.20 hours) and a moderate 28-day mortality rate of 20.50%. Patients in this group exhibited the best P/F ratio (median: 278.30 mmHg) and the lowest Sequential Organ Failure Assessment (SOFA) score (median: 6.6). Phenotype-III (n = 416) had higher tidal volumes (mean: 530.15 ml), the highest BMI (median: 29.80 kg/m 2 ) and the best lung compliance (median: 41.60 ml/cmH 2 O). They also exhibited the most laboratory abnormalities, including higher white blood cell counts (median: 11.10 K/mcL), higher creatinine levels (median: 1.20 mg/dl), and higher lactic acid levels (median: 2.00 mmol/L). This phenotype had a moderate duration of mechanical ventilation (median: 63.8 hours). The relationship between phenotypes and outcomes. The three phenotypes exhibited distinct clinical outcomes, with significant differences observed in mortality, mechanical ventilation time, and ICU stay lengths ( p < 0.05, Additional file Table S2 ). Patients in Phenotype II had shorter durations of mechanical ventilation compared to those in Phenotypes I and III. Those assigned to Phenotype I had the longest ICU stays, with a median duration of 9.13 days. Additionally, Phenotype I was associated with the highest rates of ICU mortality, in-hospital mortality, and 28-day mortality. Heterogeneity of mechanical power effects. The relationship between mechanical power and prognosis was investigated across the three phenotypes. Logistic regression analysis revealed that in Phenotype III, an increase in mechanical power was significantly associated with higher 28-day mortality, in-hospital mortality, and ICU mortalitym of ARDS patients, with odds ratios (OR) and 95% confidence intervals (CI) being 1.050 (1.018, 1.083), 1.053 (1.021, 1.085), and 1.057 (1.023, 1.091) ( p < 0.05). No significant association was found between MP and mortality in Phenotype II patients. In patients with Phenotype I, an increase in MP was significantly associated with higher in-hospital mortality and ICU mortality, with OR and 95% CI being 1.028 (1.001, 1.056) and 1.031 (1.0033, 1.060) ( p < 0.05)(Table 2 ). Table 2 The relationship between mechanical power and prognosis in difference phenotypes. Variable Phenotype Odds Ratio Lower Bound 95% CI Upper Bound 95% CI p -value 28d mortality Phenotype-I 1.027 1.000 1.056 0.050 Phenotype-II 1.019 0.981 1.058 0.334 Phenotype-III 1.050 1.018 1.083 0.002 Hospital mortality Phenotype-I 1.028 1.001 1.056 0.042 Phenotype-II 1.009 0.972 1.047 0.635 Phenotype-III 1.053 1.021 1.085 0.001 ICU mortality Phenotype-I 1.031 1.003 1.060 0.030 Phenotype-II 1.034 0.994 1.076 0.098 Phenotype-III 1.057 1.023 1.091 0.001 Each phenotype was further divided into high and low MP subgroups, using 18.7 J/min as the threshold. The impact of MP varied across different phenotypes. In Phenotype I, compared to the low MP subgroup, patients with high MP had increased mechanical ventilation time, length of ICU stay, and ICU mortality. In Phenotype II, MP had no significant effect on mortality. In Phenotype III, MP less than 18.7 J/min was associated with reduced mechanical ventilation time, length of ICU stay, and mortality (Fig. 7 ). Sensitivity analyses We evaluated the impact of unmeasured confounding by sensitivity analysis parameter, which represents the correlation between the residuals of the mediator and outcome regressions. When ACME is 0, the value of ρ is between − 0.01 and − 0.02 and the 95% confidence interval of ρ includes 0. When ρ is greater than − 0.01 or less than − 0.02, the values of the average causal mediation effect being positive or negative remain unchanged. Therefore, the mediation effect in this study is robust ( Additional file Figure S10-12 ). Discussion The present study conducted a retrospective analysis of clinical data from the MIMIC-IV database of ARDS patients admitted to BIDMC from 2008 to 2019, with the objective of investigating the impact of mechanical power on the outcome of ARDS patients and the response of different phenotypes of ARDS patients to mechanical power. The findings of our study can be summary as follows: Firstly, there is a significant association between mechanical power in the first 24h and increased ICU, in-hospital, and 28-day mortality among ARDS patients. Secondly, causal mediation analysis revealed that arterial pH and PF ratio significantly mediated the relationship between mechanical power (MP) and mortality in ARDS patients, accounting for a substantial portion of effects on 28-day, hospital, and ICU mortality. Thirdly, our study identified three clinical phenotypes in ARDS patients using consensus clustering, revealing significant differences in mortality, mechanical ventilation duration, and ICU stays, and found varying associations between mechanical power and outcomes across three phenotypes. Mechanical power (MP) provides a more comprehensive way of evaluating ventilation load than traditional lung-protective ventilation strategies, and consists of three components: the power to overcome airway resistance during gas movement, the power to inflate the lung and chest wall movement, and the power to overcome lung and respiratory system end-expiratory pressure-related recoil[ 14 ]. This makes MP not only more effective in reducing lung injury by precisely controlling energy transfer during ventilation, but also able to provide individualized treatment according to the patient's specific conditions, which may improve clinical outcomes, such as reducing mortality and shortening the length of ICU stay. Von Wedel et al[ 15 ] conducted a study on 2,103 surgical patients admitted in ICU and found that high mechanical power were associated with increased 28-day mortality.Our findings demonstrate a clear correlation between higher MP values and increased mortality rates within the first 24 hours of ventilation. The identification of an optimal MP cutoff value at 18.7 J/min. Our causal mediation analysis provides a more detailed understanding of how mechanical power (MP) influences mortality, with arterial pH and PF ratio serving as significant mediators. Excessive mechanical power can lead to alveolar overdistension and lung injury, worsening gas exchange capacity and reducing the oxygenation index [ 16 ]. High MP can lead to lung damage and affect the ventilation function, resulting in the obstruction of carbon dioxide discharge and respiratory acidosis, which may reduce the pH of the arterial blood. Additionally, inflammation induced by lung injury may indirectly affect pH through metabolic abnormalities, emphasizing the importance of optimizing mechanical power settings based on specific patient conditions in clinical practice to improve gas exchange and maintain acid-base balance. In this study, we identification three distinct ARDS clinical phenotypes through consensus clustering analysis emphasizes the heterogeneity within ARDS patient populations and underscores the necessity of personalized mechanical ventilation strategies. Phenotype I, characterized by lower tidal volumes, higher total respiratory rates, and poorer lung compliance, was associated with the highest rates of mortality and ICU stays, highlighting the potential benefits of tailored mechanical power settings in this group to mitigate adverse outcomes. Conversely, Phenotype III, which exhibited higher tidal volumes and better lung compliance, demonstrated a positive response to lower mechanical power settings, underscoring the phenotype's potential resilience to ARDS's detrimental effects. These findings suggest that mechanical power, a modifiable factor during mechanical ventilation, plays a critical role in the prognosis of ARDS patients, with its impact varying significantly across different phenotypes. The differential effects of mechanical power across phenotypes indicate a nuanced relationship between mechanical ventilation settings and patient outcomes, advocating for a more individualized approach to ventilation management in ARDS. This approach could involve adjusting mechanical power to optimize patient-specific parameters such as lung compliance and tidal volume, potentially reducing mortality and improving overall outcomes. Furthermore, the mediation analysis revealing that arterial PH and PF ratio significantly mediate the effect of mechanical power on mortality highlights the importance of monitoring and managing these parameters in ARDS treatment protocols. In summary, this study's findings illuminate the critical need for personalized mechanical ventilation strategies in ARDS, informed by patient-specific phenotypes and key physiological parameters. By adopting such an approach, clinicians can potentially enhance the efficacy of ARDS management, reducing mortality and improving patient outcomes in this heterogeneously affected patient population. There are some limitations in this study. This study was a retrospective analysis based on the MIMIC-IV database, and the results may be limited by the quality and completeness of the recorded data. And external validation is further needed to establish the reliability and universality of the identified phenotypes and MP cut-offs in different populations and clinical settings. Conclusion In conclusion, our study revealed the critical role of mechanical power in ARDS outcomes, advocating for a appreciation of patient heterogeneity in the management of ARDS. Through a deeper understanding of the phenotypic differences and their interactions with mechanical ventilation parameters, clinicians can better navigate the challenges of providing optimal care for ARDS patients. Declarations Acknowledgements The authors extend their gratitude to the researchers who established and maintained the MIMIC-IV database. The opinions expressed in this study are solely those of the authors and do not represent the views of any affiliated third parties. Author contributions FMX contributed to obtaining funding and study supervision. ZQ and LN contributed to study selection, statistical analysis, and data extraction. ZQ contributed to the charts making and drafting of the article. WF and WHY contributed to the thorough review of statistical methodologies and analyses. DRS, LY, WZY, and LY contributed to the revision of the article for important intellectual content. The article writing and figure elaboration were completed by all the authors together. All authors read and approved the final manuscript. Funding This study was supported by the grants of the program of Hebei Province to introduce overseas students (grant number C20210353) and "333 Talent Project" talent training funding project (grant number A202103005). Competing interests The authors declare no competing interests. Data availability All data and material were available at https:// mimic. mit. edu/. Ethics declarations The patient privacy information in the MIMIC-IV database has been anonymized, this study does not require ethical approval. References Meyer NJ, Gattinoni L, Calfee CS. Acute respiratory distress syndrome. Lancet. 398, 622–637; 10.1016/S0140-6736(21)00439-6 (2021). Matthay MA, et al. A New Global Definition of Acute Respiratory Distress Syndrome. Am J Respir Crit Care Med. 209, 37–47; 10.1164/rccm.202303-0558WS (2024). Bellani G, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 315, 788–800; 10.1001/jama.2016.0291 (2016). Bos LDJ, Ware LB. Acute respiratory distress syndrome: causes, pathophysiology, and phenotypes. Lancet. 400, 1145–1156; 10.1016/S0140-6736(22)01485-4 (2022). Paudel R, et al. Mechanical Power: A New Concept in Mechanical Ventilation. Am J Med Sci. 362, 537–545; 10.1016/j.amjms.2021.09.004 (2021). Serpa Neto A, et al. Mechanical power of ventilation is associated with mortality in critically ill patients: an analysis of patients in two observational cohorts. Intensive Care Med . 44, 1914–1922; 10.1007/s00134-018-5375-6 (2018). Zhang Z, Zheng B, Liu N, Ge H, Hong Y. Mechanical power normalized to predicted body weight as a predictor of mortality in patients with acute respiratory distress syndrome. Intensive Care Med. 45, 856–864; 10.1007/s00134-019-05627-9 (2019). Ranieri VM, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 307, 2526–33; 10.1001/jama.2012.5669 (2012). Jimenez JV, et al. Electric impedance tomography-guided PEEP titration reduces mechanical power in ARDS: a randomized crossover pilot trial. Crit Care . 27, 21; 10.1186/s13054-023-04315-x (2023). Grotberg JC, Reynolds D, Kraft BD. Management of severe acute respiratory distress syndrome: a primer. Crit Care. 27, 289; 10.1186/s13054-023-04572-w (2023). Zhang Z, Zheng C, Kim C, Van Poucke S, Lin S, Lan P. Causal mediation analysis in the context of clinical research. Ann Transl Med. 4, 425; 10.21037/atm.2016.11.11 (2016). Pattharanitima P, et al. Machine Learning Consensus Clustering Approach for Patients with Lactic Acidosis in Intensive Care Units. J Pers Med. 11, 1132; 10.3390/jpm11111132 (2021). Liu X, et al. Identification of distinct clinical phenotypes of acute respiratory distress syndrome with differential responses to treatment. Crit Care. 25, 320; 10.1186/s13054-021-03734-y (2021). Hoppe K, Khan E, Meybohm P, Riese T. Mechanical power of ventilation and driving pressure: two undervalued parameters for pre extracorporeal membrane oxygenation ventilation and during daily management? Crit Care. 27, 111; 10.1186/s13054-023-04375-z (2023). von Wedel D, et al. Adjustments of Ventilator Parameters during Operating Room-to-ICU Transition and 28-Day Mortality. Am J Respir Crit Care Med. 209, 553–562; 10.1164/rccm . 202307-1168OC (2024). Liu XB, Zhu F. Reinterpreting the ventilator-induced lung injury from the prospective of mechanical power. Zhonghua Shao Shang Za Zhi. 37, 292–295; 10.3760/ cma.j.cn501120-20200203-00039 (2021). Additional Declarations No competing interests reported. Supplementary Files AdditionalFile.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4441850","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":310354118,"identity":"e3184cf9-2030-4c95-8d4e-dcc5c580ab09","order_by":0,"name":"Qi Zhang","email":"","orcid":"","institution":"Hebei Medical University Third hospital","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Zhang","suffix":""},{"id":310354119,"identity":"17a751d1-8e6f-4506-8463-eb911eb9e683","order_by":1,"name":"Na Liu","email":"","orcid":"","institution":"The Fourth hospital of Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Na","middleName":"","lastName":"Liu","suffix":""},{"id":310354120,"identity":"b6e17041-9133-4ea5-9a26-2c899f9864aa","order_by":2,"name":"Fan Wang","email":"","orcid":"","institution":"School of Public Health，Hebei Medical University","correspondingAuthor":false,"prefix":"","firstName":"Fan","middleName":"","lastName":"Wang","suffix":""},{"id":310354121,"identity":"6b2bc3cd-2587-48d2-8c19-9fc92d75fb87","order_by":3,"name":"Huiyong Wang","email":"","orcid":"","institution":"School of Information Science and Engineering, Hebei University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Huiyong","middleName":"","lastName":"Wang","suffix":""},{"id":310354122,"identity":"14a98008-5a51-4f67-a7c7-895b2c90cb54","order_by":4,"name":"Renshuang Ding","email":"","orcid":"","institution":"School of Information Science and Engineering, Hebei University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Renshuang","middleName":"","lastName":"Ding","suffix":""},{"id":310354123,"identity":"c242db9d-52e6-4922-b35c-d3a6879d0f6d","order_by":5,"name":"Yan Li","email":"","orcid":"","institution":"Hebei Medical University Third hospital","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Li","suffix":""},{"id":310354124,"identity":"529ae60d-d16e-4a9c-8498-c2bd41b8f694","order_by":6,"name":"Zhiyong Wang","email":"","orcid":"","institution":"Hebei Medical University Third hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhiyong","middleName":"","lastName":"Wang","suffix":""},{"id":310354125,"identity":"28a30e03-d23b-4870-8057-d5921dd8c297","order_by":7,"name":"Mingxing Fang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDklEQVRIiWNgGAWjYDACZubjHz/8kODhZ28++OBjA0MCWJQHnxZ2tjRmyR4LGcmeY8mGM4nSws9jxsDDVmFjcCPHTJiXGC26zWxpDyR4JHgYzhxLY7bdYZOnOyOB8cHbNgZ5cxxazA4zHzcosJDgYWxvPvY490xasdmNBGbDuW0MhjsbcGlhS5AA2cLMcyzdOLftcOK2Gwls0rxtDAkGB3Bp4TGQ4GEDoRwzacu2/yAt7L8JaDEDa+EBaWFsOwC2hRm/FrZkY8kekNOAgdzbllxsduZhs+SccxKGG3BpOX/44MMPP+rs7Y8Do/Jnm12e2fHkgx/elNnI47IFG2BsABISxKsfBaNgFIyCUYABAB75XZgWeNY3AAAAAElFTkSuQmCC","orcid":"","institution":"Hebei Medical University Third hospital","correspondingAuthor":true,"prefix":"","firstName":"Mingxing","middleName":"","lastName":"Fang","suffix":""},{"id":310354126,"identity":"bd5d0068-d494-4476-ad1c-3712bc761707","order_by":8,"name":"Yan Li","email":"","orcid":"","institution":"The Second Affiliated Hospital of Xi 'an Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yan","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2024-05-18 16:41:37","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4441850/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4441850/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":57954164,"identity":"a59408b9-d615-43e0-9b57-070a3150774b","added_by":"auto","created_at":"2024-06-07 23:13:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":110506,"visible":true,"origin":"","legend":"\u003cp\u003eFlow chart of the inclusion of the study population and grouping\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/083310c68c436174c8016601.png"},{"id":57954166,"identity":"1646b1ee-f985-4c89-818b-eed112a53a8a","added_by":"auto","created_at":"2024-06-07 23:13:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":111411,"visible":true,"origin":"","legend":"\u003cp\u003eMechanical power in the first day of ventilation and mortality. a Odds ratio represents the odds of death per 1 J/min increase in MP in original cohort. b Odds ratio represents the odds of death per 1 J/min increase in MP in matched cohort.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/167aada7a8108dcd241f4d48.png"},{"id":57954169,"identity":"b2d5e572-8c65-43ee-9738-176972fdb423","added_by":"auto","created_at":"2024-06-07 23:13:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":128620,"visible":true,"origin":"","legend":"\u003cp\u003eThe selection of optimal cut-off value of mechanical power.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/ec30f2c29f184fdc36e30236.png"},{"id":57955346,"identity":"10e5b129-ef10-4676-9ed4-afc51fee8ed8","added_by":"auto","created_at":"2024-06-07 23:29:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":61328,"visible":true,"origin":"","legend":"\u003cp\u003eCausal mediation analysis in 28-day mortality. A. CMA for arterial PH. B. CMA for PF ratio.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/b828d1b96adc25d521587307.png"},{"id":57954809,"identity":"e34a1b6a-c74f-48d3-9754-5d25d35b03a6","added_by":"auto","created_at":"2024-06-07 23:21:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":47352,"visible":true,"origin":"","legend":"\u003cp\u003eA.\u003cstrong\u003e \u003c/strong\u003eConsensus cumulative distribution function (CDF)distributions with different cluster numbers k from 2 to 6. B. Delta area of consensus CDF for number of clusters cluster numbers k from 2 to 6.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/933c894e91b1f68b4de0fe74.png"},{"id":57955347,"identity":"89fa3121-3ca8-4c04-8aa8-40df0a288e3a","added_by":"auto","created_at":"2024-06-07 23:29:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":227985,"visible":true,"origin":"","legend":"\u003cp\u003et-SNE visualization of phenotype assignments by consensus clustering.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/32ddc2f589586f68b77ac101.png"},{"id":57954812,"identity":"0d22ddf5-065b-4c72-b9a0-4decb61d2fae","added_by":"auto","created_at":"2024-06-07 23:21:52","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":155869,"visible":true,"origin":"","legend":"\u003cp\u003eHeterogeneity of low and high MP in different phenotypes.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/d8c4b773e054bae9222c0980.png"},{"id":82576130,"identity":"7069a798-ba7f-4666-a379-b12672683949","added_by":"auto","created_at":"2025-05-13 05:32:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1816400,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/720fec48-ff93-422d-99a0-da1a94bfc6ea.pdf"},{"id":57954172,"identity":"9f3e15c7-f3b4-46a4-9db7-3e2226033719","added_by":"auto","created_at":"2024-06-07 23:13:52","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1649620,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile.docx","url":"https://assets-eu.researchsquare.com/files/rs-4441850/v1/d0b5b56c7bb4ef54f83eaa89.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploring the Impact of Mechanical Power on Mortality and Phenotypes in ARDS Patients: A Retrospective Analysis","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAcute respiratory distress syndrome (ARDS) is an acute diffuse inflammatory injury of the lung caused by acute inducing factors, which leads to damage to the alveolar epithelium and pulmonary capillary endothelium, increases capillary permeability, and then causes pulmonary edema, gravity-dependent atelectasis, and reduced effective lung volume. In addition, it also leads to increased shunt and dead space, and presents a variety of pathological manifestations, including inflammatory response, hyaline membrane formation, and alveolar hemorrhage[1,2]. It is a common clinical problem in critically ill patients worldwide, and its mortality rate is still as high as 30%-45%[3]. Mechanical ventilation has certain advantages in improving respiratory function and correcting hypoxemia, which is considered as the main life support strategy for ARDS patients[4]. Improper ventilator settings can increase the risk of ventilator-induced lung injury (VILI).\u003c/p\u003e\n\u003cp\u003eIn recent years, there has been more interest in quantifying and monitoring the mechanical energy exerted on the lungs by ventilators. One approach is to calculate mechanical power, a parameter that the mechanical energy transmitted from the ventilator to the respiratory system during ventilation [5]. It is currently recognized as the determinants of VILI, taking into account tidal volume, respiratory rate, driving pressure and other indicators. The prognosis of patients with ARDS is closely related to the size of mechanical energy during mechanical ventilation. Serpa et al. [6] found that mechanical energy \u0026gt; 17.0 J/min was independently associated with high ICU mortality and 30-day mortality through analysis of 8207 critically ill patients included in MIMIC-III and eICU databases. Zhang et al. [7] found that using ideal weight standardized mechanical energy to predict mortality in patients with ARDS was better than absolute mechanical energy. In the treatment of mechanical ventilation for ARDS, reasonable control of the size of mechanical energy is of great significance for improving the prognosis of patients.\u003c/p\u003e\n\u003cp\u003eDespite the pivotal role of mechanical power in ARDS management, the literature reveals a gap in comprehensive understanding and utilization of mechanical ventilation strategies to optimize patient outcomes. Currently, there are few studies on mechanical power in patients with ARDS. Therefore, this study took patients with invasive mechanical ventilation in the MIMIC-IV database as the research object to explore the effect of mechanical power on ARDS patients and the prognosis of patients with different ARDS phenotypes after clustering.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eStudy design\u003c/h2\u003e \u003cp\u003eThis study was a retrospective analysis based on the MIMIC-IV database, which included data from patients with acute respiratory distress syndrome treated in the intensive care unit at Beth Israel Deaconess Medical Center from 2008 to 2019. One author has completed the courses and tests of the Collaborating Institution Training Program for \"Data or Specimen Studies\" and has obtained a license to use the MIMIC database (Number: 9535772). Because the patient privacy information in the database has been anonymized, this study does not require ethical approval.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003ePatients selection\u003c/h2\u003e \u003cp\u003ePatients in the MIMIC-IV database who were older than 18 years, received invasive mechanical ventilation for more than 24 hours, and met the 2012 Berlin definition of ARDS were included[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. The Berlin defnition including bilateral opacities on chest imaging, oxygenation index less than 300 mmHg with positive end-expiratory pressure (PEEP)\u0026thinsp;\u0026ge;\u0026thinsp;5cmH\u003csub\u003e2\u003c/sub\u003eO, and absence of heart failure. Exclusion criteria were as follows: (1) patients younger than 18 years; (2) received less than 24 hours of invasive mechanical ventilation; (3) patients who died within the first 24 hours after ARDS diagnosis; (4) patients that did not sufficiently capture the ventilatory variables needed to calculate MP were excluded. Only the first admission data were used when patients were admitted multiply.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eVariable extraction\u003c/h2\u003e \u003cp\u003eThe collected data included:(1) Demographic characteristics: age, height, weight and body mass index (BMI). (2) Severity Scores: sequential organ failure assessment (SOFA score), acute physiology score-III (APS-III). (3) Vital signs: systolic blood pressure (ABPs), mean arterial pressure (ABPm) and diastolic blood pressure (ABPd). (4) Respiratory characteristics: tidal volume, spontaneous respiratory rate, total respiratory rate, set respiratory rate, PEEP, the plateau pressure, peak airway pressure, and mechanical ventilation time. (5) Lab events: white blood cell (WBC), platelet (PLT), blood urea nitrogen (BUN), blood glucose, creatinine, and lactate level (Lac). (6) Blood gas analysis parameters: oxygenation index on the first day (PF ratio), arterial pH value, arterial partial pressure of oxygen (PaO2), arterial partial pressure of carbon dioxide (PaCO2), and HCO3-. (7) Outcomes: 28-day mortality, ICU mortality, in-hospital mortality, and length of ICU stay. ll data were extracted using Google Cloud BigQuery and Structured Query Language (SQL) in PostgreSQL (version 12.0). Indicators with missing values exceeding 35% of the total were excluded. Outliers were defined as values outside the 1.5-fold interquartile range. After excluding outliers, the mean value of the data within 24 hours after admission was calculated. The Miss Forest model was then employed to impute missing values \u003cb\u003e(Additional file : Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/b\u003e\u003c/p\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003eMechanical power\u003c/h2\u003e \u003cp\u003eMechanical power (MP) was calculated by tidal volume (Vt), peak airway pressure (Ppeak), total respiratory rate (RR), and driving pressure (ΔP) as proposed previously [MP\u0026thinsp;=\u0026thinsp;0.098\u0026times;RR\u0026times;Vt\u0026times;(Ppeak-1/2ΔP)][\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. And driving pressure was calculated by the plateau pressure (Pplat) and PEEP (ΔP\u0026thinsp;=\u0026thinsp;Pplat-PEEP)[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eR version 4.3.1 software was performed for statistical analyses. The influence of mechanical power on the prognosis of patients with ARDS was investigated using univariate Logistic regression and univariate Cox proportional hazards regression models. Survival curves were plotted using the Kaplan-Meier method, and differences in survival status among various groups were compared using the Log-rank test. A p-value of less than 0.05 was considered statistically significant. The optimal cutoff value for continuous variables in survival analysis was determined using the surv_cutpoint() function from the survminer package, based on Maximally selected rank statistics calculated using the maxstat.test function. The patients were divided into the high MP group and the low MP group according to the optimal cutoff value.\u003c/p\u003e \u003cp\u003eComparisons between groups were made using the t-test, Wilcoxon\u0026rsquo;s test, and \u003cem\u003eχ\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e test. Data that followed the normal distribution were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, and the comparison of means between the two groups was performed using a t-test. For Non-normal continuous data, median (interquartile range) was used for representation, and the comparison between the two groups was carried out using Wilcoxon\u0026rsquo;s test. Data that fitted categorical variables were expressed as number of cases (rate), and the comparison between the two groups was assessed using \u003cem\u003eχ\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e test. Comparisons between groups were made using the t-test or Wilcoxon\u0026rsquo;s test. The propensity score matching was conducted with a 1:1 matching and 0.05 caliper width in our study to balance covariates, including Age, APSIII, SOFA score, BMI, WBC, Creatinine, Glucose, BUN, PlateletCount, ABPd, ABPm, ABPs, Lac, and Heart rate. Logistic regression was then used to assess the impact of MP on mortality in the PSM cohort.\u003c/p\u003e \u003cp\u003eCausal mediation analysis (CMA) is a statistical method that can distinguish the total effect of treatment into direct and indirect effects. It aims to identify and quantify the mechanisms that underlie the relationship between the independent and dependent variables. If the independent variable X can influence Y through a variable M, then M is called an intermediary variable. After M is controlled, the effect of X on Y is the average direct effect (ADE). The effect of X on M and the effect of M on Y are collectively called the average causal mediation effect (ACME)[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In our study, we hypothesized a particular indicator was a potential mediator. The different levels of MP might lead to changes in the indicator and such changes were associated with ARDS patients' mortality. Finally, arterial PH, partial pressure of arterial oxygen, partial pressure of arterial carbon dioxide, PF ratio, tidal volume, spontaneous respiratory rate, total respiratory rate, set respiratory rate, PEEP, peak airway pressure, and peak pressure were further investigated for the presence of mediating effects because of their differences between the high and low MP groups. ACME, ADE, and the total effect obtained through CMA to confirm our hypothesis that MP affects mortality through particular mediators.\u003c/p\u003e \u003cp\u003eThe variables with right-skewed distribution were logarithmically transformed to stabilize the variance and were standardized. The optimal number of clusters and distribution of K-means clustering under resampling conditions were evaluated by consensus clustering (CC), with a subsampling ratio set at 80%, 1000 iterations, and exploring cluster numbers (k) from 2 to 6. The determination of the best cluster number was based on the analysis of the heatmap of the Consistency Matrix (CM), the Cumulative Distribution Function (CDF), and the intra-cluster consistency scores. The intra-cluster consistency score, representing the average consistency value of all sample pairs within the same cluster, approaching 1 indicates higher cluster stability. Simultaneously, the the proportion of ambiguously clustered pairs (PAC) was employed to automatically determine the optimal number of clusters, with a PAC value close to 0 indicating better clustering stability. After the optimal cluster was determined, the results were visualized using t-distribution random neighbor embedding (t-SNE) and line graphs[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. We investigated the association between clinical features of different clustering phenotypes and mortality, with clinical features represented as percentages, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation, or median and inter quartile range. The comparisons between groups were performed using the chi-square test, variance analysis, and Wilcoxon\u0026rsquo;s test. All analyses were performed using R version 4.3.1.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003ePatients\u003c/h2\u003e \u003cp\u003eThe MIMIC-IV database contained 53,569 patients admitted to the ICU. Following the exclusion of individuals who underwent invasive mechanical ventilation for a duration of less than 24 hours, a cohort of 1,517 patients was identified as meeting the criteria outlined in the Berlin definition for ARDS in 2012. An additional 184 patients were disqualified due to the absence of essential data required for the calculation of mechanical power, culminating in a final sample size of 1333 patients. The flow diagram is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eHigher mechanical power in ARDS patients is associated with increased ICU, in-hospital, and 28-day mortality.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eIn the first 24 hours of ventilation, mechanical power was significantly associated with higher ICU mortality, in-hospital mortality, and 28-day mortality (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The optimal cut-off value for MP was determined to be 18.7 J/min using the Survminer package (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The best cutoff found in the ROC analyses was 18.8 J/min, but this had a poor predictive power [AUC was 0.553 (0.516\u0026ndash;0.590), sensitivity was 47%, and specificity was 65%](\u003cb\u003eAdditional File Figure S2\u003c/b\u003e). ARDS patients were categorized into high and low MP groups using 18.7 J/min as the threshold. Before matching, the groups showed differences in age, BMI, disease severity score, vital signs, laboratory parameters, and respiratory mechanics indicators, as presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. After 1:1 propensity score matching, a total of 778 patients (389 in the high MP group and 389 in the low MP group) were included. Post-matching, there were no statistically significant differences between the groups in terms of disease severity score, vital signs, and laboratory parameters (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). \u003cb\u003eAdditional File Figure S3\u003c/b\u003e shows the changes in SMD before and after matching in the two groups of patients. In the matched cohort, MP is still significantly correlated with high ICU mortality, in-hospital mortality, and 28-day mortality (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of the covariates between the low MP group and high MP group in original cohort and matched cohort.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003e Original\u0026nbsp;Cohort\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eMatched\u0026nbsp;Cohort\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMP\u0026nbsp;\u0026le;18.7J/min\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMP\u0026nbsp;\u0026gt;\u0026nbsp;18.7J/min\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMP\u0026nbsp;\u0026le;18.7J/min\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMP\u0026nbsp;\u0026gt;\u0026nbsp;18.7J/min\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e832\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e501\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e389\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e389\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDemographic characteristics\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e65.70\u0026nbsp;[55.00,\u0026nbsp;76.62]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e59.10\u0026nbsp;[48.20,\u0026nbsp;68.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59.40\u0026nbsp;[49.70,\u0026nbsp;69.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e61.30\u0026nbsp;[50.90,\u0026nbsp;70.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.284\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e27.80\u0026nbsp;[24.10,\u0026nbsp;31.52]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e30.90\u0026nbsp;[26.30,\u0026nbsp;35.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e28.80\u0026nbsp;[24.90,\u0026nbsp;33.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e30.10\u0026nbsp;[25.70,\u0026nbsp;33.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.047\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eThe severity of illness\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPSIII\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e67.00\u0026nbsp;[49.00,\u0026nbsp;86.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e81.00\u0026nbsp;[59.00,\u0026nbsp;105.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e76.00\u0026nbsp;[56.00,\u0026nbsp;94.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e73.00\u0026nbsp;[56.00,\u0026nbsp;98.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.959\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSOFA score\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6.90\u0026nbsp;[5.00,\u0026nbsp;9.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8.20\u0026nbsp;[6.20,\u0026nbsp;11.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.00\u0026nbsp;[5.70,\u0026nbsp;10.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.70\u0026nbsp;[5.80,\u0026nbsp;10.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.962\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eVital signs\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eABPd\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e57.45\u0026nbsp;[53.98,\u0026nbsp;61.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e58.40\u0026nbsp;[54.60,\u0026nbsp;62.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e58.40\u0026nbsp;[56.00,\u0026nbsp;62.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e58.10\u0026nbsp;[54.00,\u0026nbsp;62.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.163\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eABPm\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e75.30\u0026nbsp;[73.00,\u0026nbsp;79.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e75.00\u0026nbsp;[71.90,\u0026nbsp;79.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.055\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e75.30\u0026nbsp;[73.00,\u0026nbsp;79.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e75.00\u0026nbsp;[72.00,\u0026nbsp;79.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.169\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eABPs\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e112.40\u0026nbsp;[107.00,\u0026nbsp;119.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e108.70\u0026nbsp;[103.20,\u0026nbsp;115.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e110.00\u0026nbsp;[105.00,\u0026nbsp;115.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e109.00\u0026nbsp;[104.00,\u0026nbsp;116.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.445\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeart Rate\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e84.90\u0026nbsp;[73.88,\u0026nbsp;96.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92.60\u0026nbsp;[78.30,\u0026nbsp;105.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e89.10\u0026nbsp;[77.70,\u0026nbsp;101.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e89.10\u0026nbsp;[76.10,\u0026nbsp;102.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.892\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLaboratory test\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10.60\u0026nbsp;[8.40,\u0026nbsp;13.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11.20\u0026nbsp;[8.20,\u0026nbsp;14.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.138\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.90\u0026nbsp;[8.30,\u0026nbsp;14.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11.10\u0026nbsp;[8.40,\u0026nbsp;14.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.449\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlucose\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e125.50\u0026nbsp;[110.57,\u0026nbsp;148.57]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e131.00\u0026nbsp;[115.00,\u0026nbsp;155.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e129.00\u0026nbsp;[113.00,\u0026nbsp;153.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e129.00\u0026nbsp;[114.40,\u0026nbsp;154.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.705\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCreatinine\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.00\u0026nbsp;[0.70,\u0026nbsp;1.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.30\u0026nbsp;[0.90,\u0026nbsp;2.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.10\u0026nbsp;[0.70,\u0026nbsp;2.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.20\u0026nbsp;[0.80,\u0026nbsp;2.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.196\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBUN\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21.00\u0026nbsp;[14.88,\u0026nbsp;34.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25.40\u0026nbsp;[17.30,\u0026nbsp;38.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22.00\u0026nbsp;[15.00,\u0026nbsp;37.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24.40\u0026nbsp;[17.00,\u0026nbsp;34.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.351\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLac\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.80\u0026nbsp;[1.40,\u0026nbsp;2.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.20\u0026nbsp;[1.50,\u0026nbsp;3.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.10\u0026nbsp;[1.60,\u0026nbsp;2.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.10\u0026nbsp;[1.50,\u0026nbsp;2.90]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.612\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlatelet Count\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e173.00\u0026nbsp;[113.92,\u0026nbsp;225.52]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e172.00\u0026nbsp;[107.50,\u0026nbsp;226.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.708\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e166.70\u0026nbsp;[105.00,\u0026nbsp;221.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e171.70\u0026nbsp;[113.00,\u0026nbsp;224.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.466\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eBlood gas analysis parameters\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArterial PH\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.40\u0026nbsp;[7.30,\u0026nbsp;7.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7.30\u0026nbsp;[7.30,\u0026nbsp;7.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7.40\u0026nbsp;[7.30,\u0026nbsp;7.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.30\u0026nbsp;[7.30,\u0026nbsp;7.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaO2 (median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e113.85\u0026nbsp;[98.22,\u0026nbsp;133.83]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e105.30\u0026nbsp;[89.00,\u0026nbsp;121.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e114.50\u0026nbsp;[97.70,\u0026nbsp;134.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e106.00\u0026nbsp;[88.10,\u0026nbsp;121.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaCO2 (median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39.30\u0026nbsp;[35.00,\u0026nbsp;43.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e40.80\u0026nbsp;[36.50,\u0026nbsp;47.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38.50\u0026nbsp;[35.00,\u0026nbsp;42.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e41.00\u0026nbsp;[36.60,\u0026nbsp;47.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHCO3\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.00\u0026nbsp;[19.50,\u0026nbsp;24.72]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.00\u0026nbsp;[17.80,\u0026nbsp;24.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e21.00\u0026nbsp;[18.80,\u0026nbsp;24.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e21.60\u0026nbsp;[18.00,\u0026nbsp;24.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.382\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePF ratio\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e213.90\u0026nbsp;[160.75,\u0026nbsp;270.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e170.30\u0026nbsp;[127.80,\u0026nbsp;221.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e210.00\u0026nbsp;[158.00,\u0026nbsp;274.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e172.10\u0026nbsp;[129.20,\u0026nbsp;224.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRespiratory characteristics\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTidal Volume\u0026nbsp;(mean\u0026nbsp;(SD))\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e457.81\u0026nbsp;(70.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e478.49\u0026nbsp;(71.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e461.95\u0026nbsp;(69.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e482.77\u0026nbsp;(68.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpontaneous Respiratory Rate\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.10\u0026nbsp;[0.00,\u0026nbsp;4.23]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.00\u0026nbsp;[0.00,\u0026nbsp;1.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.70\u0026nbsp;[0.00,\u0026nbsp;3.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.00\u0026nbsp;[0.00,\u0026nbsp;1.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal Respiratory Rate\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19.40\u0026nbsp;[16.88,\u0026nbsp;21.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25.50\u0026nbsp;[22.40,\u0026nbsp;28.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e20.20\u0026nbsp;[17.60,\u0026nbsp;22.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24.90\u0026nbsp;[22.20,\u0026nbsp;27.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSet Respiratory Rate\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17.60\u0026nbsp;[15.78,\u0026nbsp;20.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24.00\u0026nbsp;[20.20,\u0026nbsp;26.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18.20\u0026nbsp;[16.00,\u0026nbsp;20.80]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e23.00\u0026nbsp;[20.00,\u0026nbsp;26.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePEEP\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.00\u0026nbsp;[5.00,\u0026nbsp;8.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.70\u0026nbsp;[7.90,\u0026nbsp;10.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5.00\u0026nbsp;[5.00,\u0026nbsp;8.30]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9.60\u0026nbsp;[8.00,\u0026nbsp;10.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlateau Pressure\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.70\u0026nbsp;[16.20,\u0026nbsp;21.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.70\u0026nbsp;[21.00,\u0026nbsp;27.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e19.00\u0026nbsp;[16.30,\u0026nbsp;22.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e23.20\u0026nbsp;[20.80,\u0026nbsp;26.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeak Pressure\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21.70\u0026nbsp;[19.00,\u0026nbsp;24.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e28.40\u0026nbsp;[25.60,\u0026nbsp;32.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22.00\u0026nbsp;[19.40,\u0026nbsp;24.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e28.10\u0026nbsp;[25.40,\u0026nbsp;31.40]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDriving pressure\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12.00\u0026nbsp;[10.00,\u0026nbsp;14.50]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14.20\u0026nbsp;[12.00,\u0026nbsp;16.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12.30\u0026nbsp;[10.30,\u0026nbsp;14.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e14.00\u0026nbsp;[11.90,\u0026nbsp;16.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical power\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.70\u0026nbsp;[10.90,\u0026nbsp;16.10]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.80\u0026nbsp;[21.10,\u0026nbsp;27.20]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.10\u0026nbsp;[11.40,\u0026nbsp;16.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e23.60\u0026nbsp;[21.00,\u0026nbsp;26.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eOutcomes\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMechanical ventilation hour\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e54.35\u0026nbsp;[35.00,\u0026nbsp;94.55]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80.00\u0026nbsp;[47.00,\u0026nbsp;133.70]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e60.00\u0026nbsp;[35.00,\u0026nbsp;110.60]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e74.20\u0026nbsp;[43.40,\u0026nbsp;115.00]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eICU staydays\u0026nbsp;(median\u0026nbsp;[IQR])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.98\u0026nbsp;[4.74,\u0026nbsp;13.68]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.85\u0026nbsp;[5.65,\u0026nbsp;15.83]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8.23\u0026nbsp;[4.98,\u0026nbsp;13.17]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8.37\u0026nbsp;[4.99,\u0026nbsp;13.99]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.927\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eICU mortality\u0026nbsp;(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e137\u0026nbsp;(16.47)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e139\u0026nbsp;(27.74)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e68\u0026nbsp;(17.48)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e116\u0026nbsp;(29.82)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e28-day mortality\u0026nbsp;(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e167\u0026nbsp;(20.07)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e148\u0026nbsp;(29.54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e80\u0026nbsp;(20.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e123\u0026nbsp;(31.62)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital mortality(%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e177\u0026nbsp;(21.27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e157\u0026nbsp;(31.34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e84\u0026nbsp;(21.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e127\u0026nbsp;(32.65)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003eCausal Mediation Analysis\u003c/h2\u003e \u003cp\u003eIn our study, the mediating role of arterial PH and PF ratio in the association between mechanical power and mortality was found to be statistically significant. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e showed that in terms of 28-day mortality in ARDS patients, the arterial PH mediated 29.2% (95% CI 0.135\u0026ndash;0.660; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) of the effect of MP\u0026thinsp;\u0026gt;\u0026thinsp;18.7 J/min (ACME: \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and the PF ratio mediated 20% (95% CI 0.053\u0026ndash;0.480; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.010) of the effect of MP\u0026thinsp;\u0026gt;\u0026thinsp;18.7 J/min (ACME: \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.010). \u003cb\u003eAdditional File Figure S4\u003c/b\u003e showed that in terms of hospital mortality in ARDS patients, the PF ratio mediated 18.2% (95% CI 0.033\u0026ndash;0.460; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.018) of the effect of MP\u0026thinsp;\u0026gt;\u0026thinsp;18.7 J/min (ACME: \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.020) and the arterial PH mediated 29.2% (95% CI 0.134\u0026ndash;0.660; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) of the effect of MP\u0026thinsp;\u0026gt;\u0026thinsp;18.7 J/min (ACME: \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.032). \u003cb\u003eAdditional File Figure S5\u003c/b\u003e showed that in terms of ICU mortality in ARDS patients, the arterial PH mediated 24.6% (95% CI 0.115- 0.500; \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001) of the effect of MP\u0026thinsp;\u0026gt;\u0026thinsp;18.7 J/min (ACME:\u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.030).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eCluster Analysis\u003c/h2\u003e \u003cp\u003eIn this study, we identified three major sample clusters through consensus clustering analysis. The cumulative distribution function (CDF) graph showed the distribution of CDF across different numbers of clusters, with the Δarea graph revealing the largest changes in the area under the CDF curve occurring between k\u0026thinsp;=\u0026thinsp;2 and k\u0026thinsp;=\u0026thinsp;4 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The heat map of the consensus matrix indicated clear boundaries at k\u0026thinsp;=\u0026thinsp;2 and k\u0026thinsp;=\u0026thinsp;3 (\u003cb\u003eAdditional file Figure S6\u003c/b\u003e). The stability evaluation of the clustering results showed that the average cluster consistency scores were similar and both higher than 0.8 when k\u0026thinsp;=\u0026thinsp;2 and k\u0026thinsp;=\u0026thinsp;3 (\u003cb\u003eAdditional file Figure S7\u003c/b\u003e), while the PAC value was the lowest at k\u0026thinsp;=\u0026thinsp;3 (PAC\u0026thinsp;=\u0026thinsp;0.051) (\u003cb\u003eAdditional file Figure S8\u003c/b\u003e, leading to the determination of three as the optimal number of clusters.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn this study, three distinct clinical phenotypes were identified in patients with ARDS using consensus clustering analysis. The phenotypes were visualized using line and t-SNE plots (\u003cb\u003eAdditional file Figure S9 and\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The characteristics of the three phenotypes are presented in \u003cb\u003eAdditional file Table S2\u003c/b\u003e. Phenotype-I (n\u0026thinsp;=\u0026thinsp;483) exhibited the highest 28-day mortality rate at 29%. Patients in this phenotype had lower tidal volumes (mean: 404.69 ml), higher total respiratory rates (median: 23.00 bpm), higher plateau pressure (median: 22.10 cmH2O), higher peak pressure (median: 25.60 cmH2O), and the worst lung compliance (median: 29.40 ml/cmH2O). Phenotype-II (n\u0026thinsp;=\u0026thinsp;434) had the lowest duration of mechanical ventilation (median: 52.20 hours) and a moderate 28-day mortality rate of 20.50%. Patients in this group exhibited the best P/F ratio (median: 278.30 mmHg) and the lowest Sequential Organ Failure Assessment (SOFA) score (median: 6.6). Phenotype-III (n\u0026thinsp;=\u0026thinsp;416) had higher tidal volumes (mean: 530.15 ml), the highest BMI (median: 29.80 kg/m\u003csup\u003e2\u003c/sup\u003e) and the best lung compliance (median: 41.60 ml/cmH\u003csub\u003e2\u003c/sub\u003eO). They also exhibited the most laboratory abnormalities, including higher white blood cell counts (median: 11.10 K/mcL), higher creatinine levels (median: 1.20 mg/dl), and higher lactic acid levels (median: 2.00 mmol/L). This phenotype had a moderate duration of mechanical ventilation (median: 63.8 hours).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eThe relationship between phenotypes and outcomes.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe three phenotypes exhibited distinct clinical outcomes, with significant differences observed in mortality, mechanical ventilation time, and ICU stay lengths (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u003cb\u003eAdditional file Table S2\u003c/b\u003e). Patients in Phenotype II had shorter durations of mechanical ventilation compared to those in Phenotypes I and III. Those assigned to Phenotype I had the longest ICU stays, with a median duration of 9.13 days. Additionally, Phenotype I was associated with the highest rates of ICU mortality, in-hospital mortality, and 28-day mortality.\u003c/p\u003e \u003cp\u003e \u003cb\u003eHeterogeneity of mechanical power effects.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThe relationship between mechanical power and prognosis was investigated across the three phenotypes. Logistic regression analysis revealed that in Phenotype III, an increase in mechanical power was significantly associated with higher 28-day mortality, in-hospital mortality, and ICU mortalitym of ARDS patients, with odds ratios (OR) and 95% confidence intervals (CI) being 1.050 (1.018, 1.083), 1.053 (1.021, 1.085), and 1.057 (1.023, 1.091) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). No significant association was found between MP and mortality in Phenotype II patients. In patients with Phenotype I, an increase in MP was significantly associated with higher in-hospital mortality and ICU mortality, with OR and 95% CI being 1.028 (1.001, 1.056) and 1.031 (1.0033, 1.060) (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05)(Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe relationship between mechanical power and prognosis in difference phenotypes.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eOdds\u0026nbsp;Ratio\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLower\u0026nbsp;Bound\u0026nbsp;95%\u0026nbsp;CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eUpper\u0026nbsp;Bound\u0026nbsp;95%\u0026nbsp;CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003e28d\u0026nbsp;mortality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-I\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.027\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.000\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.050\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.019\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.981\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.058\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.334\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.050\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.018\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.083\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eHospital\u0026nbsp;mortality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-I\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.028\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.009\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.972\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.047\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.635\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.053\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.085\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eICU\u0026nbsp;mortality\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-I\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.031\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.060\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.030\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.034\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.994\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.076\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.098\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhenotype-III\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.057\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.091\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eEach phenotype was further divided into high and low MP subgroups, using 18.7 J/min as the threshold. The impact of MP varied across different phenotypes. In Phenotype I, compared to the low MP subgroup, patients with high MP had increased mechanical ventilation time, length of ICU stay, and ICU mortality. In Phenotype II, MP had no significant effect on mortality. In Phenotype III, MP less than 18.7 J/min was associated with reduced mechanical ventilation time, length of ICU stay, and mortality (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity analyses\u003c/h2\u003e \u003cp\u003eWe evaluated the impact of unmeasured confounding by sensitivity analysis parameter, which represents the correlation between the residuals of the mediator and outcome regressions. When ACME is 0, the value of ρ is between \u0026minus;\u0026thinsp;0.01 and \u0026minus;\u0026thinsp;0.02 and the 95% confidence interval of ρ includes 0. When ρ is greater than \u0026minus;\u0026thinsp;0.01 or less than \u0026minus;\u0026thinsp;0.02, the values of the average causal mediation effect being positive or negative remain unchanged. Therefore, the mediation effect in this study is robust (\u003cb\u003eAdditional file Figure S10-12\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study conducted a retrospective analysis of clinical data from the MIMIC-IV database of ARDS patients admitted to BIDMC from 2008 to 2019, with the objective of investigating the impact of mechanical power on the outcome of ARDS patients and the response of different phenotypes of ARDS patients to mechanical power. The findings of our study can be summary as follows: Firstly, there is a significant association between mechanical power in the first 24h and increased ICU, in-hospital, and 28-day mortality among ARDS patients. Secondly, causal mediation analysis revealed that arterial pH and PF ratio significantly mediated the relationship between mechanical power (MP) and mortality in ARDS patients, accounting for a substantial portion of effects on 28-day, hospital, and ICU mortality. Thirdly, our study identified three clinical phenotypes in ARDS patients using consensus clustering, revealing significant differences in mortality, mechanical ventilation duration, and ICU stays, and found varying associations between mechanical power and outcomes across three phenotypes.\u003c/p\u003e \u003cp\u003eMechanical power (MP) provides a more comprehensive way of evaluating ventilation load than traditional lung-protective ventilation strategies, and consists of three components: the power to overcome airway resistance during gas movement, the power to inflate the lung and chest wall movement, and the power to overcome lung and respiratory system end-expiratory pressure-related recoil[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. This makes MP not only more effective in reducing lung injury by precisely controlling energy transfer during ventilation, but also able to provide individualized treatment according to the patient's specific conditions, which may improve clinical outcomes, such as reducing mortality and shortening the length of ICU stay. Von Wedel et al[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] conducted a study on 2,103 surgical patients admitted in ICU and found that high mechanical power were associated with increased 28-day mortality.Our findings demonstrate a clear correlation between higher MP values and increased mortality rates within the first 24 hours of ventilation. The identification of an optimal MP cutoff value at 18.7 J/min.\u003c/p\u003e \u003cp\u003eOur causal mediation analysis provides a more detailed understanding of how mechanical power (MP) influences mortality, with arterial pH and PF ratio serving as significant mediators. Excessive mechanical power can lead to alveolar overdistension and lung injury, worsening gas exchange capacity and reducing the oxygenation index [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. High MP can lead to lung damage and affect the ventilation function, resulting in the obstruction of carbon dioxide discharge and respiratory acidosis, which may reduce the pH of the arterial blood. Additionally, inflammation induced by lung injury may indirectly affect pH through metabolic abnormalities, emphasizing the importance of optimizing mechanical power settings based on specific patient conditions in clinical practice to improve gas exchange and maintain acid-base balance.\u003c/p\u003e \u003cp\u003eIn this study, we identification three distinct ARDS clinical phenotypes through consensus clustering analysis emphasizes the heterogeneity within ARDS patient populations and underscores the necessity of personalized mechanical ventilation strategies. Phenotype I, characterized by lower tidal volumes, higher total respiratory rates, and poorer lung compliance, was associated with the highest rates of mortality and ICU stays, highlighting the potential benefits of tailored mechanical power settings in this group to mitigate adverse outcomes. Conversely, Phenotype III, which exhibited higher tidal volumes and better lung compliance, demonstrated a positive response to lower mechanical power settings, underscoring the phenotype's potential resilience to ARDS's detrimental effects. These findings suggest that mechanical power, a modifiable factor during mechanical ventilation, plays a critical role in the prognosis of ARDS patients, with its impact varying significantly across different phenotypes.\u003c/p\u003e \u003cp\u003eThe differential effects of mechanical power across phenotypes indicate a nuanced relationship between mechanical ventilation settings and patient outcomes, advocating for a more individualized approach to ventilation management in ARDS. This approach could involve adjusting mechanical power to optimize patient-specific parameters such as lung compliance and tidal volume, potentially reducing mortality and improving overall outcomes. Furthermore, the mediation analysis revealing that arterial PH and PF ratio significantly mediate the effect of mechanical power on mortality highlights the importance of monitoring and managing these parameters in ARDS treatment protocols.\u003c/p\u003e \u003cp\u003eIn summary, this study's findings illuminate the critical need for personalized mechanical ventilation strategies in ARDS, informed by patient-specific phenotypes and key physiological parameters. By adopting such an approach, clinicians can potentially enhance the efficacy of ARDS management, reducing mortality and improving patient outcomes in this heterogeneously affected patient population.\u003c/p\u003e \u003cp\u003eThere are some limitations in this study. This study was a retrospective analysis based on the MIMIC-IV database, and the results may be limited by the quality and completeness of the recorded data. And external validation is further needed to establish the reliability and universality of the identified phenotypes and MP cut-offs in different populations and clinical settings.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, our study revealed the critical role of mechanical power in ARDS outcomes, advocating for a appreciation of patient heterogeneity in the management of ARDS. Through a deeper understanding of the phenotypic differences and their interactions with mechanical ventilation parameters, clinicians can better navigate the challenges of providing optimal care for ARDS patients.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors extend their gratitude to the researchers who established and maintained the MIMIC-IV database. The opinions expressed in this study are solely those of the authors and do not represent the views of any affiliated third parties.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFMX contributed to obtaining funding and study supervision. ZQ and LN contributed to study selection, statistical analysis, and data extraction. ZQ contributed to the charts making and drafting of the article. WF and WHY contributed to the thorough review of statistical methodologies and analyses. DRS, LY, WZY, and LY contributed to the revision of the article for important intellectual content. The article writing and figure elaboration were completed by all the authors together. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the grants of the program of Hebei Province to introduce overseas students (grant number C20210353) and \u0026quot;333 Talent Project\u0026quot; talent training funding project (grant number A202103005).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data and material were available at https:// mimic. mit. edu/.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declarations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe patient privacy information in the MIMIC-IV database has been anonymized, this study does not require ethical approval.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eMeyer NJ, Gattinoni L, Calfee CS. Acute respiratory distress syndrome. Lancet. 398, 622\u0026ndash;637; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S0140-6736(21)00439-6\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(21)00439-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMatthay MA, et al. A New Global Definition of Acute Respiratory Distress Syndrome. Am J Respir Crit Care Med. 209, 37\u0026ndash;47; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1164/rccm.202303-0558WS\u003c/span\u003e\u003cspan address=\"10.1164/rccm.202303-0558WS\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBellani G, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 315, 788\u0026ndash;800; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2016.0291\u003c/span\u003e\u003cspan address=\"10.1001/jama.2016.0291\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBos LDJ, Ware LB. Acute respiratory distress syndrome: causes, pathophysiology, and phenotypes. Lancet. 400, 1145\u0026ndash;1156; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/S0140-6736(22)01485-4\u003c/span\u003e\u003cspan address=\"10.1016/S0140-6736(22)01485-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaudel R, et al. Mechanical Power: A New Concept in Mechanical Ventilation. Am J Med Sci. 362, 537\u0026ndash;545; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.amjms.2021.09.004\u003c/span\u003e\u003cspan address=\"10.1016/j.amjms.2021.09.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSerpa Neto A, et al. Mechanical power of ventilation is associated with mortality in critically ill patients: an analysis of patients in two observational cohorts. \u003cem\u003eIntensive Care Med\u003c/em\u003e. 44, 1914\u0026ndash;1922; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00134-018-5375-6\u003c/span\u003e\u003cspan address=\"10.1007/s00134-018-5375-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Z, Zheng B, Liu N, Ge H, Hong Y. Mechanical power normalized to predicted body weight as a predictor of mortality in patients with acute respiratory distress syndrome. Intensive Care Med. 45, 856\u0026ndash;864; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00134-019-05627-9\u003c/span\u003e\u003cspan address=\"10.1007/s00134-019-05627-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRanieri VM, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 307, 2526\u0026ndash;33; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jama.2012.5669\u003c/span\u003e\u003cspan address=\"10.1001/jama.2012.5669\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJimenez JV, et al. Electric impedance tomography-guided PEEP titration reduces mechanical power in ARDS: a randomized crossover pilot trial. \u003cem\u003eCrit Care\u003c/em\u003e. 27, 21; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13054-023-04315-x\u003c/span\u003e\u003cspan address=\"10.1186/s13054-023-04315-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrotberg JC, Reynolds D, Kraft BD. Management of severe acute respiratory distress syndrome: a primer. Crit Care. 27, 289; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13054-023-04572-w\u003c/span\u003e\u003cspan address=\"10.1186/s13054-023-04572-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Z, Zheng C, Kim C, Van Poucke S, Lin S, Lan P. Causal mediation analysis in the context of clinical research. Ann Transl Med. 4, 425; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.21037/atm.2016.11.11\u003c/span\u003e\u003cspan address=\"10.21037/atm.2016.11.11\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePattharanitima P, et al. Machine Learning Consensus Clustering Approach for Patients with Lactic Acidosis in Intensive Care Units. J Pers Med. 11, 1132; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/jpm11111132\u003c/span\u003e\u003cspan address=\"10.3390/jpm11111132\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu X, et al. Identification of distinct clinical phenotypes of acute respiratory distress syndrome with differential responses to treatment. Crit Care. 25, 320; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13054-021-03734-y\u003c/span\u003e\u003cspan address=\"10.1186/s13054-021-03734-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoppe K, Khan E, Meybohm P, Riese T. Mechanical power of ventilation and driving pressure: two undervalued parameters for pre extracorporeal membrane oxygenation ventilation and during daily management? Crit Care. 27, 111; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13054-023-04375-z\u003c/span\u003e\u003cspan address=\"10.1186/s13054-023-04375-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evon Wedel D, et al. Adjustments of Ventilator Parameters during Operating Room-to-ICU Transition and 28-Day Mortality. Am J Respir Crit Care Med. 209, 553\u0026ndash;562; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1164/rccm\u003c/span\u003e\u003cspan address=\"10.1164/rccm\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. 202307-1168OC (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu XB, Zhu F. Reinterpreting the ventilator-induced lung injury from the prospective of mechanical power. Zhonghua Shao Shang Za Zhi. 37, 292\u0026ndash;295; 10.3760/ cma.j.cn501120-20200203-00039 (2021).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Acute Respiratory Distress Syndrome, Mechanical ventilation, Mechanical power, mortality, consensus clustering analysis","lastPublishedDoi":"10.21203/rs.3.rs-4441850/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4441850/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn this study, we investigated the effect of mechanical power (MP) on mortality in acute respiratory distress syndrome (ARDS) patients. Patients diagnosed with ARDS were identified from the MIMIC-IV database. Kaplan-Meier curves and Cox proportional hazards models were utilized for survival analysis. The optimal cut-off value for MP was determined by using 'survminer' package. Causal mediation analysis (CMA) further investigated the effect of MP on 28-day mortality. Key predictive indicators were used to cluster and identify characteristics of different phenotypes. A total of 1333 patients were included. MP lower than 18.7J/min was associated with reduced mortality. Arterial pH and P/F ratio separately accounted for 29.2% and 20% of the mediating effect of high MP on increased 28-day mortality. Clustering analysis showed that phenotype-I had the worst respiratory mechanical parameters and the highest 28-day mortality. Phenotype-II was correlated with less organ dysfunction, the best oxygenation index and lower mechanical ventilation hours. Phenotype-III had the most laboratory abnormalities, the worse P/F ratio and longer ICU staytime. MP is strongly associated with mortality of ARDS patients belong to phenotype-III. High MP is independently associated with increased mortality in patients with ARDS. MP of less than 18.7 J/min is safer for ARDS patients.\u003c/p\u003e","manuscriptTitle":"Exploring the Impact of Mechanical Power on Mortality and Phenotypes in ARDS Patients: A Retrospective Analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-07 23:13:46","doi":"10.21203/rs.3.rs-4441850/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"40d451d1-8e50-40c9-a481-01bdfb6fc52c","owner":[],"postedDate":"June 7th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":32793700,"name":"Health sciences/Medical research"},{"id":32793701,"name":"Health sciences/Risk factors"}],"tags":[],"updatedAt":"2025-05-13T05:23:49+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-07 23:13:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4441850","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4441850","identity":"rs-4441850","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}