Abstract
Background and Aims:
Mechanical ventilation is a cornerstone in managing acute respiratory failure, yet, it can lead to ventilator-induced lung injury. This study investigated the relationship between mechanical power (MP) and 28-day mortality in critically ill patients undergoing mechanical ventilation.
Methods:
A single-centre, prospective observational study was conducted, involving 102 mechanically ventilated patients in a tertiary care intensive care unit (ICU). Patient characteristics, ventilator parameters, and clinical scores {Acute Physiology and Chronic Health Evaluation (APACHE) II and Sequential Organ Failure Assessment (SOFA)} were recorded. MP was calculated at 24, 48, and 72 hours, and its correlation with mortality and other clinical outcomes was analysed using logistic regression.
Results:
The cohort had a median age of 40.5 years, with a mortality rate of 37.3% within 28 days. Higher baseline MP values (23.90 J/min vs. 19.0 J/min, P = 0.008) were observed in non-survivors. MP demonstrated strong correlations with driving pressure and plateau pressure (r = 0.79 to 0.86, P < 0.001). MP was measured at 24, 48, and 72 hours after ICU admission and demonstrated increasing predictive validity for 28-day mortality (Hazard ratio of 1.05, 1.09, and 1.11, respectively). The receiver operating characteristic curve supported this trend, with area under the curve values rising from 0.633 at 24 hours to 0.708 at 48 hours and 0.747 at 72 hours.
Conclusion:
This study highlights MP as a critical factor in assessing mortality risk in mechanically ventilated patients. Elevated MP correlates with increased mortality. Monitoring MP in addition to already established parameters can predict the outcome of patients on mechanical ventilation. Future research should explore MP and its implications for ventilation practices in diverse patient populations.
Keywords: Acute respiratory failure, correlation, critically ill, mechanical power, mechanical ventilation, mortality, outcome, ventilation-induced lung injury
INTRODUCTION
Critically ill patients with acute respiratory failure usually need invasive mechanical ventilation, making them prone to ventilator/ventilation-induced lung injury (VILI).[1] VILI results from the interaction between the ventilation delivered to the lung parenchyma and the acceptance by the lung parenchyma. The ventilator-generated causes of VILI include pressures, volume, flow, and respiratory rate.[2,3] In parallel, lung-related factors such as reduced functional lung size,[4] parenchymal inhomogeneity with stress amplification,[5] and cyclic alveolar collapse and reopening[6] increase vulnerability to VILI and the resulting pathophysiology encompasses barotrauma, volutrauma, atelectotrauma, and biotrauma. The distinction between the ventilator’s and the lung’s contribution to VILI may help to reach a better VILI-prevention strategy. Several studies have shown that strategies such as lower tidal volume (VT),[7,8] lower driving pressure (DP),[9] lower plateau pressure (Pplt),[10] and higher positive end- expiratory pressure (PEEP)[11] can prevent VILI. They have demonstrated the survival benefit of lung protective ventilation strategies and suggested decreasing stress and strain injury to inflamed and fragile lung tissue.
From a physical and thermodynamic perspective, lung injury can be conceptualised as a consequence of excessive energy transfer from the ventilator to the lung. Mechanical power (MP) represents the total energy delivered to the respiratory system per unit time and integrates all ventilator-related determinants of VILI, including VT, airway pressures, inspiratory flow, and respiratory rate. The dissipation of this energy within lung tissue can result in abnormal cellular deformation, extracellular matrix disruption, inflammatory activation, and microstructural injury. Thus, while ventilator-related causes of VILI have traditionally been studied separately, they can be unified within the single construct of MP. Although VT and airway pressures have received considerable attention, respiratory rate and inspiratory flow remain underrecognised contributors to lung injury, despite strong experimental and theoretical evidence supporting their role in increasing MP.[3]
To integrate these variables into a single quantitative framework, Gattinoni et al.[12] introduced the concept of MP based on the equation of motion for volume-controlled ventilation. Subsequent simplified and mode-specific equations have facilitated bedside estimation of MP in both volume- and pressure-controlled ventilation (PCV).[13,14,15,16]
Experimental studies have consistently demonstrated a direct association between increasing MP and the development of lung injury.[3,17,18,19] Clinically, Amato et al.[7] first identified DP as a key predictor of mortality in acute respiratory distress syndrome (ARDS), a finding reinforced by subsequent analyses showing its superiority over VT alone.[9] Guérin et al.[20] further demonstrated that MP above a threshold of approximately 12 J/min was associated with reduced survival. Large observational studies have since confirmed that elevated MP during early mechanical ventilation is independently associated with increased mortality and fewer ventilator-free days (VFD) in both patients with and without ARDS.[21,22]
However, identical levels of MP may exert different biological effects depending on ventilated lung size (“baby lung”) and the relative contributions of lung and chest wall elastance.[23,24,25] Normalisation of MP to predicted body weight has therefore been proposed to improve its prognostic value.[21] Additionally, recent evidence suggests that dynamic changes in MP over time may be more informative than single time-point measurements.[26] Despite its limitations—including assumptions of lung homogeneity and lack of consideration for patient effort or regional stress distribution—MP provides a physiologically sound, integrative framework for understanding VILI and may serve as an important target for individualised, lung-protective ventilation strategies.
The primary objective of this study was to explore the correlation between MP and 28-day mortality outcomes. The secondary outcome of this study was to find out the correlation of DP and Pplt with MP, the number of VFD, length of intensive care unit (ICU) stay, and hospital stay and to evaluate the impact of MP variation rate over 72 hours on mortality.
METHODS
It is a single-centre, prospective observational study conducted in the ICU of a tertiary care hospital from June 2023 to July 2024. This study was approved by the institutional ethical committee (vide approval number 2022-193-DM-129, dated 06 February 2023) and registered in the Clinical Trials Registry-India (CTRI/2023/05/052466). Informed consent was taken from the patient’s next of kin. Out of 330 patients screened, 102 patients were enroled, who required mechanical ventilation for more than 48 h. Inclusion criteria were age more than 18 years and those on invasive controlled mechanical ventilation for more than 48 hours. This study was done according to the principles of the Declaration of Helsinki, 2013 and Good Clinical Practice guidelines.
Patients’ clinical characteristics and 6-hourly ventilator parameters were recorded. Severity illness scores like acute physiology and chronic health evaluation (APACHE) II and sequential organ failure assessment (SOFA) were recorded at enrolment, and MP was calculated as per the formula given by Chiumello et al.[16] [MP = 0.098 x RR x Vt x {positive end expiratory pressure (PEEP)} + ΔPinsp], where 0.098 is a conversion factor from cmH2O l min − 1 in J/min, RR is the respiratory rate, Vt is the tidal volume in litres. PEEP, and ΔPinsp is the pressure (cmH2O) above PEEP] for PCV and Giosa et al.[13] [MP = [VE x Peak pressure + PEEP + Inspiratory flow/6J/20, where VE is the minute ventilation expressed in l/min. Peak pressure and PEEP are expressed in cmH2O] for volume-controlled ventilation (VCV) every 6 hours and averaged at 24 h, 48 h, and 72 h. The MP variation rate (%) was calculated as [Baseline MP- 24 hour interval MP/Baseline MP] x 100.
Patients were divided into survivors and non-survivors on day 28 and analysed. Use of the standard protocol of mechanical ventilation, time point-based detailed data collection, and multivariate regression analysis was used to manage potential confounding variables.
Based on the study conducted by Wu HP et al.,[27] the diagnostic accuracy of the day 1 MP score to detect the ICU patient mortality was 73% [area under the receiver operating characteristic (AUROC) curve]. At minimum two-sided 95% confidence interval and 95% power of the study, the estimated sample size for study was 74 (assuming a survivor and non-survivor patient ratio of 1:1). Finally, 102 patients were targeted in our study. Descriptive statistics of the non-uniformly distributed data such as age, severity score, and ventilatory parameters are expressed as median (interquartile range), while categorical data such as sex and severity of ARDS are represented as a percentage (%). Chi-square test was used to find the association between qualitative variables such as gender, ARDS severity, and indication of mechanical ventilation. Mann–Whitney U test was used to compare the non-uniformly distributed data. The receiver operating characteristic (ROC) curve analysis and Kaplan survival analysis methods were used to assess the predictors of ICU mortality after including MP score along with other independent variables. A P value < 0.05 was considered statistically significant. Statistical package for social sciences (SPSS) (version-23) (International Business Machines Corp. (2015)) was used for data analysis.
RESULTS
A total of 330 patients were admitted in ICU during the study period (January 2023 to June 2024), who required mechanical ventilation for acute respiratory failure. Among them, 228 patients were excluded (age < 18 years n = 45; on weaning mode n = 58; on ventilation before admission n = 49; severe haemodynamic instability n = 46; died within 48 hours n = 26; and extubated in less than 48 hours n = 4) [Figure 1]. Hence, 102 patients were included in this study and analysed. The median age was 40.5 years, and 59.8% were male. The most common indication for the requirement of mechanical ventilation was hypoxic respiratory failure [n = 75 (73.5%)] and encephalopathy [n = 23 (22.5%)]. The mean APACHE II and SOFA scores were 17 and 10, respectively, showing high severity of illness. In our study, the majority of patients (96 out of 102) were ventilated using PCV due to prevailing clinical practices in the participating ICU. A total of 38 patients died and 64 survived within 28 days of admission; however, a total of 53 patients died during the course of ICU. Table 1 demonstrates baseline clinical characteristics between survivor and non-survivor groups.
Figure 1.
A flow chart of patient inclusion and exclusion. n=number of patients
Table 1.
Baseline characteristics
| Variable | All (n=102) | Survivors (n=64) | Non-survivors (n=38) | P | |||
|---|---|---|---|---|---|---|---|
| Age | 40.50 (26.8-54) | 39 (25-54.75) | 42 (34.5-52.5) | 0.41 | |||
| Sex: Male | 61 (59.8) | 39 | 22 | 0.46 | |||
| CCI | 2.00 (0.00-3.3) | 0.50 (0.00-3) | 2 (1-4) | 0.11 | |||
| APACHE at admission | 17.00 (15-22) | 16 (13.25-20) | 21 (16.75-24) | 0.001 | |||
| SOFA at admission | 10 (7.75-12) | 9 (6.25-11) | 11.50 (9.75-13) | 0.001 | |||
| PF ratio | 229 (180.7-280) | 242.50 (190-280) | 214 (160-264.75) | 0.04 | |||
| PF ratio <150 | |||||||
| No Yes |
86 (84.30) 16 (15.70) |
56 8 |
30 8 |
0.27 | |||
| ARDS | |||||||
| Yes No |
87 (84.30) 15 (14.70) |
55 9 |
32 6 |
0.51 | |||
| ARDS Severity: | |||||||
| No Mild Moderate Severe |
15 (14.70) 50 (49) 28 (27.50) 9 (8.80) |
9 36 16 3 |
6 14 12 6 |
0.13 | |||
| Primary organ failure: | |||||||
| Abdomen Lung Liver |
35 (34.30) 22 (21.60) 21 (20.60) |
23 (35.90) 17 (26.60) 9 (14.10) |
12 (31.60) 5 (13.20) 12 (31.60) |
0.36 | |||
| Indication for MV | |||||||
| Respiratory distress Encephalopathy Haemodynamic instability Elective |
75 (73.50) 23 (22.50) 2 (2) 2 (2) |
50 11 1 2 |
25 12 1 0 |
0.26 | |||
| APACHE_24 h | 15 (12-18) | 13 (10-16.75) | 19.5 (15-22.5%) | 0.001 | |||
| SOFA_24 h | 9 (7-12) | 8 (6-11) | 12 (9.75-14) | 0.001 | |||
| SOFA_48 h | 8 (6-12) | 7 (5-9) | 12 (8-13.25) | 0.001 | |||
| SOFA_72 h | 7 (5-10) | 5 (4-8) | 10.5 (7-14) | 0.001 | |||
| Vasopressor Need: | |||||||
| No Single Dual |
20 (19.60) 52 (51) 30 (29.40) |
18 32 14 |
2 20 16 |
0.008 | |||
| Prone | |||||||
| Yes No |
9 (8.80) 93 (91.20) |
4 60 |
5 33 |
0.28 | |||
| Tracheostomy | |||||||
| Yes No |
48 (47.10) 54 (52.90) |
40 24 |
8 30 |
0.001 | |||
| LOS ICU | 27 (15-47) | 40 (26.25-56.25) | 14 (11.75-23.25) | 0.001 | |||
| LOS Hospital | 33.5 (21-55.50) | 52 (32.25-66.75) | 17 (12-27) | 0.001 | |||
| Overall Mortality | 53 (52) | ||||||
Data are presented as median (IQR) or number of patients (percentage). IQR=interquartile range; APACHE=Acute physiology and chronic health evaluation; SOFA=Sequential organ failure assessment; CCI=Charlson comorbidity index; LOS=length of stay; PF ratio=PaO2/FiO2 ratio; ARDS=acute respiratory distress syndrome; MV=mechanical ventilation; ICU=intensive care unit
Baseline MP in the whole population at 24 hours was 20.45 J/min and was significantly higher in non-survivor groups than in survivor groups (19.0 J/min vs. 23.90 J/min; P = 0.008). The 24-h MP variation rate was 6.5% (IQR -7.28% to 16.02%) and -1.43% (IQR -9.92% to 9.1%) in survivor and non-survivor groups (P = 0.05), respectively, resulting in higher MP at 48 h in non-survivors (23.55; IQR- 18.27 to 29.3) group as compared to survivors (18.64; IQR-14.62–22.33) [Table 2]. In all patients, MP showed significant correlation with DP and Pplt at 24 h (DP, r = 0.79; Pplt, r = 0.85), 48 h (DP, r = 0.78; Pplt, r = 0.86), and 72 h (DP, r = 0.79; Pplt, r = 0.86) (all P = 0.001; Figure 2), respectively. Supplementary Table 1 shows the results of the logistic regression analysis. Admission APACHE II and SOFA score, arterial oxygen partial pressure/fraction of inspired oxygen (P/F) ratio at the time of admission and SOFA score at 24-hour interval up to 72 hours, and MP at 24 hours, 48 hours, and 72 hours were significantly associated with 28-day mortality on univariate analysis. Amongst them, after adjustment with illness severity at admission, MP at 24 hours {hazard ratio(HR)-1.05, 95%confidence interval (CI)-1.001–1.103; P = 0.05}, 48 hours (HR-1.09, 95% CI-1.04–1.15; P = 0.001), and 72 hours (HR-1.11, 95% CI-1.06–1.17; P = 0.001) was significantly associated with 28-day mortality. The ROC curve analysis of MP showed the areas under the curve at 24 h, 48 h, and 72 h were 0.633 (95% CI: 0.52, 0.75), 0.708 (95% CI: 0.60, 0.81), and 0.747 (95% CI: 0.65, 0.84), respectively, and all predictive values were significant [Figure 3]. The patients were divided into high mechanical power (HMP) and low MP (LMP) groups according to whether MP exceeded 19 Jules/min at 24 h and 48 h. As 38 cases died in 28 days, analysing the Kaplan–Meier survival curve showed that the overall survival rate was 62.7%. The HMP group had a lower survival rate (56.4% and 50.9%) than the LMP (low MP) group (70.25% and 77.8%) at 24 hours (Log rank P value = 0.27) and 48 hours, respectively (Log rank P value = 0.008; statistically significant) [Figure 4].
Table 2.
Comparison of ventilatory variables between survivors and non-survivors
| Variables | All n=102 | Survivors (n=64) | Non-survivors (n=38) | P |
|---|---|---|---|---|
| Mode of ventilation | ||||
| PC VC |
96 (94.10) 6 (5.90) |
62 2 |
34 4 |
0.19 |
| RR_24h (/min) | 20 (17.50-22) | 18.38 (17.06-21) | 21 (18-24.44) | 0.001 |
| RR_48h (/min) | 19.50 (17.88-22) | 19 (16.68-20.44) | 21 (18-24) | 0.001 |
| RR_72h (/min) | 19.50 (17-22) | 18 (16-20) | 20.75 (19.88-24) | 0.001 |
| Vt/Pbw_24h (ml/kg) | 6.20 (5.75-6.74) | 6.32 (5.91-7.13) | 6.03 (5.65-6.60) | 0.006 |
| Vt/Pbw_48h (ml/kg) | 6.10 (5.73-6.67) | 6.20 (5.77-6.99) | 5.99 (5.69-6.53) | 0.09 |
| Vt/Pbw_72h (ml/kg) | 6 (5.65-6.77) | 6.16 (5.77-7.06) | 5.95 (5.54-6.48) | 0.05 |
| PEEP_24h (cm H2O) | 8 (7-9.56) | 8 (7-9) | 8 (7.69-10.56) | 0.05 |
| PEEP_48h (cm H2O) | 8 (7-9.31) | 7.75 (7-8.25) | 8.38 (8-10.88) | 0.001 |
| PEEP_72h (cm H2O) | 8 (7-9) | 7.50 (7-8.25) | 8.63 (7.88-10) | 0.001 |
| Pplat_24h (cm H2O) | 24 (21.19-27.5) | 23 (20-26) | 26 (22-29.13) | 0.001 |
| Pplat_48h (cm H2O) | 23.38 (20.81-27) | 22 (19.06-25.44) | 26.5 (22.94-28) | 0.001 |
| Pplat_72h (cm H2O) | 24 (19.5-26.5) | 21.75 (18.25-25) | 25.63 (23.56-28.06) | 0.001 |
| DP_24h (cm H2O) | 15.38 (13.5-18.56) | 15 (12.56-17.44) | 17.25 (14.5-20.06) | 0.001 |
| DP_48h (cm H2O) | 15.12 (13.38-18) | 14.25 (12-16.50) | 16.50 (14.25-20) | 0.001 |
| DP_72h (cm H2O) | 15 (12-17.75) | 14 (12-16) | 17 (14-20.25) | 0.001 |
| MV_24h (l/min) | 8.7 (7.90-10.20) | 8.70 (7.85-9.8) | 8.85 (7.90-10.70) | 0.21 |
| MV_48h (l/min) | 8.6 (7.70-9.60) | 8.30 (7.50-9.10) | 9 (8.30-10.41) | 0.02 |
| MV_72h (l/min) | 8.30 (7.50-9.40) | 7.8 (7.28-9.05) | 8.87 (8-10.40) | 0.003 |
| MP_24h (J/min) | 20.45 (16.38-27.10) | 19 (16.14-25.42) | 23.98 (17.74-31.20) | 0.008 |
| MP_48h (J/min) | 19.99 (15.87-25.08) | 18.64 (14.62-22.33) | 23.55 (18.27-29.30) | 0.001 |
| MP_72h (J/min) | 19.14 (15-24.40) | 17.60 (13.86-21.92) | 23.08 (18.74-28.22) | 0.001 |
| MP_VR_24 | 3.60 (-8.20 to 15.80) | 6.5 (-7.28 to 16.02) | -1.43 (-9.92 to 9.10) | 0.05 |
| MP_VR_48 | 5.70 (-5.39 to 14.59) | 6.51 (-4.80 to 13.82) | 5.02 (-10.23 to 15.86) | 0.19 |
| VFD (days) | 5.30 (0.0-10.25) | 8.45 (0.0-19.5) | 0.000 (0.00-0.00) | 0.001 |
| Duration of MV (days) | 19 (10-28.50) | 23 (8-38) | 14 (10-22) | 0.001 |
Data are presented as median (IQR) or number of patients (percentage). IQR=interquartile range; PC=pressure controlled; VC=volume controlled; RR=respiratory rate; Vt/Pbw=tidal volume/predicated body weight; PEEP=positive end expiratory pressure; plat=plateau pressure; DP=driving pressure; MP=mechanical power; MP_VR=mechanical power variation rate; VFD=ventilator-free day; MV=mechanical ventilation
Figure 2.
Pearson correlation of MP with DP and Pplt at 24 hours, 48 hours, and 72 hours. MP = mechanical power; DP = driving pressure; Pplt = plateau pressure
Supplementary Table 1.
Logistic regression analysis of severity score, ventilatory pressure, and mortality outcome
| Variable | Univariate analysis |
Multi-variate analysis |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P | HR | 95% CI | P | |
| APACHE_adm | 1.10 | 1.04-1.17 | 0.002 | |||
| APACHE_24h | 1.19 | 1.12-1.26 | 0.001 | |||
| SOFA_adm | 1.17 | 1.07-1.29 | 0.001 | |||
| SOFA_24h | 1.26 | 1.15-1.39 | 0.001 | 1.26 | 1.15-1.39 | 0.001 |
| SOFA_48h | 1.32 | 1.21-1.45 | 0.001 | 1.46 | 1.28-1.66 | 0.001 |
| SOFA_72h | 1.36 | 1.25-1.49 | 0.001 | 1.36 | 1.24-1.48 | 0.001 |
| Pplat_24h | 1.11 | 1.04-1.18 | 0.002 | |||
| Pplat_48h | 1.17 | 1.09-1.25 | 0.001 | |||
| Pplat_72h | 1.17 | 1.09-1.25 | 0.001 | |||
| DP_24h | 1.15 | 1.05-1.26 | 0.002 | |||
| DP_48h | 1.21 | 1.1-1.33 | 0.001 | |||
| DP_72h | 1.19 | 1.09-1.31 | 0.001 | |||
| VFD | 0.78 | 0.63-0.96 | 0.02 | |||
| MP_24h | 1.06 | 1.01-1.11 | 0.01 | 1.05 | 1.001-1.103 | 0.04 |
| MP_48h | 1.10 | 1.05-1.15 | 0.001 | 1.09 | 1.04-1.15 | 0.001 |
| MP_72h | 1.11 | 1.06-1.16 | 0.001 | 1.11 | 1.056-1.165 | 0.001 |
Data are presented as hazard ratio. HR=hazard ratio; CI=confidence interval; APACHE=Acute physiology and chronic health evaluation; SOFA=Sequential organ failure assessment; plat=plateau pressure; DP=driving pressure; VFD: ventilator-free day; MP=mechanical power
Figure 3.
Receiver operating curves showing the prognostic accuracy of MP measured at 24, 48, and 72 hours in predicting patient outcomes. (AUC = 0.633; P = 0.02; 95% CI = 0.52, 0.75 with a sensitivity of 66% and a specificity of 51% at 24 hours AUC = 0.708; P = 0.001; 95% CI = 0.60, 0.81 with a sensitivity of 71% and a specificity of 57% at 48 hours, and 0.747; P = 0.000; 95% CI = 0.65, 0.84 with a sensitivity of 73% and a specificity of 67% at 72 hours). MP = mechanical power; AUC = area under the curve; ROC = receiver operating characteristic
Figure 4.
Kaplan–Meier survival curves illustrating 28-day cumulative ICU survival based on mechanical power (MP) stratification at (a) 24 hours and (b) 48 hours. This figure compares survival outcomes between patients receiving higher mechanical power (HMP, ≥19 J/min) and those with lower mechanical power (LMP, <19 J/min). Patients with LMP had better cumulative survival at both time points, indicating an association between LMP and improved short-term ICU outcomes. ICU=intensive care unit; HMP = higher mechanical power; LMP = lower mechanical power; LOS_ccm 28 dS = 28-day survival in ICU
DISCUSSION
Mechanical ventilation is one of the life-saving therapies in the management of acute respiratory failure in critically ill patients. In this prospective study of 102 mechanically ventilated patients with high severity of illness, we identified a significant correlation between MP and 28-day mortality. Higher MP particularly during the first 72 hours of ventilation is significantly associated with increased mortality, underscoring the clinical relevance of MP as an integrative marker of ventilator-induced stress.
The baseline MP at 24 hours was significantly higher in non-survivors compared to survivors (23.90 J/min vs. 19.0 J/min, P = 0.008). The observed elevation in MP in non-survivors is consistent with recent findings by Serpa et al.,[22] who reported that higher MP was linked to increased mortality in patients with ARDS. Similarly, Xie et al.[28] reported in a secondary analysis of the LOV-Ed study that survivors exhibited significantly lower MP levels, averaging less than 17 J/min. Parhar et al.[29] observed that an MP below 22 J/min was associated with more VFD and reduced mortality. Our study supports these findings, revealing that higher MP values were prevalent in non-survivors.
Additionally, the variation rate in MP over the first 72 hours revealed critical insights. While survivors showed a slight decrease in MP, non-survivors experienced an increase, suggesting that early stabilisation or reduction in MP might be beneficial for survival. Yi Chi et al.[26] also found that the MP-improved group had a lower ICU mortality rate than the MP-worsened group (24% vs. 36%; P = 0.005) and 24-h MP variation rate was associated with ICU mortality (odds ratio, 0.906 [95% CI 0.833–0.985]; P = 0.021), similar to our study.
We also observed a strong correlation between MP and established ventilatory parameters, particularly DP and Pplt, across all time points (24, 48, and 72 hours; r = 0.79–0.86, P = 0.001). This interdependence highlights the shared physiological basis of these variables and supports prior work by Roca et al.[30] and Tonna E et al.,[31] who demonstrated that elevated DP is strongly associated with worse outcomes. While DP remains a key marker of lung stress, MP integrates DP with tidal volume, respiratory rate, and flow, providing a more comprehensive assessment of the mechanical energy applied to the lungs.
The multivariate logistic regression analysis shows that MP at 24, 48, and 72 hours remained significantly associated with 28-day mortality even after adjusting for APACHE II and SOFA scores, which is also supported by the findings of Xie et al.[28] The hazard ratios in our study indicating increased risk (HR 1.05, 1.09, and 1.11, respectively) reinforce that higher MP is a critical risk factor for mortality in this cohort. The ROC curve analysis further supports the predictive validity of MP, with increasing areas under the curve (AUC) at 24, 48, and 72 hours (0.633, 0.708, and 0.747, respectively).
Stratification based on an MP threshold of 19 J/min revealed significantly lower survival in the HMP group than in LMP group (50.9% vs. 77.80%, Log rank P value < 0.008), consistent with findings reported by Wu et al.[27] In subgroup analysis, patients with increasing of severity of ARDS exhibited higher MP, DP, and Pplt and mortality compared with non-ARDS patients. These findings reinforce the vulnerability of the “baby lung” to excessive energy delivery [Supplementary Table 2].
Supplementary Table 2.
Characteristics in between Severity of ARDS and Non-ARDS patients
| Variable | Non-ARDS (n=15) | Mild ARDS (n=50) | Moderate ARDS (n=28) | Severe ARDS (n=9) | P |
|---|---|---|---|---|---|
| Age in years, Mean (SD) | 40.7 (17.17) | 42.2 (16.32) | 42.88 (15.97) | 39.11 (14.3) | 0.92 |
| Sex: Male n, (%) | 9 (60) | 30 (60) | 17 (60.7) | 5 (55.6) | 0.99 |
| PBW in kg, Mean (SD) | 71.67 (7.17) | 69.6 (9.76) | 72.39 (7.52) | 73.77 (8.34) | 0.50 |
| CCI, Median (IQR) | 2 (0-4) | 2 (0-3) | 1 (0-4) | 1.5 (0-2.5) | 0.68 |
| APACHE at adm, Mean (SD) | 17.53 (3.76) | 17.66 (5.46) | 18.39 (4.79) | 21.33 (8.46) | 0.29 |
| SOFA at adm, Mean (SD) | 9.13 (2.82) | 10.04 (3.25) | 9.5 (3.23) | 11.55 (3.94) | 0.30 |
| PF ratio, Mean (SD) | 318 (17.80) | 252.9 (27.62) | 170.3 (22.8) | 89.55 (7.78) | 0.001 |
| APACHE_24h, Mean (SD) | 14.4 (3.77) | 15 (5.66) | 15.89 (5.96) | 18.44 (7.73) | 0.34 |
| SOFA_24h, Mean (SD) | 9.07 (2.79) | 9.7 (3.69) | 9.1 (3.86) | 11.33 (3.53) | 0.43 |
| SOFA_48h Mean (SD) | 8.2 (3.19) | 8.44 (4.02) | 8.39 (3.49) | 10.77 (3.86) | 0.34 |
| SOFA_72h Mean (SD) | 7.53 (3.41) | 7.32 (4.30) | 7.28 (3.82) | 9.77 (4.43) | 0.40 |
| Pplt at 24 h Mean (SD) | 20.38 (2.20) | 23.16 (4.32) | 26.28 (3.61) | 30.97 (4.33) | 0.001 |
| DP at 24h Mean (SD) | 13.7 (2.08) | 15.30 (3.59) | 17.27 (3.18) | 19.47 (4.26) | 0.001 |
| VT_IBW Mean (SD) (ml) | 5.9 (0.74) | 6.72 (1.06) | 6.26 (0.90) | 5.90 (0.46) | 0.008 |
| MP at 24 h Mean (SD) | 15.41 (2.91) | 20.67 (6.69) | 24.59 (6.01) | 30.8 (5.59) | 0.001 |
| MP at 24 h Mean (SD) | 15.68 (4.19) | 19.43 (5.72) | 23.26 (6.42) | 29.72 (3.23) | 0.001 |
| MP at 24 h Mean (SD) | 15.5 (3.88) | 18.96 (5.91) | 22.88 (7.19) | 26.19 (5.59) | 0.001 |
| VFD, Median (IQR) | 0.0 (0-1) | 0 (0-15.25) | 0 (0-3.75) | 0 (0-0) | 0.11 |
| Length of ventilator day Median (IQR) | 15 (10-29) | 16.5 (7-27.25) | 19 (12-35) | 22 (17-31) | 0.85 |
| Length of stay in ICU (days), Median (IQR) | 26 (13-43) | 33.5 (18.25-48.5) | 22.5 (14.25-52) | 25 (18.5-36.5) | 0.63 |
| Length of stay in hospital (days), Median (IQR) | 27 (14-58) | 37 (24-59.25) | 29 (18.75-53) | 36 (23-37) | 0.59 |
| Outcome at 28 day: Mortality, n, (%) | 6 (40) | 14 (28) | 12 (42.9) | 6 (66.7) | 0.001 |
Data are presented as median (IQR), number of patients (percentage) and mean (SD). IQR=interquartile range; SD=standard deviation; ARDS=acute respiratory distress syndrome; PBW=predicted body weight; CCI=Charlson comorbidity index; APACHE=Acute physiology and chronic health evaluation; SOFA=Sequential organ failure assessment; PF ratio=PaO2/FiO2 ratio; Pplt=plateau pressure (in cmH2O); DP=driving pressure (in cmH2O); MP=mechanical power ((joules/min); Vt_IBW: tidal volume/ideal body weight; VFD=ventilator free day; ICU=intensive care unit
Collectively, our findings suggest that lung-protective ventilation should extend beyond limiting tidal volume and plateau pressure to include minimisation of overall MP. Incorporating MP into routine ventilator assessment may provide a more holistic understanding of ventilator-induced stress. Future multicentre studies are warranted to define optimal MP thresholds and develop MP-guided ventilation strategies tailored to different patient populations.
The limitations of our study include its observational nature, small sample size, and use of APACHE II. In our study, non-survivors had a higher severity score and lower PF ratio that can affect outcome. MP was calculated at four predefined time points. Continuous measurement of MP would have been the ideal method. Additionally, the majority of patients in our cohort were ventilated using PCV. This mode, while clinically relevant, presents specific limitations in terms of MP estimation—notably, the indirect calculation of Pplt and DP, which may introduce approximation error, especially in patients with spontaneous efforts or variable compliance. Large sample size and different cohorts would have been added to strengthen our study.
The strengths of our study include its prospective nature with serial MP measurements, enabling the temporal evaluation of MP. Our study is amongst the first from India to examine MP trends within the first 72 hours of ventilation and its influence on outcome. This highlights the potential for MP trend monitoring to facilitate timely ventilator adjustments and possibly improve the outcomes. This study provides a cut-off value of MP to predict mortality outcome which needs to be validated in larger patient populations.
CONCLUSION
In conclusion, our findings underscore the relevance of MP in the context of mechanical ventilation and its association with patient outcomes. Integrating MP monitoring into routine ventilatory parameter monitoring may support more individualised ventilation strategies. However, large multicentre studies in different cohorts are needed to validate the role of MP in outcome of critically ill ventilated patients.
Presentation at conferences/CMEs and abstract publication
None.
Study data availability
De-identified data may be requested with reasonable justification from the authors (email to the corresponding author) and shall be shared after approval as per the authors’ institution’s policy.
Disclosure of use of artificial intelligence (AI)-assistive or generative tools
The authors confirm that no AI tools or language models (LLMs) were used in the writing or editing of the manuscript, and no images were manipulated using AI.
Declaration of use of permitted tools
Tools used are freely available and not copyrighted.
Supplementary material
This article has supplementary material and can be accessed at this link. Supplementary Material at http://links.lww.com/IJOA/A72.
Conflicts of interest
There are no conflicts of interest.
Acknowledgements
None
Funding Statement
Nil.
REFERENCES
- 1.Slutsky AS, Ranieri VM. Ventilator-induced lung injury. N Engl J Med. 2013;369:2126–36. doi: 10.1056/NEJMra1208707. [DOI] [PubMed] [Google Scholar]
- 2.Protti A, Maraffi T, Milesi M, Votta E, Santini A, Pugni P, et al. Role of strain rate in the pathogenesis of ventilator-induced lung edema. Crit Care Med. 2016;44:838–45. doi: 10.1097/CCM.0000000000001718. [DOI] [PubMed] [Google Scholar]
- 3.Cressoni M, Gotti M, Chiurazzi C, Massari D, Algieri I, Amini M, et al. Mechanical power and development of ventilator-induced lung injury. Anesthesiology. 2016;124:1100–8. doi: 10.1097/ALN.0000000000001056. [DOI] [PubMed] [Google Scholar]
- 4.Gattinoni L, Pesenti A, Avalli L, Rossi F, Bombino M. Pressure–volume curve of total respiratory system in acute respiratory failure: Computed tomographic scan study. Am Rev Respir Dis. 1987;136:730–6. doi: 10.1164/ajrccm/136.3.730. [DOI] [PubMed] [Google Scholar]
- 5.Cressoni M, Cadringher P, Chiurazzi C, Amini M, Gallazzi E, Marino A, et al. Lung inhomogeneity in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med. 2014;189:149–58. doi: 10.1164/rccm.201308-1567OC. [DOI] [PubMed] [Google Scholar]
- 6.Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med. 1995;151:1807–14. doi: 10.1164/ajrccm.151.6.7767524. [DOI] [PubMed] [Google Scholar]
- 7.Amato MBP, Barbas CSV, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med. 1998;338:347–54. doi: 10.1056/NEJM199802053380602. [DOI] [PubMed] [Google Scholar]
- 8.Acute Respiratory Distress Syndrome Network , Brower RG, Matthay MA, Morris A, Schoenfeld D, Thompson BT, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342:1301–8. doi: 10.1056/NEJM200005043421801. [DOI] [PubMed] [Google Scholar]
- 9.Amato MBP, Meade MO, Slutsky AS, Brochard L, Costa ELV, Schoenfeld DA, et al. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med. 2015;372:747–55. doi: 10.1056/NEJMsa1410639. [DOI] [PubMed] [Google Scholar]
- 10.Villar J, Martín-Rodríguez C, Domínguez-Berrot AM, Fernández L, Ferrando C, Soler JA, et al. Quantile analysis of plateau and driving pressures and mortality in ARDS. Crit Care Med. 2017;45:843–50. doi: 10.1097/CCM.0000000000002330. [DOI] [PubMed] [Google Scholar]
- 11.Briel M, Meade M, Mercat A, Brower RG, Talmor D, Walter SD, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and ARDS: Systematic review and meta-analysis. JAMA. 2010;303:865–73. doi: 10.1001/jama.2010.218. [DOI] [PubMed] [Google Scholar]
- 12.Gattinoni L, Tonetti T, Cressoni M, Cadringher P, Herrmann P, Moerer O, et al. Ventilator-related causes of lung injury: The mechanical power. Intensive Care Med. 2016;42:1567–75. doi: 10.1007/s00134-016-4505-2. [DOI] [PubMed] [Google Scholar]
- 13.Giosa L, Busana M, Pasticci I, Bonifazi M, Macrì MM, Romitti F, et al. Mechanical power at a glance: A simple surrogate for volume-controlled ventilation. Intensive Care Med Exp. 2019;7:61. doi: 10.1186/s40635-019-0276-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Becher T, van der Staay M, Schädler D, Frerichs I, Weiler N. Calculation of mechanical power for pressure-controlled ventilation. Intensive Care Med. 2019;45:1321–3. doi: 10.1007/s00134-019-05636-8. [DOI] [PubMed] [Google Scholar]
- 15.van der Meijden S, Molenaar M, Somhorst P, Schoe A. Calculating mechanical power for pressure-controlled ventilation. Intensive Care Med. 2019;45:1495–7. doi: 10.1007/s00134-019-05698-8. [DOI] [PubMed] [Google Scholar]
- 16.Chiumello D, Gotti M, Guanziroli M, Formenti P, Umbrello M, Pasticci I, et al. Bedside calculation of mechanical power during volume- and pressure-controlled ventilation. Crit Care. 2020;24:417. doi: 10.1186/s13054-020-03116-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vassalli F, Pasticci I, Romitti F, Duscio E, Aßmann DJ, Grünhagen H, et al. Does iso-mechanical power lead to iso-lung damage? Anesthesiology. 2020;132:1126–37. doi: 10.1097/ALN.0000000000003189. [DOI] [PubMed] [Google Scholar]
- 18.Scharffenberg M, Wittenstein J, Ran X, Zhang Y, Braune A, Theilen R, et al. Mechanical power correlates with lung inflammation assessed by PET in experimental acute lung injury. Front Physiol. 2021;12:717266. doi: 10.3389/fphys.2021.717266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Nieman GF, Satalin J, Andrews P, Habashi NM, Gatto LA. Lung stress, strain, and energy load: Engineering concepts to understand VILI. Intensive Care Med Exp. 2016;4:16. doi: 10.1186/s40635-016-0090-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Guérin C, Papazian L, Reignier J, Ayzac L, Loundou A, Forel JM; Acurasys and Proseva Investigators Effect of driving pressure on mortality in ARDS. Crit Care. 2016;20:384. doi: 10.1186/s13054-016-1556-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhang Z, Zheng B, Liu N, Ge H, Hong Y. Mechanical power normalized to predicted body weight as a predictor of mortality in ARDS. Intensive Care Med. 2019;45:856–64. doi: 10.1007/s00134-019-05627-9. [DOI] [PubMed] [Google Scholar]
- 22.Serpa Neto A, Deliberato RO, Johnson AEW, Bos LD, Amorim P, Pereira SM, et al. Mechanical power of ventilation and mortality in critically ill patients. Intensive Care Med. 2018;44:1914–22. doi: 10.1007/s00134-018-5375-6. [DOI] [PubMed] [Google Scholar]
- 23.Gattinoni L, Pesenti A. The concept of the “baby lung”. Intensive Care Med. 2005;31:776–84. doi: 10.1007/s00134-005-2627-z. [DOI] [PubMed] [Google Scholar]
- 24.Vasques F, Duscio E, Cipulli F, Romitti F, Quintel M, Gattinoni L. Determinants and prevention of ventilator-induced lung injury. Crit Care Clin. 2018;34:343–56. doi: 10.1016/j.ccc.2018.03.004. [DOI] [PubMed] [Google Scholar]
- 25.Marini JJ, Jaber S. Dynamic predictors of VILI risk: Beyond the driving pressure. Intensive Care Med. 2016;42:1597–600. doi: 10.1007/s00134-016-4534-x. [DOI] [PubMed] [Google Scholar]
- 26.Chi Y, Zhang Q, Yuan S, Zhao Z, Long Y, He H. Twenty-four-hour mechanical power variation rate and mortality in acute respiratory failure. BMC Pulm Med. 2021;21:331. doi: 10.1186/s12890-021-01691-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wu HP, Chu CM, Chuang LP, Lin SW, Leu SW, Chang KW, et al. Association between mechanical power and mortality in pneumonia using pressure-targeted ventilation. Diagnostics (Basel) 2021;11:1862. doi: 10.3390/diagnostics11101862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Xie Y, Qian Y, Liu K, Liu S, Zheng H, Cao L, et al. Correlation between mechanical power and lung ultrasound score in ARDS. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2019;31:704–8. doi: 10.3760/cma.j.issn.2095-4352.2019.06.009. [DOI] [PubMed] [Google Scholar]
- 29.Parhar K, Zjadewicz K, Soo A, Sutton A, Doig L, Lam C, et al. Epidemiology, mechanical power, and long-term outcomes in ARDS. Ann Am Thorac Soc. 2019;16:1263–72. doi: 10.1513/AnnalsATS.201812-910OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Roca O, Peñuelas O, Muriel A, García-de-Acilu M, Laborda C, Sacanell J, et al. Driving pressure as a risk factor for ARDS in mechanically ventilated patients. Respir Care. 2021;66:1505–13. doi: 10.4187/respcare.08587. [DOI] [PubMed] [Google Scholar]
- 31.Tonna E, Peltan ID, Brown SM, Grissom CK, Presson AP, Herrick JS, et al. PEEP and respiratory rate modify the association of mechanical power with mortality in ARDS. Crit Care Explor. 2021;3:e0583. doi: 10.1097/CCE.0000000000000583. [DOI] [PMC free article] [PubMed] [Google Scholar]