This prospective observational study aims to validate the dynamic mechanical power (MPdyn) formula by comparing it with the gold-standard geometric method (MPgeo) in patients with acute respiratory distress syndrome (ARDS) who are mechanically ventilated in volume-controlled ventilation (VCV) and pressure-controlled ventilation (PCV) modes.

Mechanical power (MP) is a composite parameter that integrates multiple components of ventilator-induced lung injury (VILI) and has shown strong associations with mortality in ARDS. While several formulas exist for calculating MP in VCV and PCV modes, most require inspiratory resistance, which is not readily available at the bedside. The MPdyn formula, introduced by Asar et al., allows for bedside calculation without inspiratory resistance and has shown good agreement with established formulas such as MPrs and MPLM. However, it has never been validated against the geometric method, which calculates mechanical power based on the area of the pressure-volume (P-V) loop and is considered the most accurate standard.

In this single-center study, 37 deeply sedated ARDS patients were ventilated with a Servo-U ventilator using both VCV and PCV modes. For each mode, two different I:E ratios (1:2 and 1:1) were applied, and 12 screenshots of full P-V loops were captured per patient, totaling 444 images. Geometric MP (MPgeo) was calculated using Python-based image processing with OpenCV and NumPy libraries. Dynamic mechanical power (MPdyn) was computed using ventilator-recorded values of minute volume (MVe), work of breathing per liter (WOBv), and PEEP.

The primary outcome was the agreement between MPdyn and MPgeo values under different ventilator modes and I:E ratios. Secondary outcomes included regression correlation (R²) and Bland-Altman analysis of bias and limits of agreement. This study seeks to determine whether MPdyn is a valid and reliable surrogate for geometric mechanical power, particularly in clinical settings where bedside calculation is needed and inspiratory resistance cannot be easily measured.

Mechanical power (MP) is an integrative parameter that quantifies the energy delivered to the lungs during mechanical ventilation. It accounts for multiple factors contributing to ventilator-induced lung injury (VILI), including tidal volume, airway pressures, respiratory rate, compliance, and resistance. MP has been shown to correlate strongly with morbidity and mortality in critically ill patients, particularly those with acute respiratory distress syndrome (ARDS).

Several MP calculation formulas have been proposed for volume-controlled ventilation (VCV) and pressure-controlled ventilation (PCV). While formulas such as MPrs (Gattinoni) and MPLM (Trinkle) are widely used, many of them require precise measurement of inspiratory resistance (Raw), which is often not readily accessible at the bedside-especially in PCV. To address this, a dynamic mechanical power formula (MPdyn) was developed by Asar et al., which estimates MP without the need for inspiratory resistance. MPdyn has demonstrated good agreement with MPrs in VCV and with MPLM in PCV. However, it has not been directly validated against the geometric method, which remains the gold standard for MP calculation.

The geometric method involves calculating the area enclosed by the pressure-volume (P-V) loop on the ventilator screen. This area represents the energy per breath, which can be multiplied by respiratory rate to yield MP. Modern ventilators, such as the Maquet Servo-U, provide high-resolution graphical displays that allow for precise area calculations using digital image processing.

This prospective observational study was conducted in the intensive care unit of Bakırköy Dr. Sadi Konuk Training and Research Hospital, Istanbul. A total of 37 ARDS patients who were deeply sedated and under controlled mechanical ventilation were enrolled. Each patient was ventilated in both VCV and PCV modes with two different I:E ratios (1:2 and 1:1). For each setting, three screenshots of the P-V loop were captured, resulting in 12 images per patient and 444 images in total.

Geometric mechanical power (MPgeo) was calculated using Python-based image analysis (OpenCV, NumPy, and PIL libraries). The region of interest (ROI) corresponding to the P-V loop was extracted, binarized, and segmented to compute the loop area. This pixel-based area was then scaled using the known tidal volume (TVe) and pressure (Pinsp) values to obtain mechanical energy per breath. MPgeo was calculated as:

MPgeo (J/min) = [E_breath × RR] + [PEEP × TVe × RR × 0.098]

Dynamic mechanical power (MPdyn) was calculated using the following formula:

MPdyn = MVe × [WOBv + (PEEP × 0.098)] where MVe is the expiratory minute volume (L/min), WOBv is the work of breathing per liter (J/L), and PEEP is the end-expiratory pressure (cmH₂O).

Statistical analysis included descriptive statistics, Shapiro-Wilk test for normality, univariable linear regression to assess correlation between MPdyn and MPgeo, and Bland-Altman analysis to evaluate agreement between methods. Paired measurements were analyzed both at the patient-averaged level and for all individual data points.

The primary outcome of the study was the degree of agreement between MPdyn and MPgeo under different ventilator modes (VCV, PCV) and I:E ratios (1:1, 1:2). Secondary analyses included evaluating potential biases and limits of agreement.

This study is expected to validate MPdyn as a reliable, practical alternative to geometric calculations in clinical settings, facilitating routine bedside monitoring of mechanical power without requiring complex measurements or advanced equipment.

Inclusion Criteria:

• Age ≥ 18 years
• Diagnosis of moderate-to-severe ARDS according to the Berlin criteria
• Receiving mechanical ventilation in volume-controlled (VCV) or pressure-controlled (PCV) mode
• Stable hemodynamics during data collection
• Expected to remain under mechanical ventilation for at least 48 hours
• Written informed consent from legal representative

Exclusion Criteria:

• Presence of chest tube or pneumothorax
• Known neuromuscular disease affecting respiratory function
• Pregnancy
• Patients with do-not-resuscitate (DNR) or limitation-of-therapy orders
• Incomplete ventilator data or poor-quality screenshots

The study population will consist of adult ICU patients diagnosed with ARDS and receiving mechanical ventilation (VCV or PCV mode) at a tertiary care hospital. Data will be collected between 24 and 48 hours after ICU admission.

The individual participant data (IPD) will not be shared due to institutional restrictions and ethical considerations concerning patient confidentiality.