Effects of End-inspiratory Pause on Ventilation

Reduction of tidal volume (VT) to 6 mL/kg normalized to predicted body weight (PBW) and driving pressure limitation up to 15 cmH20 are essential ventilatory strategies that positively impact clinical outcomes of acute respiratory distress syndrome (ARDS) patients. Recently, mechanical power has also been linked to ventilator-induced lung injury (VILI) and associated with mortality, mainly due to its dynamic-elastic effects. However, despite using these low VT strategies, it is possible to develop VILI in moderate to severe ARDS patients, especially when driving pressure exceeds established safety limits for lung protection. In these situations, a VT less than 6 mL/kg-PBW is recommended, but an increase in respiratory rates (RR) is required to counteract the side effects of secondary hypercapnia (PaCO2 levels greater than 45 mmHg). In turn, the RR needed to achieve this target is at least 25 breaths/min, even if a mild degree of hypercapnia is often accepted. Yet, mechanical ventilator programming with higher RR per minute may promote lung tissue inflammation and could be associated with adverse outcomes in ARDS.

Thus, in ARDS is challenging to establish a combined strategy of tidal volume (VT) and respiratory rate RR reduction since its utmost problem is an increase in the partial pressure of carbon dioxide (PaCO2) and impairment of ventilatory efficiency, the magnitude of which is linked to the disease severity. For this reason, some non-invasive strategies are available to attenuate hypercapnia and improve the lung's ability to CO2 clearance. Among them, it is worth highlighting the end-inspiratory pause (EIP). This strategy has shown its usefulness in increasing CO2 clearance by improving the mean time given to inspired gas for distribution and diffusive mixing within the lungs. Nevertheless, the role of EIP as part of a protective ventilation strategy that combines the VT and RR reductions is unknown.

Therefore, we hypothesize that adding an end-inspiratory pause (EIP) improves CO2 expiration and that such an effect allows a decrease in the VT and RR for more lung-protective purposes. Thus, the main objective of this study is to evaluate the effects of an EIP on Bohr´s dead space (VDBohr/VT) and PaCO2 when combined with a VT and the RR reduction in ARDS. The secondary aim is to evaluate ventilatory strategy on air distribution and homogeneity by electrical impedance tomography.

Patient selection Patients ≥18 years of age with mild, moderate, and severe ARDS up to 5 days of mechanical ventilation. Patients must be under deep sedation and neuromuscular paralysis. Patients with hemodynamic instability, acute heart failure, previous chronic respiratory disease, and variations in oesophageal temperature higher than 0.5 °C in the last 2 hours will be excluded.

Baseline mechanical ventilation settings Mechanical ventilation at baseline will be programmed in volume-controlled mode with a Servo-i (Maquet, Solna, Sweden) and the following parameters: a VT of 7 ml/kg-PBW, an RR adjusted to ensure an arterial pH greater than 7.30 and no intrinsic positive end-expiratory pressure (PEEP), an inspiratory insufflation time of 0.6 seconds, an inspiratory: expiratory (I: E) ratio of 1:2, and no EIP. End-expiratory transpulmonary pressure will be set at the beginning of the study by electrical impedance tomography. The PBW was calculated as follows: 50+ [0.91 *(height in cm-152.4)] for men and 45.5+ [0.91* (height in cm-152.4)] for women. Active humidification will be used in all participants, and compressible volume compensation was performed in each mechanical ventilator used before starting the protocol.

Respiratory mechanics Airway pressure, esophageal pressure, and gas flow will be measured continuously by a proximal pneumotachograph (MBMED, Buenos Aires, Argentina). Data will be downloaded on a second laptop after proper flow and pressure sensors calibration. Respiratory driving pressure will be determined as plateau pressure minus total PEEP (PEEPTOT = external PEEP + intrinsic PEEP) and respiratory system compliance as Crs = VT/Pplat-PEEPTOT.

Respiratory mechanics were calculated offline using specific software, elastic-dynamic power was calculated according to Costa et al study, and minute ventilation was indexed by PBW.

Volumetric capnography Expired CO2 will be measured by an infrared mainstream sensor (Capnostat 5®; USA) and integrated into a monitor (MBMed CO2 Module). Volumetric capnograms were reconstructed automatically using MATLAB® (Natick, MA, USA), which allows ventilation analysis in volumetric capnograms using a mathematical algorithm that adjusts the tidal volume to the exhaled CO2. The following parameters were recorded: the airway dead space fraction (VDaw/VT), VDBohr/VT, index of gas exchange (VDEnghoff/VT), CO2 elimination per breath (VTCO2,br), alveolar minute ventilation, the fraction of expired CO2 (FECO2), minute elimination of CO2 (VTCO2,br * RR), end-tidal CO2 and the mean alveolar partial pressure of CO2 (PACO2).

The capnogram at phase III (SIII) is normalized (SnIII) by the FECO2 of the corresponding expiratory cycle. SnIII allows a comparison of slopes from breaths with different CO2 excretion rates, which could be expected to occur during modifications in the setting of the VT or RR. Alveolar minute ventilation will be calculated using the formula: alveolar ventilation * RR indexed by PBW. Ventilatory efficiency is defined as the relationship between alveolar ventilation per minute and minute ventilation.

The flow sensor is automatically adjusted to values recorded with the Body Temperature \& Pressure Saturated (BTPS) conversion factor to determine the exhaled air volume.

Electrical impedance tomography

A 16-electrode belt will be placed in the mid-thoracic region, and continuous lung impedance will be assessed by electrical impedance tomography (EIT) (Dräger Medical Systems, USA). Offline analysis of EIT data will be performed, and the following parameters were calculated:

• Lung inhomogeneity is determined by the global inhomogeneity index (GI), which quantifies the homogeneity of the tidal volume distribution through quantitative lung pixel impedance dispersion.
• Regional ventilation distribution is assessed by the impedance ratio (IR), a parameter for determining the dependent and non-dependent regions' air distribution. An IR > 1 represents a ventral distribution predominance, while an IR < 1 represents mainly a dorsal distribution. Briefly, the ventral region corresponds to regions of interest 1 and 2, and the dorsal region corresponds to regions of interest 3 and 4.
• Tidal variation of impedance (TVI) represents impedance change generated by inspired gas during a respiratory cycle.
• End-expiratory lung impedance (EELI) corresponds to the impedance value at the end of expiration.

Protocol All patients will be lying in a semi-recumbent position at 40 degrees, and before starting the study, a period of stabilization of 60 minutes must be mandated. PEEP and insufflation time settings will be kept constant throughout the study. For safety reasons, we predefined that the protocol could be stopped at any time in the event of significant alterations in systemic hemodynamics and/or respiratory acidosis (pH ≤ 7.25 and PaCO2 ≥ 60 mmHg).

The following data were collected at inclusion: demographic variables (age, sex, height), Acute Physiology, and Chronic.

Sequential protocol steps

• Step I: Tidal volume (VT) 7 ml/kg-PBW, with constant positive end-expiratory pressure (PEEP), respiratory rate (RR), and insufflation time (0.6 sec), without adding an end-inspiratory pause (EIP). inspiratory: expiratory (I: E) ratio of 1:2. 60 minutes long.
• Step II: VT 5 ml/kg-PBW with constant PEEP, RR, and insufflation time (0.6 sec) without adding an end-inspiratory pause (EIP). I:E ratio of 1:2. 60 minutes long.
• Step III: EIP will be configured to achieve an I:E ratio of 1:1, keeping constant the insufflation time at 0.6 sec and the VT at 5 ml/kg-PBW. PEEP and RR will not be changed from baseline conditions. 60 minutes long.
• Step IV: The RR will be reduced by 20%, keeping constant the VT in 5 ml/kg-PBW. The insufflation time will be maintained at 0.6 sec, and the I:E ratio was equal to 1:1; consequently, the EIP will be prolonged again. 60 minutes long.

Volumetric capnography will be recorded, and an offline analysis will be performed using the mean value of the last 40 breaths. Gas exchange measurements were also performed at the end of each phase using a conventional blood gas analyzer (GEM® 4000, Instrumentation Laboratory, Lexington, USA). Hemodynamic variables and pulse oximetry will be continuously monitored (Multiparameter Spacelabs 91393). To identify the influence of cardiac output variations on CO2 dynamics, minimally invasive cardiac output monitoring will be installed (Edwards Lifesciences, USA). If needed, a norepinephrine infusion will be used to maintain the mean arterial pressure near 65 mmHg, and no bolus of fluid was administered during the protocol.

Statistical analysis The Shapiro-Wilk test will be performed to determine the distribution of continuous variables, which will be expressed as mean, standard deviation, or median and interquartile range as appropriate. Friedman´s nonparametric test, the Wilcoxon matched-pairs signed-rank test, and Dunn's post hoc comparisons will be performed to compare multiple variables.

A two-tailed p-value less than 0.05 was considered statistically significant.