Predicting Fluid Responsiveness Using Transiently Increased Intrathoracic Pressure in Mechanically Ventilated Patients

Overview

The goal of this study is to identify in patients requiring active fluid resuscitation and mechanical ventilation for circulatory shock, can a controlled increase in intrathoracic pressure (either by positive-end expiratory pressure (PEEP) or tidal volume (TV)) predict responsiveness to additional fluid resuscitation. We hypothesize that a temporary, physiologically-safe increase in positive-end expiratory pressure (PEEP) and/or a temporary increase in tidal volume (from 6 cc/kg predicted body weight (PBW) to 8 cc/kg PBW) in patients requiring invasive mechanical ventilation will predict fluid responsiveness based upon an assessment of the change in pulse pressure and stroke volume variation.

Once an eligible subject is identified and written or telephone consent is obtained from the patient or surrogate as appropriate, we will record demographic data and clinical information including age, sex, height, actual body weight, predicted body weight (PBW), APACHE II on admission, primary diagnosis, reason for ICU admission, and dose of vasopressors. Patients will be intubated and on assist-control volume ventilation receiving low tidal volumes. Sedation will be titrated per current MICU protocol to minimize patient discomfort and minimizing ventilator dyssynchrony. In keeping with standard practice, the dose of vasoactive medications will be kept constant during the intervention period. The standard practice is to titrate vasopressor therapy every five minutes, so keeping vasoactive medications constant for sixty seconds is in keeping with the standard of care. All subjects will be in an ICU setting and monitored per standard of care (vital signs, telemetry, ventilator data) by a respiratory therapist, pulmonary and critical care fellow or faculty member, and critical care nurse at the bedside throughout the entire intervention. A point of care (POC) echocardiogram will be performed by the investigator performing the intervention in order to measure the stroke volume index (surrogate for cardiac output). Subjects will be randomized to either undergo the PEEP challenge or tidal volume challenge first. After the initial challenge, subjects will cross over to whichever challenge has not yet been performed. The simple changes to the ventilator we are proposing to use as challenges in this study are done frequently and routinely in the ICU with or without the supervision of a physician, but we will have a physician present throughout the entire proposed intervention. For the subjects undergoing PEEP challenge first, we will do the following: 1. POC echocardiogram performed to measure SVI 2. The tidal volume will be set to 6 mL/kg PBW and the patients will be monitored for 5 minutes for any variability in vital parameters. The subject's ventilatory parameters, vitals, hemodynamics, and oxygenation will be recorded on initial ventilator settings. 3. PEEP challenge - PEEP will be increased by 50% for 60 seconds. The subject's ventilatory parameters, vitals, hemodynamics, and oxygenation will be recorded after 60 seconds. 4. PEEP returned to initial settings. 5. 2 minute washout period to prepare for tidal volume challenge. 6. Tidal volume challenge - tidal volume raised to 8 cc/kg PBW for 60 seconds. The subject's ventilatory parameters, vitals, hemodynamics, and oxygenation will be recorded will be recorded after 60 seconds. 7. Subjects ventilator settings returned to initial settings x 2 minutes. 8. 500 cc crystalloid fluid bolus of the ICU team's choosing is then administered to the patient. The subject's ventilatory parameters, vitals, hemodynamics, and oxygenation will be measured. 9. POC echocardiogram performed to measure SVI and determine if the subject is a fluid responder or non-responder. The procedure for those randomized into the tidal volume challenge first will be similar, but the tidal volume challenge (e, above), rather than the PEEP challenge (b, above), will be performed first. In the management of a critically ill patient in circulatory shock, physicians encounter the question of whether to administer fluid to the patient or not several times during the course of the patient's stay in the ICU. To ensure that our study is generalizable and applicable, we will perform the above protocol up to three times per subject. This is similar to previous studies using changes in intrathoracic pressure to predict fluid responsiveness. Additionally, we will have a larger number of interventions in order to better show a clear difference statistically.