Effect of PEEP on Indirect Calorimetry in Critically Ill Patients

Overview

Indirect calorimetry is used in intensive care to measure how much oxygen a patient consumes and how much carbon dioxide is produced. These measurements are used to estimate energy expenditure and help guide nutritional treatment. In mechanically ventilated patients, however, ventilator settings may potentially influence the accuracy of indirect calorimetry. This study will investigate whether changes in positive end-expiratory pressure (PEEP), a routinely used ventilator setting, affect measurements obtained by indirect calorimetry in critically ill mechanically ventilated adults. Each participant will undergo indirect calorimetry at three PEEP levels: 8, 12, and 16 cmH2O. Each PEEP level will be maintained for a stabilization period before measurements are obtained. Other ventilator settings will be kept unchanged during the measurement sequence whenever clinically feasible. The study will compare oxygen consumption (VO2), carbon dioxide production (VCO2), respiratory quotient (RQ), and resting energy expenditure (REE) across the three PEEP levels. The aim is to determine whether clinically relevant changes in PEEP systematically influence indirect calorimetry measurements and their interpretation.

Indirect calorimetry (IC) is considered a reference method for assessment of energy expenditure in critically ill patients. It measures oxygen consumption (VO2) and carbon dioxide production (VCO2), from which respiratory quotient (RQ) and resting energy expenditure (REE) are calculated. Reliable measurements are important when IC is used to guide nutritional therapy in mechanically ventilated intensive care patients. Measurement accuracy may potentially be influenced by ventilator settings and changes in respiratory mechanics. Positive end-expiratory pressure (PEEP) is frequently adjusted during intensive care, but there is limited clinical evidence regarding whether changes in PEEP within commonly used ranges systematically affect contemporary indirect calorimetry measurements. The approved research project therefore includes a physiological substudy specifically examining the influence of PEEP on VO2, VCO2, RQ, and REE. This is a prospective, single-center, repeated-measures physiological study conducted in mechanically ventilated adult intensive care patients at Sahlgrenska University Hospital. Participants undergo indirect calorimetry during a standardized sequence of three PEEP levels: 8, 12, and 16 cmH2O. Each participant therefore serves as their own control. At each PEEP level, the patient is allowed to reach a stable condition before indirect calorimetry measurements are obtained. The ethics protocol specifies a stabilization period before measurement and that other ventilator settings should remain unchanged during the PEEP sequence. The final operational study protocol uses PEEP levels of 8, 12, and 16 cmH2O. Indirect calorimetry measurements will include VO2, VCO2, RQ, and REE. Measurements will be performed under standardized conditions, with attention to measurement stability and ventilator leakage. In the underlying research protocol, acceptable IC measurement periods are defined using steady-state criteria and controlled ventilator conditions. The primary objective is to determine whether VO2, VCO2, RQ, and REE change systematically across PEEP levels of 8, 12, and 16 cmH2O. Repeated measurements will be analyzed using mixed-effects models accounting for clustering of observations within participants. PEEP changes are performed within clinically accepted ranges under continuous intensive care monitoring. The measurement sequence will be interrupted if clinically relevant deterioration in hemodynamics, oxygenation, or other physiological parameters occurs. The ethics protocol specifically requires PEEP changes to remain within clinically acceptable levels and to be discontinued if adverse physiological effects are observed. A relative difference of 10% in indirect calorimetry-derived measurements between PEEP levels was considered clinically meaningful. Assuming a within-participant standard deviation of approximately 12.5%, corresponding to a standardized paired effect size of 0.8, approximately 15 evaluable participants are required to achieve 80% power at a two-sided alpha level of 0.05. To account for incomplete measurement sequences and technically inadequate indirect calorimetry measurements, 20 participants are planned for enrollment.