Comparative Evaluation of smART+ REE Module to an FDA-cleared Device, for REE Calculation in ICU Patients

Mechanically ventilated patients are unable to take food orally and therefore are dependent on enteral nutrition for provision of both energy and protein requirements. The conditions of ICU patients dramatically change throughout their ICU stay; "diseases, like particularly sepsis, trauma and burns, cause a clinically relevant increase in resting energy expenditure (REE) of between 40%-80%," and most patients do not reach their required protein and caloric targets. The American Association for Parenteral and Enteral Nutrition (ASPEN) recommends that for patients who are at high risk for malnutrition, efforts should be made to provide >80% of the estimated or calculated energy and protein goal within 48-72 hours, starting with half the patient's calorie goal with the rate slowly increasing over time, in order to achieve the clinical benefit of enteral nutrition over the first week of hospitalization .However, studies have shown that more than 74% of ICU patients failed to receive at least 80% of their prescribed nutrition .

Predictive equations (such as Harris-Benedict equation) or a simple weight-based formula (48.1% and 48.7% respectively) have therefore been the most practiced methods of determining energy needs; however, the literature clearly indicates that each equation has a large potential for error. This makes it difficult to accurately predict an individual patient's energy requirements during critical illness. In general, predictive equations estimate accurately only 50% of the time in ICU patients , leading to over or under evaluation of the nutritional needs and inducing over or underfeeding . Numerous meta-analyses have demonstrated the poor value of predictive equations , , variability that is increased because body weight remains a value difficult to accurately assess .

To overcome the inaccuracy in determining the patient's energy consumption and the accompanying problems, the European Society for Parenteral and Enteral Nutrition (ESPEN) guidelines recommend to determine patients' REE based on indirect calorimetry (IC) or VCO2-based equation (VCO2 x 8.19). Unfortunately, this is not performed continuously during the course of enteral feeding to assess the ever-changing nutritional needs of the patient. Estimations showed that only 2% of ICUs were regularly using REE / IC, and therefore they must resort to performing a one-time calculation of the patient's energy consumption to manually calculate the caloric intake, food type and feeding rate.

This study aims to assess the accuracy of the smART+ System's REE module, compared to the FDA-cleared COSMED device, and nutritional calculations conducted with predictive equations regularly used by the clinical staff at the site.

The smART+ System is a modular nutrition management system intended to be used in a clinical setting. At the heart of the system is the console, which stores, compiles and displays data from the different system modules.

The REE Module is the only feature of the smART+ System being evaluated in this study. The REE Module is comprised of a disposable flow rate meter and a reusable CO2 detector that are added between the patient's endotracheal tube and ventilator tubing. The REE module is designed to continuously measure the patient's energy expenditure and deduce the patient's nutritional requirements according to the following equation:

REE (kcal) = [VCO2 (ml / min) * 3.941 / RQ + VCO2 (ml / min) * 1.11] * 1.44, where RQ=VCO2/VO2 A detailed description of the indirect calorimetry method, based on the above REE equation, is provided in the user manual (IFU) of the smART+ device.

The REE Module is intended to calculate the ventilated patient's REE, based on carbon dioxide monitoring, and deduce the patient's nutritional requirements.

This is a prospective, comparative single-center (three satellites) study assessing the accuracy of the smART+ REE and VCO2 measurements when compared to an FDA-cleared device indicated for REE measurement in ventilated patients (COSMED Q-NRG+ Portable Metabolic Monitor, K190800). The design of the study is as follows:

1. Each patient will undergo metabolic measurements utilizing both the smART+ System's REE Module and the COSMED device. The particular measurements collected by each device will be: REE, VCO2 and VO2 for the COSMED device, and REE, VCO2 for the smART+ System. A "measurement round" will include REE calculations (which also includes the other metabolic measurements) by both devices, which will be used simultaneously for each patient so as to avoid time-related bias which would ensue if their measurements were taken sequentially (regardless of order). To ensure adequate data is collected for a reliable REE calculation, 20 minutes (+/- 2 minutes) of "stable recording" will be taken, meaning no change in the patient's respiratory and hemodynamic status. To ensure the analysis is performed on a stable recording, the post-analysis of the data collected during the 20-minute recording will focus on the most stable 5-minute segment, determined based on the variability timeframe of the VCO₂ as indicated by the COSMED system. After completing the first measurement round, a minimum wait time of 12 hours is required before beginning the next measurement round. Each patient will undergo up to 4 measurement rounds in total. The metabolic condition of a patient is reflected by O2 consumption and CO2 production, which are continuously changing due to substrate oxidation. To accurately perform indirect calorimetry (IC) measurements, the patient's condition must be stable. That said, the metabolic state of a stable patient may differ from the previous measurement when at least 12 hours have passed. This variability was demonstrated in a study that conducted repeated measurements every hour. Other studies have employed similar approaches. According to the literature, multiple measurements are commonly used to compare metabolic data. For example:

• The Mindray study included 40 measurements from 16 patients.
• The Sabatino study analyzed 130 measurements from 42 patients.

1. For each patient, the parameters used as inputs for calculating REE via the 7 predictive equations listed below will be recorded in the patients' CRFs.
2. All clinical decisions will be made in accordance with the hospital's Standard of Care (SOC), without any influence from the measurements and calculations obtained during the study (i.e., from the test methods: smART+ System, COSMED device or predictive equations). If one of the tested methods is also used as the SOC in the participating department, the clinical team should continue to follow the department's established SOC without deviation.
3. Systems connection order on ventilation line: The smART+ REE Module shall be connected first, followed immediately by connection of the COSMED device. This ensures no airflow disruption or interference of one device on the other. Measurements from both devices will be taken simultaneously.