Aktuelle Ernährungsmedizin 2011; 36 - P4_2
DOI: 10.1055/s-0031-1276781

Evaluation of a noninvasive method, SenseWear Pro for calculation of energy expenditure in ventilated intensive care patients compared with indirect calorimetry and predictive equations

J Krüger 1, M Gründling 3, S Knigge 3, S Friesecke 2, N Giese 1, S Gärtner 1, MM Lerch 1, M Kraft 1
  • 1Department of Internal Medicine A und B
  • 2University Hospital of Greifswald
  • 3Department of Anaesthesiology and Intensive Care

Background: Continuous measurement of energy expenditure (EE) in ventilated, critically ill patients remains a challenge. Indirect calorimetry (IC) represents the gold standard method but is not always available. Practical predictive equations and formulas often lead to a considerable under- or overestimation. Due to the fluctuating energy needs of critically ill patients methods of continuously assessing energy metabolism are needed. The aim of this study was to compare SenseWearPro3 Armband (SWA) to IC and predictive equations in critically ill ventilated patients.

Methods: We prospectively included 34 critically ill, ventilated patients in the trial. During a 24-hour observation period, the metabolic rate was continuously measured using SWA. In addition, IC was performed and EE was determined by predictive equations (Harris-Benedict, Müller, Ireton-Jones, Swinamer, Fusco and Penn State). In order to identify possible confounding factors that influence EE, nursing/physiotherapy procedures, laboratory parameters, vital signs, body composition and SAPS-Score of patients were also determined. Measurement bias (ΔEE=[EE-SWA]-[EE-IC]) was calculated. Daily energy expenditure of IC vs. SWA (and SWA vs. predictive equations) was compared using the Bland Altman method. The t-test for paired samples was used for statistical analysis.

Results: 34 critically ill and ventilated patients (22 male, 12 female; means: 66,7±12,4 years; body mass index 27,6±5,7kg/m2) were investigated. A mean bias of ΔEE=-253,6±333,2kcal [range:-1406 to+488kcal; p=0,025] was calculated. Bland Altman Analysis revealed that SWA slightly underestimated IC energy expenditure in hypometabolic range[ΔEE=-33,6±228,2; p=0,565] but significantly underestimated IC values in the normometabolic range [ΔEE=-393,3±173,2; p=0,011] and in the hypermetabolic range [ΔEE=896,5 ±720,5kcal; p<0,001]. We also found significantly differences depending on heartrate and energy balance.

Conclusion: SWA is not routinely applicable to intensive care patients. Individual measurements vary too widely and should be used with caution until an improved algorithm, based on our results, has been programmed.