Bioelectrical impedance analysis (BIA)-derived phase angle (PA) is a practical aid to nutritional assessment in hospital in-patients

Abstract

Nutritional status can be difficult to assess. Bioelectrical impedance analysis (BIA)-derived phase angle (PA), and the plasma markers citrulline and transthyretin (pre-albumin) have the potential to assist, but the protocol of fasting and resting for BIA renders the investigation impractical for routine use, especially so in populations at high risk of malnutrition. 1 To clarify whether starving and resting are necessary for reliable measurement of PA. 2 To identify whether PA, citrulline and transthyretin correlate with nutritional status. Eighty consenting adult in-patients were recruited. Nutritional status was determined by subjective global assessment (SGA) used as gold standard. The Malnutrition Universal Screening Tool (MUST) was used and anthropometric measurements were performed. Serum was analysed for citrulline and transthyretin. PA was measured using Bodystat 4000. The PA was considered to define malnutrition when lower than reference ranges for sex and age, and severe malnutrition if more than 2 integers below the lower limit. Anthropometric measurements were categorised according to WHO reference centiles. Ordinal logistic regression estimated the strength of association of PA, citrulline and transthyretin with SGA. PA values in the different metabolic states were compared using paired t tests. All 80 subjects completed the BIA and the nutritional assessments in the 3 different states; 14 declined to provide blood samples for the biochemical assays. Malnutrition was identified in 32 cases, severe malnutrition in 14 cases, the remaining 34 cases were deemed not to be malnourished. PA was strongly inversely associated with SGA (Odds Ratio [OR] per unit increase=0.21, CI 0.12-0.37, p<0.001). PA was not influenced by exercise (p=0.134) or food intake (p=0.184). Transthyretin was inversely associated with malnourished/severely malnourished states (OR=0.98, 95% CI 0.97-0.99, p=0.001), but had poorer predictive values than PA. There was no significant association between citrulline concentration and SGA (OR=1.01, 95% CI 0.99-1.04, p=0.348). The BIA-derived PA reliably identifies malnutrition. It is strongly associated with SGA but requires less skill and experience, and out-performs circulating transthyretin, rendering it a promising and less operator-dependent tool for assessing nutritional status in hospital patients. Our novel demonstration that fasting and bed-rest are unnecessary consolidates that position.