Dr. Emily L. Player|Professor Alastair Forbes.
Presented at DDF 2015
INTRODUCTION: Assessment of nutritional status in the clinical setting is difficult and often poorly achieved. Evidence for malnutrition in hospital is clear, with one study showing that 4 out of 5 patients were unable to meet their nutritional demands . Bioelectrical impedance analysis (BIA)-derived phase angle (PA) is a simple and non-invasive tool which refines the determination of nutritional status, reflects body cell health, and has prognostic value. However, BIA is not much used in clinical practice because of protocols which oblige the patient to be starved and on bed-rest. The research underpinning these protocols is dated and insecure; thus the restrictions may not be necessary.
SUBJECTS /METHODS: BIA measurements using Bodystat Quadscan 4000 were taken on 50 hospital in-patients who were starved and rested. It was then repeated on two occasions: following controlled exercise and after eating breakfast. Inclusion Criteria – Adult in-patients with capacity to consent. Exclusion Criteria – Any unstable patients, Patients who were pregnant or breastfeeding, Age <18 years, Those who were unavailable for the study morning, nil by mouth, Patients in whom bioelectrical impedance testing would be impossible or uninterpretable (e.g. bilateral amputees. Data was analysed on STATA.
RESULTS: PA results were typical of a hospital patient cohort Results were collected over a 25 week period by the research team. Male (31, 62%) Female (19, 38%) age range 18-86 years, all acute medical admissions Data was parametric and analysed using paired t tests Neither exercise nor eating made a statistically nor clinically significant difference to the PA.
CONCLUSION: In summary, the current protocols of starving and resting patients appear unnecessary and outdated It is probable that PA measurement is a practical means to improve assessment of nutritional status in hospitalised patients. Limitations include a small sample size (feasibility study) and 3 clinical researchers resulting in probable user variability with the BIA machine.