• Nanna Dyhre-Petersen
Background: Water is a vital nutrient of life and a multifunctional constituent of the human body thus, especially lack of water (dehydration), but also excessive amounts of water (overhydration) cause various functionally disturbances in the body. Persons with intestinal insufficiency (INS) or intestinal failure (IF) on respectively, oral nutrition (ON) or home parenteral nutrition (HPN) are at risk of abnormal water balance due to altered/impaired gastrointestinal functions. Thus, an important aspect of disease management is therefore proper evaluation of hydration status. However, no single gold standard exists. Techniques such as bioelectrical impedance analysis (BIA) and bioelectrical impedance vector analysis (BIVA) have been presented as new possible alternatives to older techniques used in clinical settings. Though, it is still unknown how well these techniques perform as hydration assessment methods in patients with INS or IF. Objectives: To evaluate the performance of BIA and BIVA as hydration assessment techniques in INS- and IF-patients with calculated plasma osmolarity and 24-hour urine volume as reference techniques. Methods: Hydration status was evaluated in 253 metabolic stable patients with either INS (n=125) or IF (n=128) according to calculated plasma osmolarity, 24-hour urine volume, whole-body multi-frequency-BIA and BIVA. Patients were classified by each parameter as either dehydrated, euhydrated, or overhydrated according to specific reference intervals. Correlation between parameters was assessed by Pearson’s product-moment correlation while reliability was tested by weighted Kappa (κw) with linear weights. Differences between ON- and HPN-patients (i.e. patients with INS and IF, respectively) in regard to demographics and clinically values as well as hydration classification were investigated by Test of two proportions, Independent-samples T-test, and Chi-square test of homogeneity (r x 2)/Fisher’s exact test (r x 2) with post hoc test where appropriate. Results: A statistically significant correlation (p <0.05) was only found between plasma osmolarity and BIA-data as well as BIVA-data, although poor (correlation coefficient ranging from -0.150 to -0.245). Assessment by plasma osmolarity and 24-hour urine volume resulted in a hydration classification that was statistically significant different (p <0,05) between ON- and HPN-patients. Post hoc test could not confirm between which hydration status the difference existed. None of the other assessment techniques could discriminate ON-patients from HPN-patients based on hydration classification (p >0,05). Significant agreement (p <0,05) between techniques was only demonstrated for plasma osmolarity and BIVA as well as plasma osmolarity and BIA, both with an agreement less than the one expected by chance (κw <0.0). Conclusion: The study demonstrated that hydration classification of INS- and IF-patients varied with choice of hydration assessment technique and that no agreement above the one expected by chance existed between standard techniques (plasma osmolarity and 24-hour urine volume) and novel techniques (BIA and BIVA). It was neither possible to conclude if ON-patients where classified more often or fewer times as dehydrated, euhydrated, or overhydrated than HPN-patients. Further studies are recommended with improved study design in order to verify the present study’s results.
Publication date29 May 2020
Number of pages51
External collaboratorAarhus University
no name vbn@aub.aau.dk
ID: 333222163