87 PHASE ANGLE AND LONG-TERM OUTCOME 5 diminished cellular function. There is debate whether PhA can be used to determine pure disease-related malnutrition, and there is overlap as many clinical conditions, including underlying malignancies, lead to a higher protein catabolic rate. In a general population, sex, age, and BMI are important determinants of phase angle.16 With increasing age, PhA decreases as the amount of muscle tissue decreases. A higher BMI is associated with a higher PhA, albeit within the normal range. However, in this study there were no differences in sex, age, and BMI between survivors and non-survivors. In general, an AUC between 0.6 and 0.7 is considered fair, but not an all-defining silver bullet. Comparison of the odds between the independent risk factors should be done with caution. The presented odds ratios for PhA are per grade or per point in case of APACHE III score, but dichotomous with respect to the presence of malignancy. Another important factor in the interpretation of the PhA is the hydration status, since fluid overload leads to a decrease in resistance and a lower value of PhA. Vice versa, in patients with dehydration, resistance will increase which results in a higher phase angle.15 This limits the ability for sequential PhA measurements during ICU-admission, when fast shifts during fluid resuscitation and de-escalation are to be expected. To summarise, on ICU-admission BIA-derived PhA may provide additional objective and easy-to-obtain information on a patient’s body composition and, as a proxy for physical frailty, serve as an indicator for their ability for long-term recovery. Further research is needed to determine whether PhA measurement is indeed an acceptable proxy for nutritional status or physical frailty. Furthermore, the potential for PhA monitoring in the post-ICU phase in the guidance of rehabilitation interventions needs further elucidation. The main limitation of the study is the fact that in about one-third of admissions during the study period PhA measurement was not performed, creating potential bias. Post-hoc analysis revealed that characteristics of this group were comparable in all major determinants of PhA (S1 Table), and therefore probably did not influence the outcome of the study. In conclusion, in this single centre study PhA was independently associated with 1-year all-cause mortality after ICU-admission. However, more evidence is needed to establish the added clinical value of bioimpedance measurements prior to and after ICU-admission.
RkJQdWJsaXNoZXIy MjY0ODMw