ABSTRACT
BACKGROUND & AIMS: Sepsis is still a significant cause of death in the Intensive Care Unit and its early diagnosis is vital. Changes in cell permeability have been observed early in sepsis. Lower values of bioelectrical impedance (BIA) such as reactance adjusted by height (Xc/H) and phase angle (PA) have already been studied as a prognostic biomarker for many diseases and may indicate cell injury. BIA is a low cost, practical, noninvasive method that can be measured at bedside. This study investigated the utility of PA and Xc/H raw values in the pediatric critical care unit as predictors of progression to septic shock, as a clinical monitoring tool and to support the diagnosis of septic shock. METHODS: We prospectively analyzed bioelectrical impedance in 145 children aged between one month and six years who were not in septic shock on admission to the intensive care unit Serial bioelectrical impedance analysis (BIA) measures were analyzed to determine the sensitivity and specificity of accurately identifying children who subsequently developed septic shock. Kaplan-Meier septic shock-free survival curves modeled by Xc/H and PA were done. RESULTS: The free-septic shock survival curve analysis showed that patients with the lowest median values of Xc/H and PA were associated with the highest percentage of occurrence of septic shock (p = 0.0001 for Xc/H and <0.0006 for PA) and longest length of stay in the intensive care unit (p < 0.0011 for Xc/H and p < 0.004 for PA). Values of Xc/H below 48.63 Ohm/m at admission showed statistically significant odds ratio (OR) of 3.72 for developing septic shock any time during the hospitalization period, with a 87% sensitivity, 35% specificity and an area under the curve (AUC) of 0.62. The PA at admission did not show significant results. During hospitalization, patients with Xc/H below 35.72 Ohm/m were 3.38 times more likely to develop septic shock in the next day, with a sensitivity of 66.7%, a specificity of 62.3% and AUC of 0.65. PA values below 3.27 had an OR of 9.58 for a septic shock the next day with a sensitivity of 95.8%, specificity of 29.4% and AUC of 0.62. The presence of a value of Xc/H below 33 Ohm/m showed a strong association with the occurrence of septic shock on the same day of the measurement, with an OR of 11.7, as well as a value of PA below 2.64, showed an OR of 14.2. CONCLUSIONS: The bioelectrical parameters Xc/H and phase angle have limitations in predicting septic shock as isolated biomarkers, but have a potential role as a monitoring tool in the pediatric intensive care unit. The comparative value with other biomarkers remains to be elucidated.
