Thoracic Fluid Content (TFC) Measurement Using Impedance Cardiography Predicts Outcomes in Critically Ill Children

Objective: Conventional methods of fluid assessment in critically ill children are difficult and/or inaccurate. Impedance cardiography has capability of measuring thoracic fluid content (TFC). There is an insufficient literature reporting correlation between TFC and conventional methods of fluid balance and whether TFC predicts outcomes in critically ill children. We hypothesized that TFC correlates with indices of fluid balance [FIMO (Fluid Intake Minus Output) and AFIMO (Adjusted Fluid Intake Minus Output)] and is a predictor of outcomes in critically ill children. Design: Retrospective chart review. Setting: Pediatric intensive care unit of a tertiary care teaching hospital. Patients: Children <21 years, admitted to our Pediatric Intensive Care Unit (PICU) between July- November 2018 with acute respiratory failure and/or shock and who were monitored for fluid status using ICON® monitor. Interventions: None. Measurements and Main Results: We collected demographic information, data on daily and cumulative fluid balance (CFB), ventilator, PICU and hospital days, occurrence of multi-organ dysfunction syndrome (MODS), and mortality. We calculated AFIMO using insensible fluid loss. We analyzed data using correlation coefficient, chi-square test and multiple linear regression analysis. We analyzed a total 327 recordings of TFC, FIMO and AFIMO as daily records of fluid balance in 61 critically ill children during the study period. The initial TFC, FIMO, and AFIMO in ml [median (IQR)] were 30(23, 44), 300(268, 325), and 21.05(-171.3, 240.2), respectively. The peak TFC, FIMO, and AFIMO in ml were 36(26, 24), 322(286, 334), and 108.8(-143.6, 324.4) respectively. The initial CFB was 1134.2(325.6, 2774.4). TFC did not correlate well with FIMO or AFIMO (correlation coefficient of 0.02 and -0.03, respectively), but a significant proportion of patients with high TFC exhibited pulmonary plethora on x-ray chest (as defined by increased bronchovascular markings and/or presence of pleural effusion) (p = 0.015). The multiple linear regression analysis revealed that initial and peak TFC and peak and mean FIMO and AFIMO predicted outcomes (ventilator days, length of PICU, and hospital days) in critically ill children (p < 0.05). Conclusions: In our cohort of critically ill children with respiratory failure and/or shock, TFC did not correlate with conventional measures of fluid balance (FIMO/AFIMO), but a significant proportion of patients with high TFC had pulmonary plethora on chest x-ray. Both initial and peak TFC predicted outcomes in critically ill children.