To determine if the timing of excess fluid accumulation (fluid overload) is associated with adverse patient outcomes.
Secondary analysis of a prospectively collected dataset.
PICU of a tertiary care hospital.
Children 3 months to 25 years old admitted to the PICU with expected length of stay greater than or equal to 48 hours.
Patients were dichotomized by time of peak overload: peak fluid overload from ICU admission (Day0) to 48 hours (Day3–7) and peak fluid overload value after 48 hours of ICU admission, as well as time of first-time negative daily fluid balance: net fluid out greater than net fluid in for that 24-hour period.
Measurements and Main Results:
There were 177 patients who met inclusion criteria, 92 (52%) male, with an overall mortality rate of 7% (n = 12). There were no differences in severity of illness scores or fluid overload on Day0 between peak fluid overload from ICU admission (Day0) to 48 hours (Day3–7) (n = 97; 55%) and peak fluid overload value after 48 hours of ICU admission (n = 80; 45%) groups. Peak fluid overload value after 48 hours of ICU admission was associated with a longer median ICU course (8 [4–15] vs 4 d [3–8 d]; p ≤ 0.001], hospital length of stay (18 [10–38) vs 12 [8–24]; p = 0.01], and increased risk of mortality (n = 10 [13%] vs 2 [2%]; χ2 = 7.6; p = 0.006]. ICU length of stay was also longer in the peak fluid overload value after 48 hours of ICU admission group when only patients with at least 7 days of ICU stay were analyzed (p = 0.02). Timing of negative fluid balance was also correlated with outcome. Compared with Day0–2, a negative daily fluid balance on Day3–7 was associated with increased length of mechanical ventilation (3 [1–7] vs 1 d [2–10 d]; p ≤ 0.001) and increased hospital (17 [10–35] vs 11 d [7–26 d]; p = 0.006) and ICU (7 [4–13] vs 4 d [3–7 d]; p ≤ 0.001) length of stay compared with a negative fluid balance between Day0–2.
Our results show timing of fluid accumulation not just peak percentage accumulated is associated with patient outcome. Further exploration of the association between time and fluid accumulation is warranted.