The objective of this study ( ClinicalTrials.gov: NCT01351506) was to identify the threshold level of fluid accumulation measured by acute body weight (BW) change during the first week in a general surgical intensive care unit (ICU), which is associated with ICU mortality and other adverse outcomes.
Four hundred sixty-five patients were prospectively followed for a 28-day period. The maximum BW change threshold during the first week was evaluated by the maximum percentage change in BW from the ICU admission weight (Max%ΔBW). Daily screening of adverse events in the ICU were recorded. The cutoff point of Max%ΔBW on ICU mortality was defined by considering the area under the receiver operating characteristic (ROC) curve, intersection of the sensitivity and specificity, and the Youden Index. Univariable and multivariable regression analyses were used to demonstrate the associations. Statistical significance was defined as P<0.05.
The appropriate cutoff value of Max%ΔBW threshold was 5%. Regarding the multivariable regression model, in overall patients, the occurrence of the following adverse events (expressed as adjusted odds ratio [95% confidence interval]) were significantly associated with a Max%ΔBW of >5%: ICU mortality (2.38 [1.25–4.54]) ( P=0.008), ICU mortality in patients without renal replacement therapy (RRT) (2.47 [1.21–5.06]) ( P=0.013), reintubation within 72 hours (2.51 [1.04–6.00]) ( P=0.039), RRT requirement (2.67 [1.13–6.33]) ( P=0.026), and delirium (1.97 [1.08–3.57]) ( P=0.025). Regarding the postoperative subgroup, a Max%ΔBW value of more than 5% was significantly associated with: ICU mortality (3.87 [1.38–10.85]) ( P=0.010), ICU mortality in patients without RRT (6.32 [1.85–21.64]) ( P=0.003), reintubation within 72 hours (4.44 [1.30–15.16]) ( P=0.017), and vasopressor requirement (2.04 [1.04–4.01]) ( P=0.037).
Fluid accumulation, measured as acute BW change of more than the threshold of 5% during the first week of ICU admission, is associated with adverse outcomes of higher ICU mortality, especially in the patients without RRT, with reintubation within 72 hours, with RRT requirement, with vasopressor requirement, and with delirium. Some of these effects were higher in postoperative patients. This threshold value might be an indicator for caution during fluid management in surgical ICU.