Skip to main content
. 2019 May 17;19:79. doi: 10.1186/s12871-019-0753-3

Table 2.

Multivariable Cox regression analysis of 30-day mortality after ICU admission according to positive and negative fluctuations in Cl levels within 72 h after ICU admission

Variables Multivariable model P
Hazard ratio (95% CI)
Model 1
aPositive fluctuation in Cl (1 mmol L− 1 increase) 1.08 (1.04, 1.11) < 0.001
bNegative fluctuation in Cl (1 mmol L− 1 increase) 1.03 (0.98, 1.08) 0.313
 Cumulative FB groupc
  Even (0–5%) 1 (< 0.001)
  Negative (< 0%) 0.97 (0.72, 1.32) 0.868
  Positive, Mild to moderate (5–10%) 2.01 (1.57, 2.57) < 0.001
  Positive, Severe (> 10%) 4.06 (3.10, 5.32) < 0.001
 Dyschloremia at ICU admission
  Normochloremia (97–110 mmol L− 1) 1 (< 0.001)
  Hypochloremia (< 97 mmol L− 1) 1.99 (1.57, 2.51) < 0.001
  Hyperchloremia (> 110 mmol L− 1) 0.78 (0.60, 1.00) 0.054
Model 2
aPositive fluctuation of Cl > 10 mmol L− 1 3.37 (2.01, 5.67) < 0.001
bNegative fluctuation of Cl > 10 mmol L− 1 0.94 (0.38, 2.30) 0.883

ICU intensive care unit, FB fluid balance

aPositive fluctuation in Cl: (Maximum Cl – Preadmission Cl) for 72 h after ICU admission

bNegative fluctuation in Cl: (Preadmission Cl – Minimum Cl) for 72 h after ICU admission

cCumulative fluid balance (%): (Total input fluid – total output fluid, L) × 100 x weight at admission (kg)− 1