Fluid balance and cardiac function in septic shock as predictors of hospital mortality

Abstract

Introduction: Septic shock is a major cause of morbidity and mortality throughout the world. Unfortunately, the optimal fluid management of septic shock is unknown and currently is empirical.

Methods: A retrospective analysis was performed at Barnes-Jewish Hospital (St. Louis, Missouri). Consecutive patients (n = 325) hospitalized with septic shock who had echocardiographic examinations performed within 24 hours of shock onset were enrolled.

Results: A total of 163 (50.2%) patients with septic shock died during hospitalization. Non-survivors had a significantly larger positive net fluid balance within the 24 hour window of septic shock onset (median (IQR): 4,374 ml (1,637 ml, 7,260 ml) vs. 2,959 ml (1,639.5 ml, 4,769.5 ml), P = 0.004). The greatest quartile of positive net fluid balance at 24 hours and eight days post-shock onset respectively were found to predict hospital mortality, and the greatest quartile of positive net fluid balance at eight days post-shock onset was an independent predictor of hospital mortality (adjusted odds ratio (AOR), 1.66; 95% CI, 1.39 to 1.98; P = 0.004). Survivors were significantly more likely to have mild left ventricular dysfunction as evaluated by bedside echocardiography and non-survivors had slightly elevated left ventricular ejection fraction, which was also found to be an independent predictor of outcome.

Conclusions: Our data confirms the importance of fluid balance and cardiac function as outcome predictors in patients with septic shock. A clinical trial to determine the optimal administration of intravenous fluids to patients with septic shock is needed.

Figure 1

Box plots depicting daily cumulative fluid balance for survivors (white boxes) and non-surviviors (hatched boxes). The lines within the boxes represented the 50^th percentile, the lines at the bottom and top of the boxes represent the 25^th and 75^th percentiles, and the whisker lines represent the 5^th and 95^th percentiles.

Figure 2

Cox survival curves, adjusted for age, APACHE II scores, and the use of vasopressin are shown for fluid balance quartiles, 24 hours (Top) and at Day 8 (Bottom). Quartile 4 has significantly decreased survival risk compared to quartiles 1 and 2 at 24 hours and 8 days, respectively.

Figure 3

Box plots depicting cumulative fluid balance at 24 hours (Top) and 8 days (Bottom) following shock onset for patients with and without cardiac dysfunction demonstrated by echocardiographic examinations. LV, left ventricle; RV, right ventricle.