Association of negative fluid balance during the de-escalation phase of sepsis management with mortality: A cohort study

Purpose: We aimed to evaluate the impact of negative fluid balance during the fluid de-escalation phase of sepsis management. Material and methods: This is a historical cohort study of adult intensive care units (ICU) patients with septic shock and severe sepsis in a quaternary medical center, from January 2007 through December 2009. We used regression modeling to assess the impact of negative volume balance on mortality after adjustments for age, comorbidities, and illness severity. Results: Among 633 enrolled patients, 387 patients reached negative fluid balance who in comparison with others had a lower 90-day mortality rate (36% vs. 44%; P = .048), despite higher severity of illness. Each 1-L negative daily fluid balance was associated with reduced ICU, hospital, 90-day and 1-year mortality (hazard ratio [HR] 0.39 [95%CI, 0.28–0.57], 0.76 [95%CI, 0.63–0.94], 0.69 [95%CI, 0.59–0.81], 0.67 [0.58–0.78], respectively; P < .05). This protective effect of negative volume balance was maintained when cumulative ICU fluid balance was utilized. Conclusions: There is not only a significant association between outcomes of patients who were resuscitated for sepsis and achieving negative fluid balance, but also the amount of daily or cumulative negative fluid balance is associated with lower mortality of these patients. Prospective clinical trials are needed to validate this finding.