Fluid Overload and Mortality in Adult Critical Care Patients-A Systematic Review and Meta-Analysis of Observational Studies.

OBJECTIVE: Fluid administration in combination with the increase in vasopermeability induced by critical illness often results in significant fluid overload in critically ill patients. Recent research indicates that mortality is increased in patients who have received large volumes of fluids. We have systematically reviewed and synthesized the evidence on fluid overload and mortality in critically ill patients and have performed a meta-analysis of available data from observational studies.

DATA SOURCES: A systematic search was performed on PubMed, EmBase, and the Cochrane Library databases.

STUDY SELECTION AND DATA EXTRACTION: All studies were eligible that investigated the impact of fluid overload (defined by weight gain > 5%) or positive cumulative fluid balance on mortality in adult critical care patients. We excluded animal studies and trials in pediatric populations (age < 16 years old), pregnant women, noncritically ill patients, very specific subpopulations of critically ill patients, and on early goal-directed therapy. Randomized controlled trials were only evaluated in the section on systematic review. Assessment followed the Cochrane/meta-analysis of observational trials in epidemiology guidelines for systematic reviews.

DATA SYNTHESIS: A total of 31 observational and three randomized controlled trials including 31,076 ICU patients met the inclusion criteria. Only observational studies were included in the meta-analysis. Fluid overload and cumulative fluid balance were both associated with pooled mortality: after 3 days of ICU stay, adjusted relative risk for fluid overload was 8.83 (95% CI, 4.03-19.33), and for cumulative fluid balance 2.15 (95% CI, 1.51-3.07), at any time point, adjusted relative risk for fluid overload was 2.79 (95% CI, 1.55-5.00) and 1.39 (95% CI, 1.15-1.69) for cumulative fluid balance. Fluid overload was associated with mortality in patients with both acute kidney injury (adjusted relative risk, 2.38; 95% CI, 1.75-2.98) and surgery (adjusted relative risk, 6.17; 95% CI, 4.81-7.97). Cumulative fluid balance was linked to mortality in patients with sepsis (adjusted relative risk, 1.66; 95% CI, 1.39-1.98), acute kidney injury (adjusted relative risk, 2.63; 95% CI, 1.30-5.30), and respiratory failure (adjusted relative risk, 1.19; 95% CI, 1.03-1.43). The risk of mortality increased by a factor of 1.19 (95% CI, 1.11-1.28) per liter increase in positive fluid balance.

CONCLUSIONS: This systematic review and meta-analysis of observational studies reporting adjusted risk estimates suggests that fluid overload and positive cumulative fluid balance are associated with increased mortality in a general population and defined subgroups of critically ill patients.