Fluid Volume, Lactate Values, and Mortality in Sepsis Patients with Intermediate Lactate Values

RATIONALE: Patients with severe sepsis without shock or tissue hypoperfusion face substantial mortality; however, treatment guidelines are lacking. OBJECTIVES: To evaluate the association between intravenous fluid resuscitation, lactate clearance, and mortality in patients with "intermediate" lactate values of 2 mmol/L or greater and less than 4 mmol/L. MEASUREMENTS AND MAIN RESULTS: This was a retrospective study of 9,190 patients with sepsis with intermediate lactate values. Interval changes between index lactate values and those at 4, 8, and 12 hours were calculated with corresponding weight-based fluid volumes. Outcomes included lactate change and mortality. Repeat lactate tests were completed in 94.7% of patients within 12 hours. Hospital and 30-day mortality were 8.2 and 13.3%, respectively, for patients with lactate clearance; they were 18.7 and 24.7%, respectively, for those without lactate clearance. Each 10% increase in repeat lactate values was associated with a 9.4% (95% confidence interval [CI] = 7.8-11.1%) increase in the odds of hospital death. Within 4 hours, patients received 32 (± 18) ml/kg of fluid. Each 7.5 ml/kg increase was associated with a 1.3% (95% CI = 0.6-2.1%) decrease in repeat lactate. Across an unrestricted range, increased fluid was not associated with improved mortality. However, when limited to less than 45 ml/kg, additional fluid was associated with a trend toward improved survival (odds ratio = 0.92; 95% CI = 0.82-1.03) that was statistically significant among patients with highly concordant fluid records. CONCLUSIONS: Early fluid administration, below 45 ml/kg, was associated with modest improvements in lactate clearance and potential improvements in mortality. Further study is needed to define treatment strategies in this prevalent and morbid group of patients with sepsis.