By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle

Dr. Nadler reports no financial relationships relevant to this field of study.

SYNOPSIS: In critically ill patients, a negative fluid balance on day 3 either spontaneously or with deresuscitative measures may be associated with lower mortality.

SOURCE: Silversides JA, Fitzgerald E, Manickavasagam US, et al. Deresuscitation of patients with iatrogenic fluid overload is associated with reduced mortality in critical illness. Crit Care Med 2018 Jul 5. doi: 10.1097/CCM.0000000000003276. [Epub ahead of print].

Fluid administration to critically ill patients to ensure organ perfusion is widespread and supported by multiple guidelines. However, the ideal regimen, including which fluids and the duration of fluid administration, remains unclear. Excess fluid therapy is associated with pulmonary edema, renal congestion, and other negative physiologic effects. After the need for fluid administration has resolved, the question arises whether deresuscitative efforts such as diuresis or dialysis would be beneficial. The Role of Active Deresuscitation After Resuscitation (RADAR) investigators published a retrospective cohort study of 400 patients ≥ 16 years of age receiving mechanical ventilation for at least 24 hours and examined fluid balance. Exclusion criteria included drug overdose, subarachnoid hemorrhage, diabetic ketoacidosis, and other conditions for which specific fluid management is critical to management. Patients not expected to survive more than 24 hours and ICU transfers also were excluded. Demographic features were collected, and the primary outcome was 30-day mortality. Secondary outcomes included ICU length of stay and duration of mechanical ventilation.

In this cohort, the overall 30-day mortality was 31%, with a median ICU length of stay of seven days. Overall, survivors demonstrated smaller positive fluid balance over the first week, with a maximum difference in means of 0.98 L on day 3 and 2.38 L overall. Within the first three days, 34.5% of fluid was due to medications compared to 26.5% due to maintenance fluids and 24.4% from bolus fluids. In univariate logistic regressions, fluid balance on day 3 was associated independently with 30-day mortality, with an odds ratio (OR) of 1.32 (95% confidence interval [CI], 1.17-1.5 per liter).

This association remained significant in multivariate logistic regression. A similar inverse association was observed between positive fluid balance and organ dysfunction scores, ICU length of stay, and duration of mechanical ventilation.


This study highlights the evolving notion that intravenous fluids are an intervention. Like many medications, these fluids require appropriate dosing and formulation to ensure benefit. The study also highlights the limitations of retrospective cohort studies. Compared with nonsurvivors in this cohort, 30-day survivors were younger (60 vs. 69 years of age; P < 0.01), were more likely to be surgical admissions (33.7% vs. 58.7%), had lower APACHE-II scores (16.8 vs. 22.4; P = 0.01), and had less severe disease overall.

In the multivariate regression model, spontaneous negative fluid balance was associated with decreased mortality (OR, 0.21; 95% CI, 0.08-0.56; P < 0.01), but this effect size was similar to surgery as the admission type (OR, 0.22; 95% CI, 0.09-0.53; P < 0.01). The OR for 30-day mortality in patients who achieved negative fluid balance with deresuscitative measures was 0.29 (95% CI, 0.12-0.69; P < 0.01). Those individuals in whom negative fluid balance was not achieved despite deresuscitative efforts had the highest crude mortality. Thus, patient factors may play a greater role than diuretics or dialysis.

The source of fluid input in this study deserves attention. The largest single input was medications. Over the first three days, the average fluid balance was 3,325 mL. Although volume of medications represented 34.5% of the input, overall, 50.9% represented maintenance and bolus intravenous fluids. Much of the positive fluid balance was “maintenance” fluid that often is started on admission without careful consideration of duration and dose. Furthermore, there was wide practice variation between sites in this study. Median fluid bolus volumes by site ranged from 0.75 L to 3.5 L, while median maintenance fluids ranged from 0 L to 5.77 L.

This study demonstrates the significant practice variability in fluid administration in critically ill patients and the opportunity for practice improvement. It adds to previous studies, such as the FACTT trial, that showed conservative fluid management decreased duration of mechanical ventilation and ICU length of stay.1 A recent meta-analysis of conservative fluid management and deresuscitation in sepsis and ARDS also demonstrated decreased ventilator days and ICU length of stay but no change in mortality.2 Although the current trial found negative fluid balance was associated with decreased mortality, the retrospective nature of the study, along with previous larger negative trials, raises doubt about a causal relationship.


  1. National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network; Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med 2006;354:2564-2575.
  2. Silversides JA, Major E, Ferguson AJ, et al. Conservative fluid management of deresuscitation for patients with sepsis or acute respiratory distress syndrome following the resuscitation phase of critical illness: A systematic review and meta-analysis. Intensive Care Med 2017;43:155-170.