Diuretic therapy for critically ill patients: a systematic review and network meta-analysis

Fluid overload is common in critically ill patients and is associated with worse outcomes. Diuretics are the mainstay of active fluid removal in patients with preserved renal function. The optimal diuretic strategy for fluid removal remains uncertain.

We conducted a systematic review and Bayesian random-effects network meta-analysis of randomized controlled trials comparing two or more diuretic strategies for fluid removal in critically ill adults. We searched MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials via Ovid, as well as trial registries, from inception to November 20, 2025. Two reviewers independently extracted data and assessed risk of bias using ROBUST-RCT. Certainty of evidence was evaluated using the GRADE approach for network meta-analysis. Treatment effects were summarized as odds ratios (ORs) or mean differences (MDs) with 95% credible intervals (CrIs).

Twenty-six randomized controlled trials involving 1,652 participants were included. Evaluated interventions included bolus loop diuretics (19 studies), continuous loop infusion (15 studies), oral loop diuretics (5 studies), and loop diuretics combined with tolvaptan (8 studies), spironolactone (3 studies), thiazides (2 studies), acetazolamide (1 study), or triamterene (1 study). Compared with bolus loop diuretics, continuous loop infusion had an uncertain effect on mortality (OR 1.26; 95% CrI 0.62 to 2.55; very low certainty) and may increase ICU length of stay (MD 1.56 days; 95% CrI -0.02 to 3.16; low certainty). Tolvaptan monotherapy may reduce acute kidney injury compared with bolus or continuous loop diuretics (OR 0.12; 95% CrI 0.01 to 0.87; low certainty), although no studies evaluated its effect on the need for renal replacement therapy. For most other comparisons and outcomes, the certainty of evidence was low or very low.

Available evidence comparing diuretic strategies for fluid removal during ICU-level care is limited, clinically heterogeneous, and derived largely from heart failure and post-cardiovascular surgery populations. Bolus loop diuretics were at least comparable to alternate diuretic strategies for patient important outcomes in critically ill adults, but this was based mostly on low or very low certainty evidence. Tolvaptan monotherapy may decrease acute kidney injury, but its effect on subsequent need for RRT and mortality remains uncertain.