Impact of loop diuretics on critically ill patients with a positive fluid balance.

The impact of the use of loop diuretics to prevent cumulative fluid balance in non-oliguric patients is uncertain. This is a retrospective study to estimate the association of time-averaging loop diuretic exposure in a large population of non-cardiac, critically ill patients with a positive fluid balance (> 5% of body weight). The exposure was loop diuretic and the main outcomes were 28-day mortality, severe acute kidney injury and successful mechanical ventilation weaning. Time-fixed and daily time-varying variables were evaluated with a marginal structural Cox model, adjusting bias for time-varying exposure and the presence of time-dependent confounders. A total of 14,896 patients were included. Patients receiving loop diuretics had better survival (unadjusted hazard ratio 0.56, 95%CI 0.39-0.81 and baseline variables adjusted hazard ratio 0.53, 95%CI 0.45-0.62); after full adjusting, loop diuretics had no association with 28-day mortality (full adjusted hazard ratio 1.07, 95%CI 0.74-1.54) or with reducing severe acute kidney injury occurrence during intensive care unit stay - hazard ratio 1.05 (95%CI 0.78-1.42). However, we identified an association with prolonged mechanical ventilation (hazard ratio 1.59, 95%CI 1.35-1.89). The main results were consistent in the sub-group analysis for sepsis, oliguria and the study period (2002-2007 vs. 2008-2012). Also, equivalent doses of up to 80 mg per day of furosemide had no significant association with mortality. After adjusting for time-varying variables, the time average of loop diuretic exposure in non-cardiac, critically ill patients has no association with overall mortality or severe acute kidney injury; however, prolonged mechanical ventilation is a concern.