Diuretic Response in the ICU: A Practical Predictor of Renal Trajectory
Kidney replacement therapy (KRT) remains a high-stakes intervention in intensive care, yet predicting which patients will require it remains imprecise.
A new retrospective cohort study published in Journal of Critical Care by Gal Oz and colleagues evaluates the prognostic value of urine output following a first furosemide bolus in a general ICU population—a real-world test of the Furosemide Stress Test (FST) principle, but applied outside controlled protocols.
Study Designed Around Routine ICU Care
The study reviewed 1,527 adult patients treated with a furosemide bolus between 2017 and 2023 at a tertiary ICU. Patients on furosemide infusions or already receiving KRT were excluded. Researchers stratified urine output at 2 and 6 hours post-bolus into three categories: ≤25 mL/h, 25.1–50 mL/h, and >50 mL/h.
The primary endpoint was KRT initiation within 7 days. Secondary outcomes included progression in AKI severity by KDIGO staging and a combined outcome of KRT or death within 7 days.
Urine Output After Furosemide: A Clear Signal
Urine output at both 2 and 6 hours was significantly associated with downstream clinical outcomes (p < .001 for both):
- Urine output at 2 hours: KRT was initiated in 23.9% of patients with ≤25 mL/h output, compared to 8.8% in the mid-range group and 3.6% in those exceeding 50 mL/h.
- Urine output at 6 hours: KRT was initiated in 28.6%, 12.0%, and 3.5%, respectively, for the same categories.
- Odds ratios for KRT were as high as 10.91 for the lowest urine output group compared to the highest at 6 hours.
Multivariable analysis confirmed that every 10 mL/h increase in urine output at 6 hours reduced the odds of requiring KRT (adjusted OR: 0.907; 95% CI: 0.863–0.954). Importantly, baseline urine output before furosemide had no predictive value.
Why These Results Matter
Unlike prior studies that applied the FST only to patients with established AKI, this study included a broader ICU population and used naturally occurring furosemide dosing decisions. The real-world design also enhances applicability: furosemide was given as part of routine fluid management, with median doses reflecting usual care (20 mg IV). Still, the early urine response retained strong discriminatory power for predicting KRT need and AKI progression, even in patients without confirmed AKI at baseline.
Additionally, the findings support urine output response as a dynamic biomarker of renal functional reserve. Patients with poor response to furosemide had higher rates of KDIGO stage progression and ICU mortality. These associations persisted in subgroups receiving higher doses, reinforcing the robustness of the prognostic signal across dosing ranges.
Study Limitations and Interpretive Caution
However, it’s important to note that as a retrospective, single-center study, generalizability may be limited. The indication for furosemide administration—typically presumed fluid overload—was not uniformly recorded.
Another limitation was that the precise phase of critical illness (e.g., resuscitation vs. de-resuscitation) was not captured. Variability in diuretic dosing and potential unmeasured confounders also temper the strength of the conclusions.
Looking Forward
Despite these limitations, this study suggests that diuretic responsiveness—specifically urine output in the first hours post-furosemide—may offer a simple, real-time indicator to guide fluid management and anticipate renal deterioration in the ICU. While not a substitute for established markers, this functional metric could refine risk stratification, especially in ambiguous clinical scenarios.
Further validation in prospective, multicenter cohorts and integration with clinical decision tools could advance its utility from a predictive signal to a decision-support asset in ICU nephrology.
Reference:
Gal Oz A, Wald R, Stavi D, et al. Response to furosemide and the receipt of kidney replacement therapy in critically ill patients. J Crit Care. 2025;90:155171. doi:10.1016/j.jcrc.2025.155171