FENa vs FEUrea: which to use when the patient is on diuretics

Why diuretics throw off the FENa

The FENa relies on the tubules reabsorbing almost all filtered sodium when the kidney is underperfused, which is why Espinel found values below 1% in prerenal azotemia in 1976. Diuretics break that link. They block sodium reabsorption in the tubule directly, so sodium appears in the urine whatever the patient's volume status, and a prerenal patient can have a FENa well above 1%. Carvounis and colleagues measured how large the effect is in their 2002 Kidney International study of 102 episodes of acute renal failure. Among 50 patients with prerenal azotemia who had not received diuretics, 92% had a FENa below 1%. Among 27 patients with prerenal azotemia who had received diuretics, only 48% did. Gotfried and colleagues note in their 2012 review that this is common in patients with heart failure and cirrhosis, who often take diuretics while their effective circulating volume is low, which is exactly when the question of prerenal injury comes up.

How the fractional excretion of urea works

The fractional excretion of urea, or FEUrea, uses the same structure as the FENa with urea in place of sodium: urine urea times plasma creatinine, divided by plasma urea times urine creatinine, times 100. Kaplan and Kohn proposed it in the American Journal of Nephrology in 1992, because prior diuretics hamper the FENa while urea excretion still follows volume status. Urea is reabsorbed by facilitated diffusion through urea transporters in the proximal tubule, loop of Henle and medullary collecting duct; when volume is low, antidiuretic hormone rises and less urea is excreted. Carvounis noted that this depends mainly on passive forces and so is less influenced by diuretics. A FEUrea of 35% or less suggests a prerenal cause, the cut-off used by Kaplan and Kohn and by Carvounis. Gotfried's review adds that a FEUrea above 50% suggests an intrinsic cause, with values in between indeterminate. Urea must be in the same unit in urine and plasma, either as urea nitrogen in mg/dL or as urea in mmol/L.

Gotfried's review lists two limits of FEUrea itself. It is less effective in patients with infection, because cytokines interfere with the urea transporters, and both FENa and FEUrea rise with age, which makes the standard cut-offs less reliable in older people.

What the FEUrea studies found

The early FEUrea studies were encouraging, but the later ones were not. Kaplan and Kohn studied 6 patients prospectively and 87 retrospectively and found a low FEUrea a sensitive index of reduced kidney perfusion despite prior diuretics; in Gotfried's summary, 39 of 40 patients given diuretics had a high FENa while the FEUrea was low in all of them. Carvounis then compared the two prospectively. The mean FEUrea was 27.9% in prerenal azotemia without diuretics and 24.5% with diuretics, against 58.6% in acute tubular necrosis, and 89% of the prerenal patients on diuretics had a FEUrea of 35% or less. The authors concluded that a low FEUrea was more sensitive and specific than the FENa, especially after diuretics. Pépin and colleagues reached the opposite conclusion in 2007 in 99 patients: on diuretics, FEUrea had a sensitivity of 79% but a specificity of only 33%, against 58% and 81% for the FENa.

The studies differ in design. Carvounis studied intensive care patients and grouped them by the attending nephrologist's diagnosis, while Pépin classified transient and persistent acute kidney injury by whether creatinine returned to baseline within 7 days, and exposure to diuretics varied, so a long time may have passed between the diuretic and the urine sample.

What the meta-analyses say about FENa vs FEUrea

Three meta-analyses now frame the FENa vs FEUrea question, and none shows FEUrea as a clear replacement. Abdelhafez and colleagues pooled 19 studies of the FENa in CJASN in 2022: at a 1% cut-off it had a sensitivity of 90% and a specificity of 82% for intrinsic injury, rising to 95% and 91% in oliguric patients without chronic kidney disease or diuretics, but falling to a specificity of 54% in patients on diuretics. The same group pooled 11 FEUrea studies with 1,108 patients in 2024. In four studies of 244 patients on diuretics, FEUrea had a lower sensitivity than the FENa for intrinsic injury, 52% against 92%, and a higher specificity, 82% against 44%. They concluded that FEUrea is not a superior alternative even with diuretics. Pan and colleagues pooled 12 studies with 1,240 patients in BMJ Open in 2026 and found a sensitivity of 74% and a specificity of 78%, with specificity rising to 87% in the diuretic-only studies, at low certainty.

Read together, the numbers suggest the two indices err in opposite directions on diuretics. A high FENa catches most intrinsic injury but also flags many prerenal patients, while a high FEUrea is more specific for intrinsic injury but misses about half of it. Neither replaces the history, the urine sediment and the course of the creatinine.

Worked examples: reading FENa and FEUrea together

Two illustrative patients show how the FENa and FEUrea can agree or disagree. The first has heart failure, took a diuretic the morning of the sample, and has a rising creatinine. Urine sodium is 80 mmol/L, plasma sodium 136 mmol/L, urine creatinine 60 mg/dL and plasma creatinine 2.4 mg/dL, so the FENa is (80 × 2.4) ÷ (136 × 60) × 100 = 192 ÷ 8,160 × 100 = 2.35%, in the indeterminate band and above the 1% prerenal cut-off. With urine urea nitrogen 280 mg/dL and blood urea nitrogen 56 mg/dL, the FEUrea is (280 × 2.4) ÷ (56 × 60) × 100 = 672 ÷ 3,360 × 100 = 20%, well under 35%, which supports a prerenal cause despite the FENa. The second patient has had prolonged hypotension, took no diuretic, and has muddy brown casts on urine microscopy. Here both indices point the same way, towards acute tubular necrosis.

For the second patient, (70 × 4.2) ÷ (138 × 35) × 100 = 294 ÷ 4,830 × 100 = 6.09%, above 3%, and (420 × 4.2) ÷ (70 × 35) × 100 = 1,764 ÷ 2,450 × 100 = 72%, above 50%. Enter both patients' sodium and creatinine values in the FENa calculator to check the first column.