Fractional excretion of sodium (FENa)

Clinical background

In acute kidney injury, management differs fundamentally according to whether the cause is prerenal — that is, impaired perfusion with intact tubular function — or intrinsic, that is, injury to the renal parenchyma. Prerenal azotaemia requires correction of volume status and perfusion pressure, whereas intrinsic injury, above all acute tubular necrosis, calls for a restrictive fluid balance and treatment of the underlying cause. Confusing these two conditions has direct clinical consequences: inappropriate fluid administration in acute tubular necrosis worsens pulmonary oedema and mortality, and fluid restriction in prerenal azotaemia prolongs the ischaemia and can drive the patient into irreversible kidney injury. The fractional excretion of sodium (FENa) was developed precisely to distinguish these two mechanisms quickly and simply in oliguric acute kidney injury, using laboratory values available in serum and urine [1].

Calculating the fractional excretion of sodium (FENa)

FENa expresses the proportion of filtered sodium that is excreted in the urine, and is calculated as:

$$\text{FENa} = \frac{\text{Urine sodium} \times \text{Serum creatinine}}{\text{Serum sodium} \times \text{Urine creatinine}} \times 100$$

where urine and serum sodium are given in mEq/L and urine and serum creatinine in mg/dL. The value is dimensionless and expressed as a percentage. Mathematically, FENa divides sodium clearance by creatinine clearance and thereby captures how efficiently the tubules reabsorb sodium. In prerenal azotaemia the kidneys respond to hypoperfusion with increased sodium retention, giving a low FENa. In acute tubular necrosis the tubules lose the ability to reabsorb sodium, giving a high FENa.

The derivation cohort consisted of patients in the oliguric phase of acute kidney injury, described by Espinel in 1976 [1]. Patients with prerenal azotaemia had a FENa below 1 per cent, and patients with acute tubular necrosis a FENa above 3 per cent. The difference was statistically significant. The original study therefore defined the range between 1 and 3 per cent as an indeterminate zone, but later systematic reviews have mainly used 1 per cent as the single threshold.

Interpretation in practice

FENa is interpreted according to where the value falls relative to the 1 per cent threshold, with an indeterminate zone that clinicians should be alert to:

The original study used >3 per cent as the cut-off for acute tubular necrosis [1], and some textbooks retain this higher threshold. A systematic review and meta-analysis from 2022 included 15 studies that all used 1 per cent as the threshold and found that this gave an acceptable balance between sensitivity and specificity [2].

A low FENa should prompt assessment of volume status, review of blood pressure and medication, and, where indicated, volume expansion. A high FENa should prompt restrictive fluid management, investigation of the cause of the tubular injury and regular monitoring of fluid balance and electrolytes. A value in the indeterminate zone (1–2 per cent) should not be interpreted in isolation but placed in clinical context, particularly against the background of ongoing diuretic treatment or chronic kidney disease.

Validation and performance

The largest systematic review and meta-analysis of FENa to date included 19 studies with a total of 1,287 patients [2]. At the 1 per cent threshold, the pooled sensitivity was 90 per cent (95 per cent CI 81 to 95) and the pooled specificity 82 per cent (95 per cent CI 70 to 90) for distinguishing intrinsic from prerenal acute kidney injury. In a subgroup of eight studies with 264 oliguric patients without chronic kidney disease or diuretic treatment, sensitivity rose to 95 per cent (95 per cent CI 82 to 99) and specificity to 91 per cent (95 per cent CI 83 to 95), confirming that FENa performs best precisely in the population for which the test was originally developed.

Performance falls in other populations, however. In patients with chronic kidney disease or on diuretics, the pooled sensitivity was 83 per cent and the specificity 66 per cent. In a subgroup of five studies with 238 patients on diuretics, specificity fell to 54 per cent (95 per cent CI 31 to 75) [2], which means that a high FENa in a patient on diuretics may to a considerable extent be a false positive.

A retrospective study from a German university hospital with 431 patients on a nephrology ward found that a FENa below 1 per cent had a sensitivity of 91.4 per cent but a specificity of only 36.1 per cent for a prerenal cause [3]. In this cohort a FENa above 1 per cent was not significantly correlated with intrinsic acute kidney injury, and the fractional excretion of urea performed worse than all the other parameters. The study included a complex patient population with a high proportion of chronic kidney disease and multimorbidity, which probably explains the low specificity and underlines that the performance of the test is population-dependent.

Limitations

FENa has several known weaknesses that limit its usefulness. The most important are:

Diuretics. Loop diuretics inhibit sodium reabsorption in the thick ascending limb of the loop of Henle and thereby increase urine sodium and FENa whatever the cause of the kidney injury. In the meta-analysis, specificity fell from 82 to 54 per cent when patients on diuretics were included [2]. During diuretic treatment the fractional excretion of urea (FEurea) should be used instead, since urea is reabsorbed passively and is less affected by loop diuretics. A meta-analysis of FEurea with 12 studies and 1,240 patients found a pooled sensitivity of 0.74 and a specificity of 0.78, with a summary area under the SROC curve of 0.83 [4]. In a diuretic subgroup, specificity rose to 0.87 and heterogeneity fell, which supports FEurea over FENa when diuretics are involved. The traditional threshold for FEurea is <35 per cent for a prerenal cause. The same German study that showed low specificity for FENa nonetheless found that FEurea had the lowest sensitivity and specificity of all the parameters tested [3], showing that no index is faultless.

Chronic kidney disease. In chronic kidney disease the ability to retain sodium is already impaired and FENa may be raised whatever the current cause of the acute kidney injury. FENa should therefore be interpreted with caution when renal function is already impaired.

Conditions in which a low FENa is misleading. A low FENa occurs not only in prerenal azotaemia. Contrast-induced nephropathy early in its course, early urinary tract obstruction, some forms of glomerulonephritis and the hepatorenal syndrome can all give a FENa below 1 per cent. In the hepatorenal syndrome, sodium retention is extreme but the mechanism is not prerenal in the classical sense, and volume expansion does not help.

Non-oliguric acute kidney injury. FENa was derived specifically for patients in the oliguric phase [1]. In non-oliguric patients the urine flow is higher and the sodium concentration is diluted, which reduces the discriminatory power of the test. The meta-analysis showed that performance was considerably better among oliguric patients without chronic kidney disease and without diuretics [2].

References

1. Espinel CH. The FENa test. Use in the differential diagnosis of acute renal failure. JAMA. 1976. PMID: 947239
2. Abdelhafez M et al. Diagnostic Performance of Fractional Excretion of Sodium for the Differential Diagnosis of Acute Kidney Injury: A Systematic Review and Meta-Analysis. Clin J Am Soc Nephrol. 2022. PMID: 35545442
3. Buckenmayer A et al. Evaluation of simple diagnostic parameters in acute kidney injury in hospitalized patients, diagnostic recommendations for non-nephrologists. Intern Emerg Med. 2023. PMID: 37452960
4. Pan HC et al. Assessing the utility of fractional excretion of urea in distinguishing intrinsic and prerenal acute kidney injury in hospitalised patients: a systematic review and meta-analysis. BMJ Open. 2026. PMID: 41535079