Clinical background
In acute kidney injury, the central diagnostic step is to distinguish a prerenal cause (impaired perfusion with preserved tubular function) from intrinsic renal injury (chiefly acute tubular necrosis). The distinction has treatment consequences: prerenal azotaemia responds to volume expansion, whereas excessive fluid administration in intrinsic injury can worsen oedema and outcome. Clinical signs, history and urinalysis are often insufficient to make the distinction reliably, particularly in the emergency department, where time and information about the baseline creatinine may be lacking.
The urea-to-creatinine ratio was developed precisely to fill this gap. The idea rests on the fact that urea, unlike creatinine, is reabsorbed in the proximal tubule, and that this reabsorption increases in hypovolaemia through antidiuretic hormone. In prerenal azotaemia the urea should therefore rise faster than the creatinine and the ratio be raised. The concept was introduced by Fishberg in 1939 and has since featured in textbooks of internal medicine, nephrology and intensive care [1].
The problem is that the evidence for the diagnostic performance of the ratio is thin and that modern research actively questions its usefulness for the intended purpose.
Calculating the urea-to-creatinine ratio
If urea is reported in mmol/L, which is standard in Swedish laboratory practice, it is first converted to mg/dL:
The conventional interpretation rests on a threshold of 20 (in mg/dL units). A ratio above 20 has traditionally been interpreted as a sign of prerenal azotaemia, and a ratio below 20 as a sign of intrinsic renal injury. In SI units (urea in mmol/L, creatinine in µmol/L) this corresponds to a threshold of approximately 100 [1].
The derivation has no formal development cohort in the modern sense. The ratio derives from physiological observations from the late 1930s and has never been systematically validated in a dedicated derivation study with a defined population and outcome measure [1].
Interpretation in practice
| Ratio (mg/dL) | Traditional interpretation | Clinical action |
|---|---|---|
| > 20 | Prerenal azotaemia | Consider volume expansion, investigate and treat the underlying cause of hypoperfusion |
| ≤ 20 | Intrinsic renal injury | Avoid excessive fluid administration, consider nephrology referral, investigate with urinalysis and possibly biopsy |
This is the interpretation that textbooks traditionally present, and the one on which the calculator is based. It should, however, be applied with great caution, since the diagnostic performance is poorly supported (see the next section).
A ratio above 20 may also arise for reasons unrelated to renal perfusion. Upper gastrointestinal bleeding increases urea production through the breakdown of absorbed blood protein. A high protein intake, catabolism from fever, trauma, infection or thyrotoxicosis, and drugs such as corticosteroids and tetracyclines, all increase protein turnover and hence the urea level independently of renal perfusion [1]. Conversely, a low protein intake, liver disease and rhabdomyolysis give low ratios that may mask a prerenal component [1].
Validation and performance
The largest study specifically to examine the diagnostic performance of the ratio was conducted in the emergency department of Nantes University Hospital [1]. Over 13 months, 1,103 adult patients with a plasma creatinine above 133 µmol/L (1.5 mg/dL) on arrival were included. Prerenal AKI was defined as a creatinine returning to at most 110% of baseline within 7 days, and intrinsic AKI as failure to meet this criterion. The distribution was 45% prerenal and 55% intrinsic AKI.
The result was unequivocally negative. The mean ratio was 90.6 and 91.3 (in mmol/L units) in the prerenal and intrinsic groups respectively, with no statistical difference (p = 0.758). ROC analysis gave an AUC of exactly 0.5, meaning that the ratio had no discriminatory ability whatsoever. At the conventional threshold of 100, sensitivity was 70.1% but specificity only 32.6% [1].
A retrospective study from the Austin Hospital in Melbourne included 20,126 hospital patients, 3,641 of whom had AKI by the RIFLE criteria [2]. The distribution of the ratio showed no bimodality, which argues against its separating two distinct conditions. Patients with a ratio above 20 had, paradoxically, higher in-hospital mortality than those with a ratio below 20 (29.9% versus 18.4%), even after adjustment for confounders in multivariable analysis (OR 5.73 versus 3.32). The authors found a J-shaped curve with the lowest mortality at a ratio of 15 to 20 and concluded that the ratio cannot be used to distinguish prerenal from intrinsic AKI diagnostically, but may have prognostic value as a marker of illness severity [2].
Rachoin et al. studied two cohorts of critically ill patients, a derivation cohort of 1,010 patients and a validation cohort of 10,228 patients [3]. In both cohorts a ratio above 20 was associated with increased mortality and a lower likelihood of starting dialysis. The authors suggest that the lower rate of dialysis may in part be because clinicians misinterpret a high ratio as prerenal and are therefore reluctant to escalate treatment. They explicitly recommend that the ratio not be used to classify AKI in critically ill patients [3].
Limitations
The fundamental weakness of the ratio is that urea production is influenced by many factors other than renal perfusion. Gastrointestinal bleeding, protein intake, catabolism and corticosteroids increase urea independently of renal function. Liver disease and malnutrition lower urea and may conceal a prerenal cause. Rhabdomyolysis lowers the ratio because creatinine rises rapidly from muscle breakdown while urea may be normal [1].
The ratio presupposes a dichotomy between prerenal and intrinsic AKI that probably rarely exists in pure form. Clinical AKI is rather a spectrum in which functional and structural injury coexist in varying proportions and shift over time [1]. A single measurement on arrival therefore cannot capture this dynamic.
In critically ill patients the ratio is particularly unreliable, since BUN correlates with age and illness severity rather than with renal perfusion [3]. In patients with reduced muscle mass, common in chronic illness and advanced age, creatinine rises more slowly for a given fall in GFR, which can artificially raise the ratio independently of the pathophysiology [2].
The 2012 KDIGO guidelines do not recommend the urea-to-creatinine ratio as a diagnostic tool for classifying the type of AKI [4]. The guideline emphasises instead clinical assessment, urinalysis and, where needed, urinary markers such as the fractional excretion of urea, which has better evidence than the urea-to-creatinine ratio, particularly during diuretic treatment.
References
- Manoeuvrier G, Bach-Ngohou K, Batard E, et al. Diagnostic performance of serum blood urea nitrogen to creatinine ratio for distinguishing prerenal from intrinsic acute kidney injury in the emergency department. BMC Nephrol 2017. PMID: 28545421
- Uchino S, Bellomo R, Goldsmith D. The meaning of the blood urea nitrogen/creatinine ratio in acute kidney injury. Clin Kidney J 2012. PMID: 29497527
- Rachoin JS, Daher R, Moussallem C, et al. The fallacy of the BUN:creatinine ratio in critically ill patients. Nephrol Dial Transplant 2012. PMID: 22207331
- Kellum JA, Lameire N, KDIGO AKI Guideline Work Group. Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (Part 1). Crit Care 2013. PMID: 23394211
