Diagnostic performance of fractional excretion of urea in the evaluation of critically ill patients with acute kidney injury: a multicenter cohort study

Abstract

Introduction: Several factors, including diuretic use and sepsis, interfere with the fractional excretion of sodium, which is used to distinguish transient from persistent acute kidney injury (AKI). These factors do not affect the fractional excretion of urea (FeUrea). However, there are conflicting data on the diagnostic accuracy of FeUrea.

Methods: We conducted an observational, prospective, multicenter study at three ICUs in university hospitals. Unselected patients, except those with obstructive AKI, were admitted to the participating ICUs during a six-month period. Transient AKI was defined as AKI caused by renal hypoperfusion and reversal within three days. The results are reported as medians (interquartile ranges).

Results: A total of 203 patients were included. According to our definitions, 67 had no AKI, 54 had transient AKI and 82 had persistent AKI. FeUrea was 39% (28 to 40) in the no-AKI group, 41% (29 to 54) in the transient AKI group and 32% (22 to 51) in the persistent AKI group (P = 0.12). FeUrea was of little help in distinguishing transient AKI from persistent AKI, with the area under the receiver operating characteristic curve being 0.59 (95% confidence interval, 0.49 to 0.70; P = 0.06). Sensitivity was 63% and specificity was 54% with a cutoff of 35%. In the subgroup of patients receiving diuretics, the results were similar.

Conclusions: FeUrea may be of little help in distinguishing transient AKI from persistent AKI in critically ill patients, including those receiving diuretic therapy. Additional studies are needed to evaluate alternative markers or strategies to differentiate transient from persistent AKI.

Figure 1

(a) Boxplot of the fractional excretion of sodium (FeNa) in the overall study population according to renal function. The dotted line represents FeNa of 1% (P = 0.04). (b) Boxplot of the fractional excretion of urea (FeUrea) in the overall study population according to renal function. The dotted line represents FeUrea of 35% (P = 0.12). (c) Boxplot of the urine/plasma (U/P) urea ratio in the overall study population according to renal function. The dotted line represents a U/P urea ratio of 10 (P < 0.0001).

Figure 2

Receiver-operating characteristic (ROC) curve depicting the ability of the fractional excretion of urea (FeUrea) and urine/plasma (U/P) urea ratio to detect persistent AKI in the subgroup of patients with AKI. The ROC curve shows the relationship between the proportion of true positives (Sensitivity) and the proportion of false positives (1-Specificity) with various FeUrea and U/P urea ratio cutoffs. Diagonal segments are produced by ties. The area under the ROC curve is 0.59 (95% confidence interval, 0.49 to 0.70; P = 0.06) for FeUrea. The area under the ROC curve is 0.71 (95% confidence interval, 0.62 to 0.80; P = 0.04) for U/P urea ratio.