RESUMO
Acute kidney injury (AKI) is a frequent, serious complication in critically ill patients. Even if renal replacement therapy is rapidly initiated, AKI may lead to the acute accumulation of metabolic waste products called uremic toxins (UTs). Although the accumulation and effects of UTs have been extensively described in the setting of chronic kidney disease (CKD), few data are available for AKI. A rapid, sensitive, specific method with simple sample preparation is required to facilitate routine blood monitoring of UTs in a context of acute accumulation. We have developed and validated two fast liquid chromatography tandem mass spectrometry methods for the quantification of seven UTs in human serum. The first method (in negative ionization mode) enables the quantification of five UTs (hippuric acid (HA), indoxyl sulfate (IxS), para-cresyl sulfate (pCS), para-cresyl glucuronide (pCG), 3-carboxy-4-methyl-5-propyl-2-furanpropionate (CMPF)). The second method (in positive ionization mode) enables the quantification of two UTs (indole-3-acetic acid (IAA), and trimethylamine N-oxide (TMAO)). Sample preparation consisted of the deproteinization of a small volume of serum (50 µL). The run-times required to assay all the UTs in negative and positive ionization modes were only 2.5 and 2 min, respectively. In order to obtain a reliable, toxin-free matrix for the preparation of calibration standards and quality controls, serum was pretreated with activated charcoal. We used these methods to determine the time course of UT accumulation in eight patients who developed an AKI after cardiac surgery. The calibration curves ranged from 0.1 to 100 µg mL-1 for all the UTs (except for IAA: 0.5 to 100 µg mL-1), and the correlation coefficients were above 0.999 for all. The methods were reproducible, repeatable, and accurate, with all coefficients of variation and biases below 15%. The highest concentrations measured in patients with AKI were lower than those reported in CKD stages 4 and 5 but higher than those observed in patients with no impairment of renal function (particularly for IxS and pCS). Our results also highlighted low accumulation of the other toxins (IAA, HA, TMAO, pCG, and CMPF). The UT concentrations did not rise earlier than that of creatinine; although the return to baseline took longer than for creatinine for some compounds. Lastly, assessment of the time course of UT accumulation as a prognostic marker for AKI (particularly for pCS and IxS) appears to be promising and should be continued in a larger number of patients.