Assessment of kidney proximal tubular secretion in critical illness.

BACKGROUND Serum creatinine concentrations (SCr) are used to determine the presence and severity of acute kidney injury (AKI). SCr is primarily eliminated by glomerular filtration; however, most mechanisms of AKI in critical illness involve kidney proximal tubules, where tubular secretion occurs. Proximal tubular secretory clearance is not currently estimated in the ICU. Our objective was to estimate the kidney clearance of secretory solutes in critically ill adults. METHODS We collected matched blood and spot urine samples from 170 ICU patients and from a comparison group of 70 adults with normal kidney function. We measured seven endogenously produced secretory solutes using liquid chromatography-tandem mass spectrometry. We computed a composite secretion score incorporating all seven solutes, and evaluated associations with 28-day major adverse kidney events (MAKE28), defined as doubling of SCr, dialysis dependence, or death. RESULTS The urine/plasma ratio of six of seven secretory solutes were lower in critically ill patients compared with normal individuals after adjustment for SCr. The composite secretion score was moderately correlated with SCr and cystatin C (r = -0.51 and r = -0.53, respectively). Each standard deviation higher composite secretion score was associated with a 25% lower risk of MAKE28 (95% CI 9% - 38% lower) independent of severity of illness, SCr and tubular injury markers. Higher urine to plasma ratios of individual secretory solutes isovalerylglycine and tiglylglycine were associated with MAKE28 after accounting for multiple testing. CONCLUSIONS Among critically ill adults, tubular secretory clearance is associated with adverse outcomes and measurement could improve assessment of kidney function and dosing of essential ICU medications. TRIAL REGISTRATION None. FUNDING PKB was supported by grants from the Digestive and Kidney Diseases K23DK116967, the University of Washington Diabetes Research Center P30DK017047, and an unrestricted gift to the Kidney Research Institute from the Northwest Kidney Centers. EDS was supported by the Vanderbilt O'Brien Kidney Center (NIDDK 5P30 DK114809-03) The funding sources had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.