The potential utility of urinary biomarkers for risk prediction in combat casualties: a prospective observational cohort study

Abstract

Abstract Introduction Traditional risk scoring prediction models for trauma use either anatomically based estimations of injury or presenting vital signs. Markers of organ dysfunction may provide additional prognostic capability to these models. The objective of this study was to evaluate if urinary biomarkers are associated with poor outcomes, including death and the need for renal replacement therapy. Methods We conducted a prospective, observational study in United States Military personnel with traumatic injury admitted to the intensive care unit at a combat support hospital in Afghanistan. Results Eighty nine patients with urine samples drawn at admission to the intensive care unit were studied. Twelve patients subsequently died or needed renal replacement therapy. Median admission levels of urinary cystatin C (CyC), interleukin 18 (IL-18), L-type fatty acid binding protein (LFABP) and neutrophil gelatinase-associated lipocalin (NGAL) were significantly higher in patients that developed the combined outcome of death or need for renal replacement therapy. Median admission levels of kidney injury molecule-1 were not associated with the combined outcome. The area under the receiver operating characteristic curves for the combined outcome were 0.815, 0.682, 0.842 and 0.820 for CyC, IL-18, LFABP and NGAL, respectively. Multivariable regression adjusted for injury severity score, revealed CyC (OR 1.97, 95 % confidence interval 1.26-3.10, p = 0.003), LFABP (OR 1.92, 95 % confidence interval 1.24-2.99, p = 0.004) and NGAL (OR 1.80, 95 % confidence interval 1.21-2.66, p = 0.004) to be significantly associated with the composite outcome. Conclusions Urinary biomarker levels at the time of admission are associated with death or need for renal replacement therapy. Larger multicenter studies will be required to determine how urinary biomarkers can best be used in future prediction models.