Stress Biomarkers Do Not Correlate With Risk Factors for Kidney Injury After Cardiac Surgery

Abstract

Background: The urinary cell cycle arrest biomarkers (UBs) insulin-like growth factor-binding protein-7 and tissue inhibitor of metalloproteinases-2 provide early detection of kidney stress, and elevations may predict cardiac surgery-associated acute kidney injury (CS-AKI). We sought to determine whether known clinical risk factors for CS-AKI correlated with increased UB values.

Methods: UBs were measured over a 12-month period the morning after on-pump cardiac surgery. Patients with a preoperative serum creatinine level greater than 2.0 mg/dL or patients undergoing dialysis were excluded. Known clinical AKI risk factors in patients with elevated UB (>0.3 (ng/mL)²/1000), that is known to correlate with kidney stress, were compared with patients with low scores (≤0.3 (ng/mL)²/1000) by using logistic regression; the analysis was repeated with UB as a continuous variable.

Results: A total of 412 patients met inclusion criteria. Unadjusted results demonstrated a clinically similar CS-AKI risk profile in patients with either elevated or low UB values. The Pearson correlation between preoperative estimated glomerular filtration rate and UB was low (r = 0.16). Clinical risk factors for CS-AKI were not associated with elevated UB values in the logistic regression model, thus producing an area under the receiver operating characteristic curve of 0.63. Linear regression analysis also found few associations between CS-AKI clinical risk factors and UB when measured as a continuous variable, (R²) = 0.15.

Conclusions: Traditional CS-AKI clinical risk factors do not differ between patients with normal or elevated UB values. This UB test may identify patients at increased risk for AKI who otherwise would appear to be at low risk by traditional metrics.