Blood transfusions are associated with urinary biomarkers of kidney injury in cardiac surgery

Abstract

Objective: Cardiac surgery is a major cause of acute kidney injury. In this setting, receipt of blood transfusions seems to be associated with a higher risk of acute kidney injury, as measured using serum creatinine values. We examined this association further by using urinary biomarkers of kidney injury.

Methods: A total of 1210 adults underwent cardiac surgery and were divided into 3 groups on the basis of the receipt of intraoperative packed red blood cell units: no blood (n = 894), 2 or less packed red blood cell units (n = 206), and more than 2 packed red blood cell units (n = 110). Acute kidney injury was defined as (1) doubling of serum creatinine from the preoperative value; (2) first postoperative urinary interleukin-18 in the fifth quintile; and (3) first postoperative urinary neutrophil gelatinase-associated lipocalin in the fifth quintile. We determined the relative risk for acute kidney injury outcome according to packed red blood cell units group after adjusting for 12 preoperative and surgical variables. By using the Sobel test for mediation analysis, we also evaluated the role of biomarkers in causing acute kidney injury through alternative pathways.

Results: Acute kidney injury was more common in those who received more than 2 packed red blood cell units. In patients receiving more than 2 packed red blood cell units, the adjusted relative risks were 2.3 (95% confidence interval, 1.2-4.4, P .01), 1.36 (95% confidence interval, 1.0-1.9, P .05), and 1.34 (95% confidence interval, 1.0-1.8, P .06) for doubling of serum creatinine, urinary interleukin-18 in the fifth quintile (>60 pg/mL), and urinary neutrophil gelatinase-associated lipocalin in the fifth quintile (>102 ng/mL), respectively. Furthermore, the effect of packed red blood cell units transfusion on acute kidney injury was partially mediated by interleukin-18.

Conclusions: Receipt of 2 or more packed red blood cell units during cardiac surgery is associated with a greater risk of acute kidney injury defined by serum creatinine and kidney injury biomarkers.

Figure 1

Panel A describes the direct effect of independent variable (X) on dependent variable (Y) denoted as C. Panel B describes the effect of X on Y through a mediator (M) where the association between X and M is denoted as ‘a’, association between M and Y is denoted as ‘b’ and association of X and M to Y is denoted as Ć, such that C = Ć + ab

Figure 2

The distribution of urinary biomarkers in different groups by blood transfusion is shown in box plot. The bars represent the inter-quartile range and the solid lines in the bars represent the median values of first post-operative urinary IL-18 and urinary NGAL. The median values of both urinary IL-18 and urinary NGAL were significantly elevated in the group that received > 2 PRBC units compared to those who received < 2 (P values for Wilcoxon test <.0001 and <.0001 respectively) or no blood (P values for Wilcoxon test <.0001 and <.0001 respectively). * No Blood vs. More than 2 PRBC p value <.0001, ^ Less than 2 PRBC vs. More than 2 PRBC p value <.0001.