Hemolysis during cardiac surgery is associated with increased intravascular nitric oxide consumption and perioperative kidney and intestinal tissue damage

Abstract

Introduction: Acute kidney injury (AKI) and intestinal injury negatively impact patient outcome after cardiac surgery. Enhanced nitric oxide (NO) consumption due to intraoperative intravascular hemolysis, may play an important role in this setting. This study investigated the impact of hemolysis on plasma NO consumption, AKI, and intestinal tissue damage, after cardiac surgery.

Methods: Hemolysis (by plasma extracellular (free) hemoglobin; fHb), plasma NO-consumption, plasma fHb-binding capacity by haptoglobin (Hp), renal tubular injury (using urinary N-Acetyl-β-D-glucosaminidase; NAG), intestinal mucosal injury (through plasma intestinal fatty acid binding protein; IFABP), and AKI were studied in patients undergoing off-pump cardiac surgery (OPCAB, N = 7), on-pump coronary artery bypass grafting (CABG, N = 30), or combined CABG and valve surgery (CABG+Valve, N = 30).

Results: FHb plasma levels and NO-consumption significantly increased, while plasma Hp concentrations significantly decreased in CABG and CABG+Valve patients (p < 0.0001) during surgery. The extent of hemolysis and NO-consumption correlated significantly (r (2) = 0.75, p < 0.0001). Also, NAG and IFABP increased in both groups (p < 0.0001, and p < 0.001, respectively), and both were significantly associated with hemolysis (R s = 0.70, p < 0.0001, and R s = 0.26, p = 0.04, respectively) and NO-consumption (Rs = 0.55, p = 0.002, and R s = 0.41, p = 0.03, respectively), also after multivariable logistic regression analysis. OPCAB patients did not show increased fHb, NO-consumption, NAG, or IFABP levels. Patients suffering from AKI (N = 9, 13.4%) displayed significantly higher fHb and NAG levels already during surgery compared to non-AKI patients.

Conclusions: Hemolysis appears to be an important contributor to postoperative kidney injury and intestinal mucosal damage, potentially by limiting NO-bioavailability. This observation offers a novel diagnostic and therapeutic target to improve patient outcome after cardiothoracic surgery.

Keywords: acute kidney injury; cardiopulmonary bypass; hemolysis; intestinal fatty acid binding protein; nitric oxide.

Figure 1

Change of plasma free hemoglobin (fHb, A), NO-consumption (B), and haptoglobin (D) in patients undergoing CABG+Valve (■ red line), CABG (• blue line), and OPCAB surgery (▲ green line), and correlation between hemolysis and NO-consumption (C). The total release of fHb and change in NO-consumption (C) during surgery, estimated using the area under the curve (AUC) were significantly correlated (r² = 0.75, p < 0.0001). (D) Depicts the change in perioperative Hp levels in all groups. All values were corrected for hematocrit because of significant intraoperative hemodilution. Values are mean ± s.e.m. (A,B,D) or a scatter plot with mean (black line) ± 95% confidence interval (dotted line, C). Stars indicate significant changes compared to baseline within groups: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, ^****p < 0.0001. Numbers 1–8 on the y-axis (A,B,D) refer to collection time-points of blood specimens: 1, preoperatively, after induction but prior to sternotomy; 2, before start of CPB; 3, end CPB; 4, 15 min after cessation of CPB; 5, 2 h after cessation of CPB; 6, 4 h after cessation of CPB; 7, day 1 postoperatively; 8, day 2 postoperatively. CPB was not used in OPCAB patients.

Figure 2

Change of urinary NAG (A) and plasma IFABP (B) in patients undergoing CABG+Valve (■ red line), CABG (• blue line), and OPCAB surgery (▲ green line), and correlation between hemolysis and NAG (C) or IFABP release (D). The total release of fHb and NAG (C), and fHb and IFABP (D), estimated using the area under the curve (AUC), were significantly correlated (R_s = 0.70, p < 0.0001, and R_s = 0.26, p = 0.04, respectively). Plasma IFABP values were corrected for hematocrit because of significant intraoperative hemodilution, urinary NAG levels were corrected for urine creatinine values. Values are mean ± s.e.m. (A,B) or a scatter plot with mean (black line) ± 95% confidence interval (dotted line, C,D). Stars indicate significant changes compared to baseline within groups: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, ^****p < 0.0001. Numbers 1–8 on the y-axis (A,B) refer to collection time-points of blood specimens: 1, preoperatively, after induction but prior to sternotomy; 2, before start of CPB; 3, end CPB; 4, 15 min after cessation of CPB; 5, 2 h after cessation of CPB; 6, 4 h after cessation of CPB; 7, day 1 postoperatively; 8, day 2 postoperatively. CPB was not used in OPCAB patients.

Figure 3

Patients with postoperative AKI (gray bars) display significantly higher fHb (A) and urinary NAG levels (B), compared to non-AKI patients (white bars). Values are mean + s.e.m. Stars indicate significant changes compared to baseline within groups: ^*p < 0.05,^**p < 0.01, ^***p < 0.001, ^****p < 0.0001. Hash symbols indicate significant differences between groups: ^#p < 0.05, ^##p < 0.01, ^###p < 0.001. Numbers 1–8 on the y-axis refer to collection time-points of blood specimens: 1, preoperatively, after induction but prior to sternotomy; 2, before start of CPB; 3, end CPB; 4, 15 min after cessation of CPB; 5, 2 h after cessation of CPB; 6, 4 h after cessation of CPB; 7, day 1 postoperatively; 8, day 2 postoperatively.