Role of hemodilutional anemia and transfusion during cardiopulmonary bypass in renal injury after coronary revascularization: implications on operative outcome.

Objective: Acute renal injury and failure (ARF) after cardiopulmonary bypass (CPB) has been linked to low on-pump hematocrit (hematocrit). We aimed to 1) elucidate if and how this relation is modulated by the duration of CPB (TCPB) and on-pump packed red blood cell transfusions and 2) to quantify the impact of post-CPB renal injury on operational outcome and resource utilization.

Design: Retrospective review.

Setting: A Northwest Ohio community hospital.

Patients: Adult coronary artery bypass surgery patients with CPB but no preoperative renal failure.

Interventions: None.

Measurements And Main Results: We quantified post-CPB renal injury via 1) the peak postoperative change in serum creatinine (Cr) level relative to pre-CPB values (%DeltaCr) and 2) ARF, defined as the coincidence of post-CPB Cr > or =2.1 mg/dL and >2 times pre-CPB Cr. The separate effects of lowest hematocrit, intraoperative packed RBC transfusions, and TCPB on %DeltaCr and ARF were derived via multivariate regression, overlapping quintile subgroup analyses, and propensity matching. Lowest hematocrit (22.0% +/- 4.6% sd), TCPB (94 +/- 35 mins), and pre-CPB Cr (1.01 +/- 0.23 mg/dL) varied widely. %DeltaCr varied substantially (24 +/- 57%), and ARF was documented in 89 patients (5.1%). Both %DeltaCr (p < .001) and ARF (p < .001) exhibited sigmoidal dose-dependent associations to lowest hematocrit that were 1) modulated by TCPB such that the renal injury was exacerbated as TCPB increased, 2) worse in patients with relatively elevated pre-CPB Cr (> or =1.2 mg/dL), and 3) worse with intraoperative packed red blood cell transfusions (n = 385; 21.9%), in comparison with patients at similar lowest hematocrit. Operative mortality (p < .01) and hospital stays (p < .001) were increased systematically and significantly as a function of increased post-CPB renal injury.

Conclusions: CPB hemodilution to hematocrit <24% is associated with a systematically increased likelihood of renal injury (including ARF) and consequently worse operative outcomes. This effect is exacerbated when CPB is prolonged with intraoperative packed red blood cell transfusions and in patients with borderline renal function. Our data add to the concerns regarding the safety of currently accepted CPB practice guidelines.