Risk factors and outcomes of severe hemolysis during extracorporeal membrane oxygenation：a 5-year ;single-center retrospective analysis / 中华危重病急救医学

ABSTRACT

Objective To investigate the risk factors of severe hemolysis during extracorporeal membrane oxygenation (ECMO). Methods The clinical data of adult patients undergoing ECMO after cardiac surgery admitted to Fuwai Hospital from December 2010 to October 2015 were retrospectively analyzed. Demographic characteristics, renal function, primary disease, operation data, ECMO related data and outcomes were recorded. Patients were divided into normal free hemoglobin (FHB) group (FHB ≤ 500 mg/L) and severe hemolysis group (FHB > 500 mg/L) according to the FHB level during ECMO support. The parameters before and after ECMO support were compared between the two groups. Logistic regression was used to identify the independent risk factors of severe hemolysis. Results A total of 81 patients including 19 patients with severe hemolysis was enrolled, and 62 in normal FHB group. There was no difference in cardiopulmonary bypass (CPB) time, clamping time, lactate level before ECMO, cardiopulmonary resuscitation, intra-aortic balloon pump use and central catheter insertion between two groups. The maximums of serum creatinine (SCr) and FHB levels were higher in severe hemolysis group as compared with those in normal FHB group [maximal SCr (μmol/L) 281.02±164.11 vs. 196.67±87.31, maximal FHB (mg/L) 600 (600, 700) vs. 200 (100, 300)], the incidence of clots in circuit or oxygenator, infection, and hemofiltration in severe hemolysis group was increased [26.3% (5/19) vs. 4.8% (3/62), 31.6% (6/19) vs. 12.9% (8/62), 36.8% (7/19) vs. 14.5% (9/62), all P < 0.1]. As well as outcomes including the rate of site of surgery or intubation bleeding and acute renal failure [ARF, 57.9 % (11/19) vs. 30.6% (19/62), 94.7% (18/19) vs. 41.9% (26/62)], and the survival rate was lowered [10.5% (2/19) vs. 51.6% (32/62), all P < 0.05]. As result of univariate analysis, clots in circuit or oxygenator, infection and hemofiltration were associated with severe hemolysis. It was showed by logistic regression analysis that the clots in circuit or oxygenator was a risk factor of severe hemolysis during ECMO [odds ratio (OR) = 6.262, 95% confidence interval (95%CI) = 1.244-31.515, P = 0.026]. Conclusions The clots in circuit or oxygenator were independent risk factors of severe hemolysis during ECMO. Severe hemolysis can induce the increase of the rate of bleeding in the operation site or intubation and the rate of ARF, and decrease of the survival rate.