Traditional and non-traditional anticoagulation management during extracorporeal membrane oxygenation.

Unfractionated heparin (UFH) is the anticoagulant of choice during extracorporeal membrane oxygenation (ECMO) support. Despite its favorable pharmacologic properties, management of heparin anticoagulation during ECMO remains a major challenge. To date, little is known about the optimal monitoring strategy or the heparin dose offering the best safety/efficacy profile. Therefore, it remains unclear if the heparin dose should be adapted to target a specific "clotting time" [e.g., activated clotting time (ACT) or activated partial thromboplastin time (aPTT)] or a heparin concentration, measured by coagulation factor anti-Xa assay. In addition, no study has compared the relevance of modern viscoelastic coagulation tests over the single value of a clotting time or heparin concentration value. Although guidelines for anticoagulation during ECMO support have been published, the absence of evidence limits the quality of the recommendations provided, which explains the major intra- and inter-institutional variability observed. Large prospective multicenter trials are urgently needed to investigate the optimal anticoagulation management strategy during ECMO support.

Risk factors for adverse outcomes after left ventricular assist device implantation and extracorporeal cardiopulmonary resuscitation.

Left ventricular assist device implantation following extracorporeal cardiopulmonary resuscitation has been associated with ambivalent results. In a series of patients who underwent left ventricular assist device implantation after extracorporeal cardiopulmonary resuscitation, we investigated whether the outcome can be predicted by preoperative risk factors or established risk scores. Primary endpoint was a composite of mortality and severe neurological disabling over 1 year of follow-up. To assess predictors of the primary endpoint, we performed univariate and multivariable Cox regression analyses. Of the 40 patients included, 24 patients (60%) experienced the primary endpoint. Renal replacement therapy and the Vasoactive-Inotropic Score were independently associated regarding the primary endpoint with a hazard ratio for renal replacement therapy of 2.50 (95% confidence interval: 1.09-5.70; P = 0.021) and for the Vasoactive-Inotropic Score of 1.02 per unit (95% confidence interval: 1.00-1.03; P = 0.040). The risk of experiencing an unfavorable outcome during follow-up in patients with a Vasoactive-Inotropic Score of 20 who needed renal replacement therapy or did not need renal replacement therapy was 78% and 54%, respectively. Our data indicate that a decision to implant a left ventricular assist device in patients requiring renal replacement therapy and revealing a high Vasoactive-Inotropic Score after extracorporeal cardiopulmonary resuscitation should be reached with caution.

Extracorporeal resuscitation as a further modifier of clinical outcome in patients with left ventricular assist device implantation and Interagency Registry for Mechanically Assisted Circulatory Support level 1.

In cardiogenic shock patients with Interagency Registry for Mechanical Circulatory Support (INTERMACS) level 1, the need for temporary circulatory support is a predictor and modifier of patient outcome. Because this group includes patients with and without cardiopulmonary resuscitation (CPR) and is thus very heterogeneous, we investigated whether a further subclassification is useful. We compared 30-day and 1-year mortality of patients who underwent left ventricular assist device implantation after extracorporeal CPR with the aid of an extracorporeal life support system (CPR+ group; n = 40) with cardiogenic shock patients in which the extracorporeal life support system was implanted under non-CPR conditions (CPR- group, n = 68). In the CPR+ and CPR- groups, 30-day mortality was 27.5% (n = 11) and 8.8% (n = 6), respectively (P = 0.014). The values for 1-year mortality were 57.5% (n = 23) and 36.8% (n = 25), respectively (P = 0.023). The age- and gender-adjusted hazard ratios of 30-day and 1-year mortality for the CPR+ group versus the CPR- group were 3.88 (95% confidence interval 1.29-11.7; P = 0.016) and 1.79 (95% confidence interval 1.01-3.17; P = 0.045), respectively. In conclusion, our data show that left ventricular assist device implantation with extracorporeal life support following CPR is associated with high 30-day and 1-year mortality. Further multicentre studies are needed to confirm these results and potentially add CPR as a new modifier to the INTERMACS profile.

Bivalirudin anticoagulation for left ventricular assist device implantation on an extracorporeal life support system in patients with heparin-induced thrombocytopenia antibodies.

OBJECTIVES: Heparin-induced thrombocytopenia (HIT) requires alternative anticoagulation strategies. We investigated outcomes in patients with HIT antibodies undergoing low-dose bivalirudin anticoagulation during left ventricular assist device implantation on an extracorporeal life support system (ECLS) and compared the results with non-HIT patients treated with heparin and receiving left ventricular assist device implantation with ECLS support. METHODS: The institutional ventricular assist device database was searched for the period from March 2012 to March 2016. The primary end-point was the need for early (<7 days) surgical re-exploration due to persistent haemorrhage or cardiac tamponade postoperatively. The secondary clinical end-points were delayed chest closure, stroke, intracranial bleeding, re-thoracotomy >7 days and mortality up to 1 year. Unadjusted comparison was used for the entire groups. Because of non-random group assignment, propensity score matching was also performed to compare treatment effects. RESULTS: Twenty-one patients were treated with bivalirudin and 36 patients with heparin. INTERMACS levels were lower, inotropic score was higher and the prevalence of mechanical ventilation and preoperative ECLS implants was also significantly higher in the heparin group than in the bivalirudin group (P-values <0.05). The primary end-point was reached by 19% in the bivalirudin group and 16.7% in the heparin group (bivalirudin group: odds ratio 1.18, 95% confidence interval 0.29-4.76; P = 0.820). The propensity score-matched groups also showed no difference in this regard (P = 0.455). All secondary clinical end-points were comparable between groups, both in the unadjusted analysis and in the propensity score-matched groups. CONCLUSIONS: In patients with HIT antibodies, low-dose bivalirudin anticoagulation on ECLS support appears to be a feasible option for left ventricular assist device implantation.