Discrepancy between parallel anti-Xa and activated partial thromboplastin time is related to mortality in patients with Impella-supported cardiogenic shock
Charlotte J. Van Edom 1),2); Francesca Fiorelli 3); Tim Balthazar 4); Maria Monteagudo-Vela 5); Thomas Vanassche 1),2); Vasileios F. Panoulas 3); Christophe Vandenbriele 3),6),7).
1) Department of Cardiovascular Diseases, University Hospitals of Leuven, Leuven, Belgium
2) Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium
3) Department of Cardiology and Cardiac Intensive Care, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, UK
4) Departments of Cardiology and Intensive Care, University Hospital Brussels, Jette, Belgium
5) Department of Cardiothoracic Surgery, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, UK
6) Heart Center, OLV Aalst, Aalst, Belgium
7) Division of Anaesthetics, Pain Medicine and Intensive Care, Faculty of Medicine, Imperial College London, London, UK
Abstract
Background
Managing unfractionated heparin (UFH) during percutaneous mechanical circulatory support for cardiogenic shock (CS) is challenging due to potential discrepancies between coagulation tests.
Objectives
To study the causes and consequences of discrepancies between anti-Xa and activated partial thromboplastin time (APTT) for UFH monitoring during microaxial flow pump support (Impella) for CS.
Methods
We assessed patients in CS supported with Impella in 2 tertiary care centers over 62 months. UFH was titrated based on anti-Xa levels with parallel APTT measurements. In-range anti-Xa levels were considered between 0.20 and 0.30 IU/mL or 0.31 and 0.50IU/mL, and the corresponding APTT values were 40 to 55 seconds and 56 to 80 seconds, respectively. Pearson correlation was calculated between anti-Xa and APTT. Samples with in-range anti-Xa but prolonged APTT were analyzed for abnormalities in international normalized ratio (INR; ≥1.5) and/or fibrinogen (<1.5g/L). Mortality during Impella support was then compared in those with and without additional coagulation abnormalities (chi-squared test).
Results
Correlation between anti-Xa and APTT was weak (r = 0.50, P < .001, N = 2447). When anti-Xa was in range (N = 1914 samples), 24% had short, 52% had in-range, and 24% had prolonged APTT. Of the 57 patients with prolonged APTT, 28 had abnormal same-day INR and/or fibrinogen, whereas 29 had normal fibrinogen and INR. Mortality was higher in patients with abnormal INR and/or fibrinogen than in those with normal fibrinogen and INR (32% vs 10%; P = .043).
Conclusion
Anti-Xa/APTT discrepancies are frequent during percutaneous mechanical circulatory support for CS, highlighting the importance of a multiple testing strategy. Outcomes of patients with prolonged APTT were related to the presence of abnormal INR and/or fibrinogen, suggesting a serious concomitant underlying disease.