An Agreement Study Between Point-of-Care and Laboratory Activated Partial Thromboplastin Time for Anticoagulation Monitoring During Extracorporeal Membrane Oxygenation

Yuan Teng; Shujie Yan; Gang Liu; Song Lou; Yang Zhang; Bingyang Ji

doi:10.3389/fmed.2022.931863

An Agreement Study Between Point-of-Care and Laboratory Activated Partial Thromboplastin Time for Anticoagulation Monitoring During Extracorporeal Membrane Oxygenation

Front Med (Lausanne). 2022 Jun 29:9:931863. doi: 10.3389/fmed.2022.931863. eCollection 2022.

Authors

Yuan Teng¹, Shujie Yan¹, Gang Liu¹, Song Lou¹, Yang Zhang², Bingyang Ji¹

Affiliations

¹ Department of Cardiopulmonary Bypass, National Center for Cardiovascular Diseases, Fuwai Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China.
² Center of Laboratory Medicine, National Center for Cardiovascular Diseases, Fuwai Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China.

Abstract

Background: Laboratory activated partial thromboplastin time (LAB-aPTT) is a widely used laboratory assay for monitoring unfractionated heparin (UFH) therapy during extracorporeal membrane oxygenation (ECMO). But LAB-aPTT is confined to a central laboratory, and the procedure is time-consuming. In comparison, point-of-care aPTT (POC-aPTT) is a convenient and quick assay, which might be a promising method for anticoagulation monitoring in ECMO. This study was aimed to evaluate the agreement between POC-aPTT (hemochron Jr. Signature instruments) and LAB-aPTT for anticoagulation monitoring in adult ECMO patients.

Methods: Data of ECMO-supported adult patients anticoagulated with UFH in our institute from January 2017 to December 2020 was retrospectively reviewed. POC-aPTT and LAB-aPTT results measured simultaneously were paired and included in the analysis. The correlation between POC-aPTT and LAB-aPTT was assessed using Spearman's correlation coefficient. Bias between POC-aPTT and LAB-aPTT were described with the Bland-Altman method. Influence factors for bias were identified using multinomial logistic regression analysis.

Results: A total 286 pairs of aPTT results from 63 patients were included in the analysis. POC-aPTT and LAB-aPTT correlated weakly (r = 0.385, P < 0.001). The overall bias between POC-aPTT and LAB-aPTT was 7.78 [95%CI (-32.49, 48.05)] s. The overall bias between POC-aPTT and LAB-aPTT ratio (to normal value) was 0.54 [95%CI (-0.68, 1.76)]. A higher plasma fibrinogen level [OR 1.353 (1.057, 1.733), P = 0.017] was associated with a higher chance of POC-aPTT underestimating LAB-aPTT. While a lower plasma fibrinogen level [OR 0.809 (0.679, 0.963), P = 0.017] and lower UFH rate [OR 0.928 (0.868, 0.992), P = 0.029] were associated with a higher chance of POC-aPTT overestimating LAB-aPTT.

Conclusion: The present study showed poor agreement between POC-aPTT and LAB-aPTT. POC-aPTT was not suitable for anticoagulation monitoring in adult ECMO patients.

Keywords: activated partial thromboplastin time; anticoagulation; extracorporeal membrane oxygenation; point of care; unfractionated heparin.