Pre-analytical variables in haemostasis: Findings from the United Kingdom National External Quality Assessment scheme for Blood Coagulation (UK NEQAS BC) haemolysis exercise.

Abstract

Haemolysis is considered one of the major contributors of nonconformities and sample rejection in coagulation testing. Two lyophilized plasmas were distributed to 800 centres registered for prothrombin time (PT), activated partial thromboplastin time (APTT) and either Clauss fibrinogen or thrombin time (TT) in the UK NEQAS BC programme. The same pool of normal plasma was used to prepare both samples, to one of which red blood cell haemolysate was added to mimic haemolysis at 3g/L haemoglobin concentration. Participants were asked to complete a questionnaire about their laboratory approach to dealing with haemolysed samples, including strategies used to deal with different levels of haemolysis. Results for tests performed did not show great differences between the two samples. It should be noted that artificially constructed haemolysed samples may not behave in the same way as patient samples (ie, may not be commutable). However, the possibility of carrying out a large multicentre study for detection of haemolysis was demonstrated. Inconsistency in practice was observed with 226/551 (41%) of centres indicated they reject haemolysed samples solely on visual checks, and 163 (30%) using initial visual checks with further sample rejection evaluation by analyser flags. Furthermore, 333 (72%) of centres indicated that the level of haemolysis affects sample rejection decisions, while 132 (28%) stated it did not. Variability of responses for dealing with haemolysed samples reflects a lack of clear consistency in the pre-analytical area of sample processing.