A prospective multicentre study assessing the random error rates of routine coagulation tests

Abstract number: P1596

Malan^* E.

^*Monash Medical Centre, Australia

New random access coagulation analyzers are now in routine use in most diagnostic laboratories. We have undertaken this study to assess their reliability to assist in the choice of single or duplicate testing. Important clinical decisions are made on the basis of the International Normalised Ratio (INR), and Activated Partial Thromboplastin Time (APTT) results and as such it would be important to know the error rate. A prospective multicenter study was conducted involving 6 major diagnostic laboratories across Australia using different Analyzer/reagent combinations. A total of 10 169 INRs and 7950 APTTs were performed in a routine random access operating mode. Instruments used included ACL Futura (Instrumentation Laboratories), STA and STAR (Diagnostica Stago) and CA 1500 (Sysmex). INR reagents used were Haemoliance Recombiplastin, Dade-Behring Thromborel S (ThS), Diagnostica Stago Neoplastine (Neo) and Dade-Behring Innovin. APTT reagents used were Dade-Behring Actin FSL, Organon-Teknika Platelelin LS, Organon-Teknika PTT Automate(Auto) and Dade-Behring Actin FS. A result was considered unacceptable if its duplicate exceeded 10%. Samples whose duplicates exceeded 10% were retested to ascertain which result was valid. It was considered clinically significant if the deviation altered the result from normal to abnormal or therapeutic to non therapeutic. Our study showed that a clinically significant error occurred at a rate of one in every 100–1000 tests, with the errors appearing to be random in nature. Duplicate analysis is recommended for routine coagulation testing and each laboratory should determine individual error rate as well as its clinical implication in their own laboratory.

Table 1