An audit of coagulation specimens requiring direct human involvement at an academic laboratory : peer reviewed original article

Introduction A trend exists towards automation in laboratory testing, including coagulation investigations. Some coagulation specimens cannot be summarily tested in fully automated systems because of suboptimal, pre-analytical factors that would influence results, while others require additional preparation prior to testing. The aim of this audit was to determine the proportion of coagulation specimens requiring direct human involvement for screening and additional sample preparation, and to analyse contributing factors. Methods All specimens received for coagulation testing over a ten-month period were audited. Specimens requiring direct human involvement (screening for suboptimal, pre-analytical influences; double centrifugation) were recorded. Results Of 8 614 specimens, 1 091 (12.7%) fell into one or more of the following categories: under-filled (2.9%); clotted (1.7%); incorrect specimen tubes (1.5%); lipaemic (0.1%); icteric (0.5%); haemolysed (0.9%); haematocrit >0.55 L/L (0.02%); old (1.9%); unlabelled or incorrectly labelled (0.2%); lupus anticoagulant (4.0%); activated protein-C resistance (0.1%). In total, 353 (4.1%) specimens required double centrifugation and 829 (9.6%) required screening. Conclusion A substantial proportion of specimens required direct human involvement and would not have been suitable for coagulation testing in current fully automated systems.