THE USE OF PFA-100 IN PEDIATRICS IN THE INVESTIGATION OF HEMOSTATIC DISORDERS, A RETROSPECTIVE COHORT STUDY

Abstract

Abstract BACKGROUND The screening of hemostatic disorders in paediatrics is challenging because often children have never undergone hemostatic challenges. The PFA-100 was introduced as an alternative to the in vivo bleeding time test and is now routinely used in the screening of these disorders. Numerous studies have assessed its limits and there remains uncertainty regarding its utility. OBJECTIVES The goal of our study was to describe the utilization of the PFA-100 in a tertiary paediatric center, to evaluate its accuracy and to determine the average cost generated by the investigation following an abnormal result. DESIGN/METHODS We did a retrospective descriptive study by reviewing the charts of 250 paediatric patients who had a PFA-100 done in our center over a one year period. Accuracy and reliability of the PFA-100 were calculated with a ROC curve (95% CI). RESULTS Of 225 patients included in our study, 99 (44%) PFA-100’s were done in a preoperative context, and in average, each patient had 1.52 PFA-100’s (range 1–6). Personal hemorrhagic diathesis was not described in 59 patients (26.2%). 91 patients (40.4%) had an abnormal PFA-100 result. Further investigation revealed no disorder in 55 patients (60.4%), von Willebrand disease in 20 patients (8.9%) and non-specific platelet function defects in 15 patients (6.7%). No severe platelet function defects were found in our study cohort. There was a high variability of the PFA-100 result in a same patient (coefficient of variation 15.8%). A higher cut-off value of 140 seconds for PFA-100 col_EPI (AUC 0,87) and 111 seconds for PFA-100 col_ADP (AUC 0,68) produced more accurate sensitivity (89% and 74%), specificity (78% and 63%) and positive (44% and 49%) and negative (97% and 84%) predictive values, compared to our center’s threshold of 130 seconds and 100 seconds respectively. The average cost per patient for the investigation following an abnormal PFA-100 was 434,92$ ± 232,66 (CI 95%). CONCLUSION PFA-100 was mainly used in a preoperative context in our study and patient history was often incomplete. The PFA-100 shows high sensitivity and negative predictive value making it a potentially good screening test however it is to be used with caution considering the high variability of results in a single patient.