RESUMO
INTRODUCTION: A precise and reliable screening assay for glucose 6-phosphate dehydrogenase (G6PD) deficiency would greatly help avoiding unwanted outcomes due to bilirubin neurotoxicity in neonatal jaundice and antimalarial-induced haemolytic anaemia in malaria patients. Currently, available assays are laborious and require sophisticated laboratory expertise. This study aimed to evaluate the performance of a recently introduced automated screening assay for G6PD deficiency by comparing with a routine spectrophotometric assay. METHODS: An automated UV-based enzymatic (Mindray, PRC) and spectrophotometric assays were performed simultaneously in parallel to determine G6PD activity in 251 blood samples from the subjects. RESULTS: The median G6PD activity value from spectrophotometric assay was significantly lower than that of from the automated assay. The mean difference was -2.0 U/g haemoglobin (-7.3 to 3.2; P < 0.0001). The mean activity values of both assays were strongly correlated with Pearson's correlation coefficient of r = 0.8. Cohen's kappa statistics between assays was 0.77 (0.70-0.83). The sensitivity, specificity, positive and negative predictive values of the automated assay were 85.7%, 99.2%, 85.7%, 99.2%, respectively. The sensitivity and positive predictive values of the automated assay for identifying intermediate G6PD activity levels were 40.0% and 25.0%, respectively. Genotyping was performed to confirm G6PD deficient and intermediate samples. The turnaround time for 40 samples was 60 minutes for the automated assay and 300 minutes for spectrophotometric assay. CONCLUSION: The automated assay for the detection of G6PD deficiency is comparable to a routine spectrophotometric assay and help reducing sample handling time. However, the assay shows limitation in identifying individuals with G6PD intermediate.