ABSTRACT
ABSTRACT Introduction: In Brazil, the blood donor screening for hepatitis B virus (HBV) includes laboratory testing for serological (HBsAg and Anti-HBc) and molecular (HBV DNA) markers. This study aims to correlate serology reactive results with HBV DNA detection among blood donors with at least one HBV infection marker detected in a blood bank in northern Brazil. Method: A retrospective search for HBV reactive blood donor data from January 2017 to December 2019 was performed. Serological screening was performed by chemiluminescent microparticle immunoassays Architect HBsAg and Architect Anti-HBc, whereas molecular screening was performed by the HBV nucleic acid test (HBV NAT). Main results: A total of 556 HBsAg reactive results were detected, between positive (47.66%) and inconclusive (52.34%). A total of 3,658 Anti-HBc reactive results were detected, between positive (83.71%) and inconclusive (16.29%). None of the inconclusive results were associated with HBV DNA detection. The HBV DNA detection rates were 47.55% among HBsAg positive samples and 4.08% among Anti-HBc positive samples. The signal-to-cutoff (S/CO) ratio median of HBV NAT positive samples was superior in comparison to HBV NAT negative samples (p < 0.0001). The thresholds found to optimize sensitivity and specificity were 404.15 for Architect HBsAg and 7.77 for Architect Anti-HBc. Three blood donors were in the window period and 1 occult HBV infection case was detected. Conclusion: High S/CO ratios were more predictive of HBV DNA detection. However, a number of HBV NAT positive samples gave low values, while some HBV NAT negative samples showed high values, reaffirming the significance of molecular testing to enhance transfusion safety.
ABSTRACT
The present study aims to correlate the sample-to-cutoff ratios (S/CO) distributions of reactive results for HTLV-1/2 antibodies with the detection of proviral DNA in a population of blood donor candidates. It was carried out a retrospective data search of 632 HTLV-1/2 reactive samples, submitted to confirmatory testing from January 2015 to December 2019. Serological screening was performed by chemiluminescent microparticle immunoassay Architect rHTLV-I/II, whereas confirmatory testing was performed by in-house real-time polymerase chain reaction method. 496 out of 632 samples (78%) had undetectable HTLV-1/2 proviral DNA and 136 (22%) had detectable proviral DNA. HTLV infection was not confirmed in any individual for whom the S/CO ratio value was <4, and proviral DNA detection rates gradually escalated as S/CO ratio values increased. The sensitivity and predictive positive value found for the Architect rHTLV-I/II was 100% and 22%, respectively. The receiver operating characteristic (ROC) curve analysis showed that the optimal S/CO ratio value for predicting the presence of HTLV-1/2 was 18.11. High S/CO ratios were more associated with the detection of proviral DNA. The S/CO ratio value <4 suggests excluding true HTLV infection and the risk of blood transmission (AU).
O estudo tem como objetivo correlacionar às distribuições das razões sample-to-cutoff (S/CO) de resultados reagentes para anticorpos HTLV-1/2 com a detecção de DNA proviral em uma população de candidatos à doação de sangue. Realizou-se uma busca retrospectiva de dados de 632 amostras reagentes para HTLV-1/2 submetidas à testagem confirmatória entre janeiro de 2015 a dezembro de 2019. A triagem sorológica foi realizada pelo imunoensaio quimioluminescente de micropartículas Architect rHTLV-I/II, enquanto o teste confirmatório foi realizado pelo método de PCR em tempo real in-house. 496 de 632 amostras (78%) apresentaram DNA proviral indetectável e 136 (22%) apresentaram DNA proviral detectável. A infecção por HTLV não foi confirmada em nenhum indivíduo com valor de S/CO <4 e as taxas de detecção de DNA proviral escalonaram gradualmente à medida que as razões S/CO aumentaram. A sensibilidade e valor preditivo positivo encontrados para o Architect rHTLV-I/II foram 100% e 22%, respectivamente. Utilizando análise de curva ROC, o valor de razão S/CO ideal para predizer a presença de DNA proviral foi de 18,11. Razões S/CO elevadas foram mais associadas à detecção de DNA proviral. Em suma, o valor de S/CO <4 sugere a exclusão de infecção por HTLV e o risco de transmissão pelo sangue (AU).