Utility of oral fluid samples for hepatitis B antibody detection in real life conditions

Abstract

Background: Hepatitis B virus (HBV) testing in oral fluid samples may provide advantages in diagnosis, screening or prevalence studies, especially among individuals with venous access difficulties. This study aims to optimize one commercially available assay for detecting total anti-HBc marker in oral fluid samples and to evaluate its utility under real life conditions in different settings for the purposes of prevalence and diagnostic studies.

Methods: Oral fluid was collected using a Salivette device and some parameters were initially evaluated: type of elution buffer and sample volume. Thereafter, the utility of oral fluid samples for detection of anti-HBc was evaluated in real life conditions in which, 1296 individuals gave serum and oral fluid samples. All serum samples were submitted to commercial EIAs to detect total anti-HBc, according to the manufacturer's instructions and oral fluid samples according to previous optimization.

Results: In optimization evaluation, PBS/BSA 0.5% and 100 μL of oral fluid (volume was two-fold increased compared to serum in EIA) were chosen as transport buffer and sample volume. In the field study, anti-HBc was detected in 211 out of 1296 serum samples giving overall oral fluid sensitivity of 52.6% and specificity of 96%. Concordance was higher in ambulatory setting (67.7) compared to general population (31.8). Mean ± standard deviation values of optical density/cutoff (OD/CO) in serum samples were higher in false-negative oral fluid samples than those seen in true positive samples. Sensitivity was higher in those presenting active infection compared to anti-HBc isolate and past infection. Sensitivity also increased in the ambulatory group when HCV individuals were excluded.

Conclusions: It was possible to optimize a commercial EIA for detecting anti-HBc in oral fluid samples and where the highest concordance was found in ambulatory settings and among individuals with active infection.

Keywords: Diagnosis; Enzyme immunoassay; Hepatitis B virus; Oral fluid; Prevalence.

Fig. 1

DotPlot Optical Density obtained in oral fluid samples according to transport buffer. Transport Buffers: (1) PBS pH 7.2; (2) PBS/Tween 20 0.05%; (3) PBS/Tween 20 (0.05%)/Sodium azide (0.005%); (4) PBS/Tween 20 (0.2%)/BSA 5%; (5) PBS/BSA 0.5%. Notes: (a) Correlation coefficient Pearson: 0.961; (b) Mean and standard deviation not shown due to the low number of observations among those transfer buffers

Fig. 2

DotPlot Optical Density obtained according to different volumes of oral fluid sample on assay and their respective ROC curves

Fig. 3

OD ratios of each serum and oral fluid sample plotted according to EIA in negative and positive samples. The y-axis represents the OD ratio. The solid lines represent the average OD ratios for the serum and oral fluid samples, which were 0.881 ± 0.239 and 0.392 ± 0.130 for negative serum and oral fluid samples respectively, and 0.144 ± 0.100 and 0.021 ± 0.048 for positive serum and oral fluid samples respectively