EBV usually establishes lifelong persistent infection and latency; 50% of children under 5 years of age are infected, and up to 90% of adults acquire the infection. Therefore, the age proportion in each group is important for the analysis of EBV primary infection or reactivation (Cohen et al. 2011; Porter et al. 1969). From May 2016 to April 2017, 831 patients were evaluated, and 252 patients were excluded based on the exclusion criteria. Among the evaluated patients, 186 were enrolled in the HIV/AIDS HAART use (H) group, 186 were enrolled in the HIV/ AIDS non-HAART use (NH) group, and 207 were enrolled in the HIV-negative control group (Fig. 1). No differences were observed in the average age (P = 0.291, analyzed by Mann–Whitney U test), gender, marital status, or infection route between the HIV/AIDS (H) and the HIV/AIDS (NH) groups (Table 1).
Table 1. Basic characteristics of the HIV/AIDS patients in this study.
We analyzed the seroprevalence of EBV in the HIV/AIDS patients and the HIV-negative controls. For EBV IgG and IgM, we observed a significantly higher proportion of patients positive for EBV IgG in the HIV/AIDS (H) and HIV/AIDS (NH) groups than in the HIV-negative control group (Fig. 2A). However, EBV IgM displayed a significantly higher positive rate in the HIV/AIDS (NH) group than in the HIV/AIDS (H) group (68.98% ± 2.80% vs. 52.68% ± 6.07%, P < 0.001), but both HIV/AIDS (H) and (NH) groups showed significantly higher values than the HIV-negative control group (68.98% ± 2.80% and 52.68% ± 6.07% vs. 46.69% ± 2.40%, P < 0.001, respectively). An EBV IgG-positive result represents past infection and an IgM-positive result represents EBV primary infection or reactivation, according to a previous study (Fourcade et al. 2017). Notably, the HIV/AIDS (NH) group's IgM/IgG ratio (0.74 ± 0.03) was significantly higher than that of the HIV/AIDS (H) group (0.59 ± 0.02, P < 0.001) and the HIV-negative control groups (0.57 ± 0.03, P < 0.001) (Fig. 2B).
Figure 2. EBV serological analysis. A The proportions of EBV-specific IgG and IgM in the three groups. B The EBV-specific IgM/IgG ratio. HAART use information was collected from the hospital database. The data shown are the mean ± SD (HIV/AIDS, n = 186/group; HIV-negative controls, n = 207), and each experiment was performed in duplicate. NS, not statistically significant, ***P < 0.001.
To test whether EBV salivary shedding, which can increase the risk of transmission, was significantly higher in the HIV/AIDS (NH) group than in the HIV-negative controls, salivary EBV DNA was detected by qRT-PCR. Surprisingly, the salivary EBV DNA-positive rate was 100% in the HIV/AIDS (NH) group. The HIV/AIDS (H) group (76.83% ± 0.15%) had a relatively higher EBV DNApositive rate than the HIV-negative control group (35.37% ± 0.31%). The EBV DNA-positive rate was nearly twofold higher in the HIV/AIDS groups than in the HIV-negative control group (Fig. 3A). Moreover, the HIV/ AIDS groups had a significantly higher viral load of EBV (≥ 100, 000 copies/mL in saliva) when compared to the HIV-negative control group (Fig. 3B).
Figure 3. EBV in saliva. A The proportion of the EBV DNApositive rate among the three groups. B The viral quantity distribution among the three groups. The data shown are mean ± SD (HIV/AIDS, n = 186/group; HIV-negative controls, n = 207) and each experiment was performed in duplicate. The data in the figure are representative of the analysis results. NS, not statistically significant, **P < 0.01; ***P < 0.001.
The results revealed that HIV infection could increase the salivary viral positive rate and viral load of EBV. In contrast, HIV infection can increase the risk of acquiring HBV, with risk being inversely associated with CD4 counts (Liu et al. 2018; Kelly et al. 2018). The correlation between acquiring EBV infection and CD4 counts or HAART use in HIV-infected patients was poorly understood. Subsequently, we further investigated whether higher CD4 counts can reduce the risk of EBV primary infection or reactivation. HAART use has been demonstrated to help CD4 immune reconstitution in HIV/AIDS patients (Yao et al. 2013); in particular, HAART use in HIV/AIDS patients is likely critical for decreasing the risk for complications. We also analyzed the CD4 cell counts and viral load of HIV in the HIS database of EBV-negative and -positive HIV/AIDS patients and found that higher CD4 cell counts were observed in the EBV-negative HAART use group (P < 0.001) (Fig. 4). Among the HIV-positive patients, 67% had a HIV viral load under the detection limit after more than 6 months of HAART use, which was correlated with the EBV-negative rate (Table 1).
Figure 4. CD4 cell counts in EBV DNA-positive and -negative groups. CD4 cell counts of HIV/AIDS patients were determined by flow cytometry. The reports were kept in the hospital database. The data are representative of the analysis results. Statistical significance was measured by a two-way ANOVA and a two-tailed unpaired Student's t test. ***P < 0.001.