The human parvovirus B19/human immunodeficiency virus co-infection in healthy eligible voluntary blood donors at the Blood Transfusion National Center in Kinshasa

Introduction Parvovirus B19 (PVB19) is one of several viruses transmissible by blood transfusion. Levels of exposure to PVB19 among HIV-infected voluntary blood donors are comparable to those among HIV-negative controls because, in blood donors, the PVB19 infection is transmitted mainly via the respiratory route. Thus, we hypothesize that the seroprevalence of PVB19 in HIV-positive blood donors is equal to the seroprevalence of PVB19 in HIV-negative blood donors. The objective of this study was to compare the seroprevalence of PVB19 between asymptomatic HIV-positive and HIV-negative blood donors. Methods A random sample of 360 eligible blood donors were firstly examined for HIV antibodies by using ELISA automaton and so were categorized as HIV-positive donors and HIV-negative donors. Then the two categories of donors were examined for PVB19 IgG and IgM by using ELISA kits. The seroprevalence of PVB19 in HIV-positive donors was compared to that of HIV-negative donors by using chi-square test or Fisher's exact test. All statistical analyzes were performed with SPSS 21. Results The prevalences of PVB19 IgG and IgM in HIV-positive blood donors were 92.1% (35 of 38) and 44.7% (17 of 38), respectively and those in control group were 89.1% (287 of 322) and 46.3% (149 of 322), respectively. But for both IgG and IgM the difference was not statistically significant (p > 0.05). Conclusion This research confirms our hypothesis: the seroprevalence of PVB19 in HIV-positive blood donors is equal to the seroprevalence of PVB19 in HIV-negative blood donors.

The transmission of PVB19 is mainly via the respiratory route, from the mother to the fetus, through the blood product and through transplants [9]. Iatrogenic transmission occurs through blood transfusion or organ transplantation from the seropositive donor.
Iatrogenic transmission is favored by three important features of the virus: (i) persistent virus infection in the bone marrow of an asymptomatic carrier [10]; (ii) prolonged replication after infection or initial reinfection [11]. In immunocompromised patients, infection may persist by reactivation or reinfection [12]; (iii) the resistance of the virus to many inactivation methods used in the manufacture of blood derivatives, plasma derivatives and labile blood products [13][14][15][16][17]. The transmission of PVB19 by blood transfusion occurs during the period of high viraemia in the donor. Viremia occurs approximately 1 week after primary infection and persists at elevated titres of up to 10 14 viral particles/mL in plasma for approximately 7 days [13,18].
Several cases of transfusion transmission of PVB19 have been reported, and many contaminated blood donations have been retrospectively or prospectively detected [15,16,[19][20][21][22]. PVB19, also called erythrovirus B19, is the basis of several syndromes whose clinical manifestations may be moderate or severe. They vary according to the hematological and immunological status of the infected person [23].
The child, the pregnant woman, the persons suffering from chronic hemolysis and the immunocompromised, are the most affected persons [24]. In the immunocompetent, the infection is usually asymptomatic or nonspecific. It can cause subclinical and limited aplasia of red blood cells followed by skin rash or arthralgia. The best known clinical manifestation in children is erythema infectiosum (fifth pediatric eruptive disease) [25]. It is a moderately intense facial erythema with cheeks, the prodrome of which is characterized by fever, colds, headache and nausea. An association between PVB19 infection and arthropathy was established in 1985. In non-immunized pregnant women, PVB19 carries a risk of fetal anasarca. In people with chronic hemolysis, such as sickle cell and thalassemic and not yet immunized, PVB19 can cause profound central anemia. In immunocompromised patients, such as patients receiving chemotherapy or people infected by HIV, PVB19 infection may be the cause of chronic anemia (red blood cell aplasia) following continuous and uncontrolled replication of the virus causing destruction of erythroblasts [26]. About 5% of adults and 10% of children suffering with hematological malignancy and chemotherapy are chronically infected with PVB19 and therefore develop severe and sometimes fatal cytopenia [27]. The risk of PVB19 transmission in HIV-infected people is comparable to the risk in HIV-negative controls, since PVB19 infection is transmitted through the respiratory route [28]. Thus, we hypothesize that the seroprevalence of PVB19 in HIV-positive blood donors is equal to the seroprevalence of PVB19 in HIV-negative blood donors. The objective of this study is to compare the seroprevalence of PVB19 between HIV-positive blood donors and HIV-negative blood donors. Exclusion criteria: any blood donor who did not give informed consent or who did not meet the inclusion criteria above was excluded from the study. Thus, blood donors with chronic illness such as diabetes or high blood pressure, alcoholic or addicted to tobacco and menstruating women were excluded from this study.

Methods
Sampling technique: our sampling was probabilistic. The minimum sample size was calculated by the following formula [29]: n is the sample size; z is a constant from the normal distribution at a certain confidence level (usually 95% and z = 1.96); p is the assumed prevalence of the investigated disease; e is the margin of sampling error chosen or the degree of precision desired. As the prevalence of PVB19 infection in blood donors in the DRC is unknown, the prevalence of 94% in blood donors, reported in a previous study in Nigeria [30] served as a baseline. Indeed, according to this study, 83 out of 88 blood donors had anti-parvovirus B19 antibody: 13 donors were IgM positive; 33 IgG positive and 37 positive for both subtypes (IgM and IgG). Thus, the size of our sample for a degree of accuracy of 5% is: The minimum number required is 87. Taking into account nonrespondents, the sample size was increased by 10%. Thus, the final size was 96. We selected 360 donor samples; Collection of data: sociodemographic data included age, sex, province of origin, district and profession/occupation; biological data from blood donors included PVB19 IgM and IgG serological status. All  Table 1 shows the general characteristics of the study population.

Discussion
We used the HIV ELISA test allowed us to classify blood donors into two categories: HIV-positive donors and HIV-negative donors. Table   1  Then we looked for anti-PVB19 antibodies in both blood donor categories. Table 2  The seroprevalence of anti-PVB19 IgG was 92.1% in HIV-infected donors, which was not significantly higher than that (89.1%) in the normal blood donor controls (p >0.05). These results are in agreement with the findings by Van Elsacker et al. [40] and contrast those published by Bremner et al. [41], Naides et al. [ [44]. It seems that the seroprevalence of IgG is inversely correlated with the degree of immunodeficiency [35].
Patients with a CD4 count of over 300x10 6 cells/L are usually capable of producing neutralizing antibodies [45]. In the present paper we studied asymptomatic blood donors.
Although we could not determine the CD4 T-cell count in order to elucidate the subjects' immune status, the HIV infection in these individuals were described as clinical stage 1 (asymptomatic), according to the revised WHO clinical staging of HIV/AIDS for adults and adolescents [46]. As the current study group was asymptomatic and recently diagnosed with HIV infection, this seems likely explanation to the equality of anti-PVB19 IgG seroprevalence between HIV-infected donors and normal blood donors. Another potential explanation of this PVB19 seroprevalence equality between the HIVpositive and HIV-negative blood donors is that HIV-infected blood donors may have been exposed to PVB19 and therefore developed anti-PVB19 IgG antibodies before their HIV infection. In fact the results of Dockrell et al. study suggest that, although HIV-positive individuals with decreased CD4 counts may have less effective immune response to T-lymphocyte-dependent antigens, they may still retain antibodies to antigens to which they have previously been exposed [47]. This explanation is likely because, as shown in Table 2, two out of 38 HIV+ donors have never been exposed to PVB19 and are therefore IgG-/IgM-; while 35 out of 38 HIV+ donors are IgG+.
PVB19 has a worldwide distribution. The literature shows that the infection occurs normally in childhood [48]. In adulthood, donors already have IgG antibodies against PVB19. The degree of immunodeficiency and other confounding factors could explain the difference found in the IgG seroprevalence in different studies [45].
These variations in seroprevalence might be explained by seasonal, epidemiological or demographical characteristics that resulted in different rates of exposure to the virus [49].
Confounding factors may be the number of individuals included in the study, the sampling season. Epidemiological studies have shown that PVB19 activity occurs periodically, commonly in the form of outbreaks in the late spring and summer. These outbreaks represent an occasion in which susceptible individuals are at a higher risk of contracting PVB19 infection [50]. This study has limitations. The sensitivity and specificity of the ELISA kit were less than 100% for the PVB19 IgM

Conclusion
Our results indicate that PVB19 seroprevalence in HIV-seropositive blood donors is not significantly different from PVB19 seroprevalence in HIV-seronegative blood donors and thus confirm our hypothesis.
What is known about this topic  There is no data on HIV-PVB19 coinfection in the Democratic Republic of Congo.


This article provides data on HIV-PVB19 coinfection in the Democratic Republic of Congo and shows that the seroprevalence of PVB19 in HIV-seropositive blood donors is equal to the seroprevalence of PVB19 in HIV-seronegative blood donors.