Hepatitis C Virus Viral Load And Mother-To-Child Transmission: a Systematic Review and Meta - Analysis

Background: Hepatitis C virus (HCV) infection is increasing among pregnant women own to the opioid epidemic, and mother-to-child transmission (MTCT) was considered to be the leading cause of hepatitis C in the future. Thus, our aim of this study was to assess the association of maternal HCV viral load and human immunodeciency virus (HIV) coinfection with the risk of MTCT in pregnant women infected by HCV. Methods: We performed a literature searching using MEDLINE, EMBASE, CENTRAL, SCIE, CPCIS, SCOPUS, LILACS and WHO Global Index Medicus from inception to 1 March 2021. Manuscripts were included if they reported the incidence of HCV MTCT. Pooled effect estimates were obtained by using random-effects model and Holm-Bonferroni correction was performed for multiple pooled associations. Results: a total of 26 studies involving 4491 newborns with maternal HCV infected were identied. Pregnant women with HCV viremia had increased risk of MTCT (odds ratio [OR]: 8.14, 95% condence interval [CI]: 4.62-14.31) compared with those with negative HCV-RNA. Multiple-subgroup analysis showed HCV viremia/HIV+ group experienced the highest risk for HCV MTCT and then HCV viremia mono-infected group; HCV-RNA negative women experienced the lowest risk for HCV MTCT. Among HCV viremia women, elevated risk of MTCT was found in subjects with viral load ≥ 6 log copies/ml as compared with those with viral load < 6 log copies/ml (OR=4.93, 95% CI: 2.38-10.21). Conclusions: The HCV MTCT incidence is increased in pregnant women with detectable HCV viremia, and is even higher in those with viral load ≥ 6 log copies/ml. HIV coinfection further increases the risk of HCV MTCT. The 6 log copies/ml of maternal HCV viral load may be considered as the threshold for MTCT. Further randomized controlled trails are needed to conrm the indication of DAA treatment in pregnant women.


Background
Hepatitis C virus (HCV) infection is a major global health issue which affect 2-3% of population in the world 1 . HCV infection causes acute and chronic hepatitis, and increases the risk of liver cirrhosis, hepatocellular cancer, liver failure 2,3 . In the last decade, the prevalence of HCV in reproductive aged women was increasing own to opioid use 4 , resulting in a dramatic rise of HCV infection among pregnant women 5 . HCV viremia were detected in 30-40% infected neonates within the rst three days of life, indicated that intrauterine transmission may occur 6 . Therefore, mother-to-child transmission (MTCT) was considered to be the leading cause of hepatitis C in the future 7 , and current guidelines recommend universal HCV screening during pregnancy 8, 9 .
Antiviral treatment is considered as proven interventions to decrease the risk of perinatal transmission and neonatal infection, such as human immunode ciency virus (HIV) and hepatitis B virus infection during pregnancy 10,11 . For HCV antiviral treatment, direct-acting antiviral (DAA) regimens are able to reduce HCV viremia to undetectable level within two weeks 12,13 . Although it was recently reported effective to reduce HCV-RNA titer in pregnant women with DAA regime 14 , there are no epidemiologic studies or clinical trials have con rmed the reduction of HCV viral load in pregnancy contributes to the prevention of MTCT.
Although with some debates, HCV viremia has been considered as a main risk factor of MTCT during pregnancy 15 . Existing studies show when the viral load greater than or equal to 6 log10 copies/ml, the risk of MTCT is higher than women with a lower viral load [16][17][18] . Placenta express multiple HCV-receptors and/or entry cofactors (e.g., claudin-1, occludin, SR-B1, LDLr or DC-SIGN), resulting in HCV transmission of placenta barrier and fetal infection 19,20 . In addition, exposure to maternal infected blood during birth is another important pathway of HCV vertical transmission 21 , potentially because of peripheral mononuclear cells infection of HCV 20 . However, some clinical studies did not support this hypothesis 22,23  Injection drug use and the opioid epidemic is the leading cause of both HCV and HIV infection in the last decades 24 , and HIV/HCV co-infection is frequently observed in HCV pregnant women 25 . The most recent systematic review and meta-analysis showed maternal HIV coinfection is a main risk for MTCT of HCV.
HIV down regulates the immune responses, in particular of the T cell mediated reactivity with expansion of Treg cells 19 . In immunode ciency status, HCV replication may be increased and therefore facilitate transmission 7,26 , and it is suggested that HIV antiviral treatment may prevent HCV MTCT. However, whether HCV-viremia mono-infection or HCV-viremia coinfection of HIV increases MTCT of HCV has not been fully clari ed.
In this study, therefore, we aim to compare the risk of MTCT of HCV between pregnant women with and without HCV viremia detected in circulation, and to study how HCV viral load associated with MTCT did. In addition, the relationship of HIV coinfection with MTCT was investigated in the current meta-analysis. To minimize differences between studies, we imposed the following methodological restrictions for the inclusion criteria: (i) study designs: cohort studies or case-control studies, (ii) including data on both anti-HCV positive mothers and their children with documented HCV serology, (iii) comparing MTCT incidence between HCV viremia women and those with negative HCV-RNA. MTCT was con rmed when HCV-RNA was detected in newborns. In instances of multiple publications, the most up-to-date or comprehensive information was used. Citations initially selected by systematic search were rst retrieved as a title or abstract and preliminarily screened. Potentially suitable citations were then retrieved as complete manuscripts and assessed for compliance to inclusion criteria.

Data extraction
Articles were reviewed and crosschecked independently by two authors. Data on the following characteristics were independently extracted: rst author, publication year, county, study type, year of enrollment, number of mother-child pairs, maternal age of delivery, HCV-RNA detection method, number of live births, duration and frequency of newborns follow-up, HCV MTCT incidence, maternal HCV-RNA titer, and crude numbers of newborn with HCV infected and not. HCV MTCT incidence was de ned as the ratio of the number of infected newborns to the total number of newborns. High level of viremia was de ned as the HCV-RNA titers greater than or equal to 6 log10 copies/ml 16 . The most comprehensive data were included if multiple articles were reported on the overlapping study population. Any disagreements were resolved by consensus.

Quality assessments
We used the Newcastle-Ottawa Scale (NOS) to evaluate the methodological quality of the included observational studies. NOS includes three dimensions to assess the overall quality and a maximum of 4, 2, and 3 stars can be rated for selection, comparability, and outcome assessment respectively. Two authors independently assessed the risk of bias of included studies. Study scored ≤5 on the NOS was regarded as being of low quality and was excluded 27 .

Statistical analyses
We assessed the risk of the HCV MTCT by conducting two meta-analysis to calculate the odds ratios (ORs) and 95% con dence intervals (CIs) for individual studies and then pool the estimates using the DerSimonian and Laird random-effects models. First, we compared the HCV MTCT risk between HCV viremia women and those with negative HCV-RNA. Then, we performed a meta-analysis for relationship between HCV-RNA titers and HCV MTCT incidence.
Multiple-subgroup analysis among maternal HCV viremia/HIV+ group, HCV viremia mono-infected group and HCV-RNA negative group was further conducted after the comparison between HCV viremia and negative HCV-RNA women. In order to control the increased type I errors triggered by multiple comparison 28 , Holm-Bonferroni method 29 was used to gure out adjusted P-value.
We used the χ 2 -based Q test to assess heterogeneity across studies, expressed as Cochran's Q and I 2 statistics, together with 95% con dence intervals. We also determine the possibility of publication bias by using the funnel plot, Begg's test, and Egger's test. Sensitivity analyses were further conducted by only included prospective studies with a sample size larger than 100. Subgroup analysis was conducted by country income according to the World Bank's list of economies 30 , and region of different incidence of HCV infection according to the WHO Global Hepatitis Report 23 . Lower middle income countries and upper middle income countries were further aggregated into middle income countries. We also undertook meta-regression by publication year and MTCT incidence.
All statistical analyses were performed using the software R version 4.0.2.

Characteristics of included studies
We identi ed 717 papers and 202 of which were removed due to duplications. Finally, 26 publications, published between 1993 and 2015, were eligible for inclusion ( Figure 1). The studies examined 4491 newborns of HCV infected mother, with a follow up of at least 6 months. There were 3 retrospective cohorts and 23 prospective cohorts. Of the primary studies, 100% had described independent, consecutive sampling of their cohort. All studies detected HCV-RNA by polymerase chain reaction, and the MTCT incidence ranged from 1.3-38.5%.
Of the twenty-six studies, twenty-three independent eligible studies evaluated the risk of HCV MTCT between maternal HCV viremia group and negative HCV-RNA group (Table 1). Thirteen independent eligible studies evaluated the risk of HCV MTCT between HCV viremia/ HIV+ co-infected mothers and HCV viremia mono-infected mothers (Table 1). Nine independent eligible studies estimated the risk of HCV MTCT based on different HCV-RNA titers ( Table 2).   All of the 26 studies included in the present meta-analysis were regarded as acceptable quality, with NOS scores between 6 and 9. (supplementary Table 2, Table 3.)
Heterogeneity was not presented in the included studies (Q = 8.21, P = 1.000; I 2 = 0.00%). The pooled OR of MTCT in maternal HCV viremia compared to negative HCV-RNA group was 8.14 (95% CI, 4.62-14.31; P < 0.0001) by random-effects model ( Fig. 2A). A sensitivity analysis including only prospective studies with a sample size larger than 100 showed the pooled OR of MTCT was 8.83(95% CI: 4.18-18.65), indicating good robustness of the results (supplementary appendix Fig. 1A) The funnel plot showed symmetric distribution which indicated the absence of publication bias (supplementary appendix Fig. 2A). However, Begg's test (P = 0.010) and Egger's test (P = 0.357) suggested the existence of publication bias.
Multiple-subgroup analysis was conducted among maternal HCV viremia/HIV+ group, HCV viremia mono-infected group and HCV-RNA negative group. Meta-regression analysis indicated that publication year (Z = 0.13, P = 0.897) and MTCT incidence (Z = 0.92, P = 0.359) were not correlate to HCV MTCT risk.

Discussion
The current meta-analysis, involving 4382 newborns with maternal HCV antibody positive, found a signi cantly increased risk of HCV MTCT in pregnant women with HCV viremia, and an even higher MTCT incidence in pregnant women with high level of viremia. In addition, we found that HIV coinfection further increase risk of MTCT in women with HCV viremia.
In spite of the clear association of HIV co-infection and HCV MTCV during pregnancy, HCV viremia was been con rmed as an independent risk factor HCV transmission according to the result of the present meta-analysis. Although HCV MTCT was found in 4 out of 1291 newborns with maternal HCV-RNA negative, this could be explained by variation of HCV RNA titers during pregnancy, or false negative results due to low sensitivity of testing kits 33,42,49,51 . In addition, mothers with HCV viremia titer greater than or equal to 6 log copies/ml were with even higher incidence of MTCT than those with HCV viremia titer lower than 6 log copies/ml. It indicates pregnant women with HCV viremia should be considered for antiviral treatment and 6 log copies/ml coule be considered as the threshold for who is proposed to receive antiviral treatment. DAA treatment is an effective and safe strategy to reduce HCV RNA viral load in patient with chronic HCV infection 55 . Recently, DAA treatment with Ledipasvir and sofosbuvir regimens showed its effectiveness among pregnant women 14 .
However, up to 79% women refused to receive DAA treatment during pregnancy, concerning the safety and cost of DAA 56 . The ndings of this study suggested the emergency of DAA treatment for HCV infect pregnant women, and strengthening health promotion among pregnant women and reducing the cost of DAA are future measures to be taken.
Maternal HIV coinfection is considered as another risk factor of MTCT 57 . In this meta-analysis, pregnant women co-infected of HIV experience signi cantly highest prevalence of HCV-MTCT. This result is consistent with the report by Benova et al 57 . The role of HCV/HIV coinfection in HCV transmission seems to depend on HCV-RNA titer, since women taken HAART therapy, by reducing HIV-RNA level, also reduces the risk of HCV-MTCT 18 . Whereas, for the same level of HCV-RNA titer, the risk of HCV MTCT is probably higher in HCV/HIV co-infected women than HCV viremia mono-infected women according to the present meta-analysis. This suggests that other mechanisms could be involved to explain HCV-MTCT. Further studies are necessary to explore the interaction between HCV and HIV. Our ndings suggested for HIV/HCV co-infected pregnant women antiviral treatments for both HIV and HCV virus could be considered; however, further clinical trials are needed to con rm the safety and effectiveness of the combined treatment.
Inevitably, the present meta-analysis has some limitations. First, because of data source, we were only able to include cohort study for analysis. Second, due to the heterogeneity of studies regarding to HCV viral load, the dose-response relationship between HCV RNA and MTCT could not be obtained. In addition, the difference of MTCT among diverse HCV gene type is not clear, because the gene type of HCV was not available in literature 14 . Although with these limitations, our study is the rst meta-analysis to determine the association between HCV viral load and MTCT during pregnancy.

Conclusion
In conclusion, detectable HCV viremia is an independent factor of MTCT in pregnant women. HCV/HIV co-infection can further increase the risk of HCV-MTCT.
The 6 log copies/ml of maternal HCV viral load may be considered as the threshold for MTCT. Further randomized controlled trails are needed to con rm the indication of DAA treatment in pregnant women.

Declarations
Availability of data and materials All data can be obtained from the manuscript.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Figure 2
A Forest plot of MTCT risk of maternal HCV viremia group compared with HCV-RNA-negative group. MTCT= mother-to-child transmission, HCV=Hepatitis C virus B Forest plot of MTCT incidence for maternal HCV-RNA titer ≥6 log copies/mL Compared with <6 log copies/ml. MTCT= mother-to-child transmission,