Pregnant women in South Africa suffer from HIV and syphilis infections resulting in negative pregnancy outcomes. Little is known about the prevalence, incidence, seroconversion, and associated risk factors for those attending a midwife run obstetric unit.
A retrospective cohort study was undertaken among pregnant women attending antenatal clinic from January to December 2018. Logistic regression was conducted to determine the risk factors for HIV and syphilis.
The prevalence of HIV and syphilis were 44.3% (95% confidence interval [CI]; 41.6:46.7) and 3.8% (95% CI; 3.1:4.1), respectively. The seroconversion and incidence for HIV were 4.0% (95% CI; 3.6:4.6) and 17.1 per 100 person-years, and for syphilis 2.6% (95% CI; 2.3:2.8) and 10.9 per 100 person-years, respectively. Significant predictors for HIV prevalence were ages: ages < 20 years, Odds ratio (OR) = 0.11 (
The present study found high rates of seroprevalence, seroconversion and incidence of HIV and syphilis among pregnant women.
Human Immunodeficiency Virus (HIV) and syphilis infections are continuing as a public health concern in developing countries, especially in South Africa (SA). In SA, there are over 7.5 million people living with HIV, causing the highest disease burden in the world, of which 4.7 million women are aged 19 years and above.
More than 95% pregnant women in SA attend the public health facilities for antenatal care (ANC) and receive ART, which also works as a preventive measure for mother-to-child transmission of HIV.
To monitor the temporal trends of HIV epidemic and syphilis, the National Department of Health SA (DOH) undertakes National Antenatal Sentinel HIV and Syphilis seroprevalence Surveys among pregnant women. The last survey conducted in 2017 highlights different rates of HIV prevalence (national 30.7%) and a great variation between SA provinces exist. The KwaZulu-Natal province (KZN) has the highest rate of 41.1%.
A retrospective cohort study was undertaken among the antenatal clinic attendees using the clinic register. All pregnant women who attended for booking visits between January and December 2018 were included and followed up until their last visits to the clinic.
This study was undertaken at a Primary Health Care (PHC) set up of Kwadabeka Community Health Centre (KCHC). It provides first level care to over 150 000 predominantly Zulu-speaking black African populations living in the outskirts of Durban, which is known to be a peri-urban community. The residences are reliant on KCHC as a public health facility that provides free services to its clients. Since 2015, antiretroviral treatment (ART) has been made available at KCHC for all HIV-positive pregnant women.
Data were collected from the antenatal clinic register (ACR). The manual ACR was developed and implemented by the national department of health that catered for the recording of selective ANC findings of all (booking and follow up) visits in a structured format.
All pregnant women were counselled and given information about antenatal serological screening tests at their first and follow-up visits. The routine practice at KCHC is that all mothers are screened for anaemia, HIV, syphilis, and rhesus factors and recorded the results in the ACR. The syphilis screening test at the booking and the rescreening in the third trimester or later visits were recorded as ‘reactive or non-reactive’. In accordance with the DOH policy on HIV screening and treating purposes, finger prick blood samples were tested with two rapid tests of two different suppliers. If the first test became positive, a second rapid test using a different test kit from a different supplier was performed. If the first test became positive, a second rapid test using a different test kit from a different supplier was performed. If both the tests became negative, the mother was considered HIV negative. If both the test results were positive, the mother was considered HIV positive. If one test was negative, then a confirmatory Enzyme-Linked Immunosorbent Assay (ELISA) test was performed at the laboratory. Every confirmed positive HIV woman was recorded as HIV status positive and offered ART using highly active ART (HAART) on the same day as part of the universal ART programme in SA should the client accepted ART. Those pregnant women who had HIV negative results were rescreened after 12 weeks or later when they became available for a repeat ANC visit. The Rapid Plasma Reagin (RPR) test was used to yield reactive or non-reactive results for syphilis. For the consideration of syphilis, the occurrence of one of these situations was considered: a record of a reactive result in the ACR and or the mother was treated with penicillin, as it was ‘syphilis positive or exposed’ (case definition). Those women who had non-reactive RPR results were re-screened at or after 32 weeks of GA, and the results recorded accordingly.
Data were captured in Microsoft Excel (for Windows) and exported to Statistical Package for the Social Sciences (SPSS) (version 22.0) for analysis. Ages of the pregnant women were categorised into; < 20 years (teenage), 20–24 years, 25–29 years, 30–34 years and ≥ 35 years, GA into first trimester (0–13 weeks), second trimester (14–27 weeks) and third trimester (≥ 28 weeks). Parity was categorised as nil-parity (primipara or primigravida), parity 1–2, 3–4 and ≥ 5 (grandmultiparity). The screening results of HIV and syphilis at the booking and subsequent visits were used to estimate the prevalence of HIV and syphilis. The 12 weeks or later follow-up visit retest data were used for the estimation of seroconversion and incidence of HIV. For syphilis, the rescreening results at or around 32 weeks GA follow-up visits were used for estimation of seroconversion and incidence of maternal syphilis. We had calculated total person years followed up for repeat HIV and syphilis tests and measured incidence of HIV and syphilis per 100 person-years. Categorical variables were presented using frequencies and proportions. Differences in proportions of syphilis and HIV for different demographic and obstetric variables at the booking visit were examined using Pearson chi-square (χ2) and
Umgungundlovu Health Ethics Review Board approved the study protocol (Reference no. UHERB 015/2020). Written permission from the Management of the KCHC was obtained to use the AR for the study. Informed consent was waived as we used secondary data. Full confidentiality and privacy were maintained to present the report.
A total of 1503 pregnant women had booking visits at KCHC for the study period. The information on ANC booking visit and follow-up at ANC clinic of pregnant women are shown in
Flow diagram of study subjects at different stages of analysis.
Demographic, obstetric and outcome variables of the study sample at booking visit.
Variables | Frequency | % | 95% CI |
---|---|---|---|
< 20 | 271 | 18.2 | - |
20–24 | 466 | 31.3 | - |
25–29 | 391 | 26.2 | - |
30–34 | 248 | 16.6 | - |
≥ 35 | 115 | 7.7 | - |
Up to 13 | 190 | 13.9 | - |
14–27 | 854 | 62.6 | - |
≥ 28 | 321 | 23.5 | - |
Nil | 461 | 30.8 | - |
1–2 | 831 | 55.6 | - |
3–4 | 186 | 12.4 | - |
≥ 5 | 17 | 1.1 | - |
Negative | 837 | 55.7 | - |
Positive | 665 | 44.3 | 41.6–46.7 |
662 | 99.5 | - | |
Yes | 32 | 4.0 | 3.6– 4.6 |
HIV incidence (per 100 person-year) | 32 | 17.3 | 11.2–22.1 |
Negative | 1441 | 96.8 | - |
Positive | 58 | 3.8 | 3.1–4.1 |
Positive (Seroconversion) | 29 | 2.6 | 2.3–2.8 |
Syphilis incidence (per 100 person-year) | 29 | 10.9 | 7.1–14.6 |
HIV, human immunodeficiency virus; RPR, Rapid Plasma Reagin; ART, anti-retroviral treatment; CI, confidence interval.
Cross table analysis of syphilis and human immunodeficiency virus prevalence with demographic and obstetric variables at the booking visit with Chi-square and
Variables | Syphilis positive |
HIV status: Positive |
||||
---|---|---|---|---|---|---|
% | % | |||||
- | - | - | - | - | 0.001 | |
< 20 | 4 | 1.5 | 0.001 | 57 | 21.0 | - |
20–24 | 13 | 2.8 | - | 155 | 33.3 | - |
25–29 | 12 | 3.1 | - | 201 | 51.7 | - |
30–34 | 13 | 5.2 | - | 160 | 64.5 | - |
≥ 35 | 14 | 12.2 | - | 87 | 75.7 | - |
- | - | - | - | - | 0.036 | |
Up to 13 | 6/190 | 3.2 | 0.255 | 85 | 44.7 | - |
14–27 | 31/854 | 3.6 | - | 349 | 41.1 | - |
≥ 28 | 16/321 | 5.0 | - | 158 | 49.2 | - |
- | - | - | - | - | 0.001 | |
Nil parity | 14/461 | 3.0 | 0.004 | 142 | 30.8 | - |
1–2 parity | 27/831 | 3.2 | - | 394 | 47.4 | - |
3–4 parity | 15/186 | 8.1 | - | 114 | 61.3 | - |
≥ 5 parity | 2/17 | 11.2 | - | 10 | 58.8 | - |
- | - | - | - | - | - | |
Negative | 14/837 | 2.5 | 0.002 | - | - | - |
Positive | 34/665 | 5.6 | - | - | - | - |
- | - | - | - | - | 0.001 | |
No | - | - | - | 3 | 0.5 | - |
Yes | - | - | - | 662 | 95.5 | - |
- | - | - | - | - | 0.002 | |
Negative | - | - | - | 62.8 | 43.5 | - |
Positive | - | - | - | 37 | 63.8 | - |
- | - | - | - | - | 0.076 | |
Negative | - | - | - | 12 | 1.9 | - |
Positive | - | - | - | 17 | 3.5 | - |
HIV, human immunodeficiency virus; RPR, Rapid Plasma Reagin; ART, anti-retroviral treatment.
Regression analysis (
Logistic regression output for human immunodeficiency virus positive at the booking visit.
Variables | Adjusted OR | 95% CI for OR |
||
---|---|---|---|---|
Lower | Upper | |||
< 0.010 | - | - | - | |
< 20 | < 0.010 | 0.119 | 0.062 | 0.379 |
20–24 | < 0.010 | 0.199 | 0.119 | 0.332 |
25–29 | < 0.010 | 0.389 | 0.233 | 0.650 |
30–34 | 0.096 | 0.633 | 0.369 | 1.084 |
≥ 35 | - | 1.0 | - | - |
0.024 | - | - | - | |
Up to 13 | 0.364 | 0.835 | 0.566 | 1.232 |
14–27 | 0.007 | 0.682 | 0.515 | 0.902 |
≥ 28 | - | 1.0 | - | - |
0.009 | - | - | - | |
Nil parity | 0.816 | 0.875 | 0.285 | 2.693 |
1–2 parity | 0.706 | 1.237 | 0.409 | 3.741 |
3–4 parity | 0.360 | 1.704 | 0.544 | 5.331 |
≥ 5 parity | - | 1.0 | - | - |
0.031 | 0.458 | 0.227 | 0.929 | |
- | 1.0 | - | - | |
0.008 | 6.114 | - | - | |
Reference group ≥ 35 years, Parity ≥ 5, GA ≥ 28 weeks and Syphilis positive. CI, confidence interval; OR, odds ratio. |
Logistic regression output for incidence or seroconversion of human immunodeficiency virus infection.
Variables | Adjusted OR | 95% CI for OR |
||
---|---|---|---|---|
Lower | Upper | |||
0.077 | - | - | - | |
< 20 | 0.012 | 0.127 | 0.025 | 0.639 |
20–24 | 0.015 | 0.176 | 0.044 | 0.709 |
25–29 | 0.090 | 0.339 | 0.097 | 1.182 |
30–34 | 0.103 | 0.297 | 0.069 | 1.276 |
≥ 35 | - | 1.0 | - | - |
Reference group ≥ 35 years.
CI, confidence interval; OR, odds ratio.
At the booking visit, 58 pregnant women had RPR reactive results thus estimating the prevalence of syphilis at 3.8% (95% CI; 3.1%:4.1%). All of them (58) received their first dose of penicillin treatment for being syphilis reactive. When we examined RPR results at the follow up visits (
Ages of the pregnant women were significantly associated with positive syphilis status (
Logistic regression output for syphilis at booking visit.
Variable | OR | 95% CI for OR |
||
---|---|---|---|---|
Lower | Upper | |||
< 0.010 | - | - | - | |
< 20 | < 0.010 | 0.119 | 0.037 | 0.379 |
20–24 | < 0.010 | 0.213 | 0.092 | 0.489 |
25–29 | 0.001 | 0.442 | 0.171 | 0.554 |
30–34 | 0.044 | 0.523 | 0.226 | 0.959 |
≥ 35 | - | 1.0 | - | - |
0.022 | 0.444 | 0.221 | 0.891 | |
- | 1.0 | - | - | |
< 0.010 | 0.102 | - | - |
Reference group ≥ 35 years, HIV positive.
CI, confidence interval; OR, odds ratio.
Logistic output for incidence or seroconversion of syphilis.
Variables | Significance ( |
Adjusted OR | 95% CI for OR |
|
---|---|---|---|---|
Lower | Upper | |||
0.020 | - | - | - | |
0–12 | - | 1 | - | - |
13–26 | 0.178 | 3.983 | 0.532 | 29.829 |
> 27 | 0.035 | 9.212 | 1.172 | 72.407 |
0.000 | 0.013 | - | - |
Reference group gestational age 0–12 weeks.
CI, confidence interval; OR, odds ratio.
The prevalence of maternal syphilis in our study is high (3.8%) at the booking visit. This is higher than the rates reported by the national surveys for SA (nationally 1.5% and 2%) and KZN (0.4% and 2.3%) during 2011 and 2015, respectively.
Similarly, the higher incidence of STI is estimated at 20 per 100 women years from Durban.
The result of syphilis seroconversion and incidence rates should be interpreted with caution as not every pregnant woman who attended KCHC for their first antenatal visit and who was syphilis negative had a retest done at or around 32 weeks of gestation. The follow up rate was 77%. It was possible that healthcare workers might overlook or forget to conduct a rescreening test for syphilis and or recording in the ACR. This may be a reason why some women are missing syphilis rescreening results. A possible explanation could be that the pregnant women could have experienced an array of complications during the antenatal period and would thus have been referred to a higher care institute for antenatal and delivery care. It might also be possible that some pregnant women who attended KCHC on their booking ANC visits either changed to different healthcare facilities, experienced loss of pregnancy or delivered prematurely. The actual rates of pregnancy complications and reasons for referrals to hospital and lost to follow-up during the antenatal period were not assessed in this study. However, it is known that approximately 60% – 70% of all pregnant women who utilise a PHC facility in public health settings in SA would most likely require the service of a hospital at some point in their antenatal period.
The present study found a significant association between the age of the pregnant women and both HIV and syphilis infections among these pregnant women. Older pregnant women had a higher rate of these infections compared to younger ones. Reports from other parts of Africa had shown that the prevalence of syphilis was found to increase from 0.6% in the younger age group to 1.1% in the older women in Rwanda, 0.9% in the younger women which increased to 3.7% in the older women in Uganda and in Kenya where the rate was 0.9% in the younger as compared to 2.5% in the older pregnant women.
Similar to syphilis, the prevalence of HIV among younger pregnant women (age ≤ 29 years) are found to decline from 21.8% in 2010 to 18.5% in 2017 in SA.
This co-infection rate is considered lower than the rate of 10% reported from the multi-country study conducted in SA, Brazil, Argentina and the USA.
We estimated a higher prevalence of HIV (44.3%) among our study population. This is higher than the rate estimated for KZN (41.1%) in 2017 during the annual survey.
We found that pregnant women who are syphilis negative, are 55% (OR = 0.459,
Therefore, it is seen that the prevalence of syphilis infection can increase HIV infection in pregnant women. It is already known that the syphilis infection can increase HIV transmission to the foetus two to four-fold, as in mothers infected with syphilis is significantly associated with HIV transmission from mother to the unborn child.
Analysis of risk factors for HIV had shown that older age of pregnant women, advance in pregnancy (last trimester), higher parity and lastly the infection with syphilis are associated with higher rate of HIV in our study population. These higher rates of syphilis and HIV infection among pregnant women suggest that these infections continue to exist as an important public health concern with the hypothesis that these prevalence rates do vary at different geographical areas and may have masked the national and provincial rates. Thus, there is a need to strengthen intervention efforts at the institutional and or community levels.
This study has some limitations namely, the retrospective nature and limited study variables for risk factors, errors in data abstraction, selection bias because of loss to follow up. We only studied those pregnant women who attended our ANC clinic. Our rates might have underestimated or overestimated the prevalence of syphilis and HIV as a significant number of women were missing because of inaccessibility to the ANC clinic or as a result of personal, behavioural or socio-economic (cost to travel) reasons. However, it is known that a high proportion (> 95%) of pregnant women do attend ANC at least once in KZN.
Our study finds higher prevalence, incidence and seroconversion rates of HIV and syphilis among pregnant women. Maternal age, parity, and HIV status were found to be risk factors for syphilis. Similarly, maternal age, GA, parity, and syphilis were risk factors for maternal HIV. Higher rates of seroconversion and incidence of HIV and syphilis suggest that these women may potentially transmit these infections to the unborn babies and affect the pregnancy outcomes negatively. The strategies should be strengthened to local areas and health facilities including counselling and testing, ART and health education to change the behaviour on prevention of HIV and syphilis among pregnant women in particular and the general population at large. Further studies are recommended to identify the behavioural factors of pregnant women related to higher rates of HIV and syphilis infections among high-risk groups.
We thank all practising midwives of Kwadabeka Community Health Centre working in the antenatal clinic for recording study variables in ACR, Muqeet Hoque who captured data in Microsoft Excel and Maariyah Hoque for coding and assisting in the analysis of data for the report.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
M.H. contributed to the conceptualisation, study design, participation in data collection, data analysis, preparation, and finalisation of the manuscript. S.B. contributed to writing the results, editing and finalisation of the manuscript. M.E.H. and G.v.H. contributed to the conceptualisation and finalisation of the manuscript. All the authors read and approved the final manuscript.
The authors received no funding from any agency in the public, commercial or not-for-profit sectors.
The data that support the findings of this study are available from the corresponding author, M.E.H., upon reasonable request.
The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy of any affiliated agency of the authors.