INTRODUCTION
Syphilis and HIV present public health challenges that particularly affect resource limited settings including up to 1.4 million pregnant women worldwide.1–7 Both can be transmitted by mother-to-child transmission during pregnancy and at birth, leading to infection in the infant.4,8–11 Maternal infection may also lead to adverse birth outcomes including stillbirth, low birth weight, and neonatal death.4,8,12–14 If untreated, greater than 50% of syphilis-affected pregnancies result in adverse events such as mid–late trimester fetal losses, stillbirths, low birth weight, preterm delivery, or congenital syphilis.15–18
Most African countries have low screening rates for syphilis in pregnancy.7,17,19–22 However, Botswana is one of the few countries with highly prevalent HIV23,24 and syphilis,25,26 with high rates of screening for both, allowing evaluation for risks associated with coinfection. We used data from birth outcomes surveillance performed in 8 healthcare facilities across Botswana to describe syphilis and HIV trends among antenatal women between 2008 and 2016, and their impact on adverse birth outcomes.
MATERIALS AND METHODS
Study Population
This was a secondary data analysis of an ongoing birth outcomes surveillance study in Botswana, the “Tsepamo Study,” which aims to study the safety of antiretroviral therapy and HIV on birth outcomes including stillbirth, preterm delivery, small for gestational age, and congenital abnormalities.27–30 Data were abstracted from the maternal obstetric records of all women delivering at government maternity wards during 2 time periods. From 2008 to 2011 data were available from 5 sites (Mochudi, Francistown, Kanye, Maun, and Ramotswa). From 2014 to 2016, data were available from 8 surveillance sites. Of the 8 total sites, 2 were at tertiary referral hospitals (Gaborone and Francistown) and the rest were at district or primary hospitals (Ghanzi, Maun, Mahalapye, Molepolole, Selebi-Phikwe, and Serowe). Births at these sites were estimated to represent approximately 45% of all births in the country.31
More than 95% of women in Botswana32 deliver in a maternity ward and 99.8% have their maternity obstetric card available for data abstraction at the time of discharge from the ward. This study included all pregnant women with data available who attended at least one antenatal visit; had at least one rapid plasma reagin (RPR) test result; and had an HIV test result recorded in pregnancy. Women with multiple pregnancies were excluded from analysis, as were those who delivered before arrival at healthcare facilities or before 24 weeks of gestation.
All laboratory testing and antenatal visits were performed by Botswana government antenatal care nurses as part of routine care. The services received during the antenatal period include testing for HIV hemoglobin and syphilis. HIV rapid testing is done at the first antenatal visit and repeated at week 36 of gestation and in labor if previous result is HIV negative.33 Screening for syphilis through RPR testing is recommended at 2 different time points; at booking/registration for antenatal clinic, and at week 34–36 gestation. Women with a reactive RPR are treated for syphilis with 3 weekly doses of benzathine penicillin (or with azithromycin/erythromycin in women with penicillin allergy).33
Data Collection
Deidentified data were abstracted from obstetric cards at the time of discharge from the postnatal wards. Data collected included maternal demographics, medical history, medications administered during the pregnancy, smoking and alcohol use, maternal blood pressure, weight, gestational age at first antenatal care visit, results for HIV testing, RPR, and hemoglobin levels. If HIV-infected, the obstetric card captures date of diagnosis, most recent CD4 cell count value and antiretroviral drugs prescribed. For the infant, the obstetric record captures gestational age at delivery, APGAR score, birth weight, length at birth, head circumference, congenital anomalies, and other examination findings at time of birth and discharge.
Exposure Definitions
All women with reactive RPR were considered to be syphilis-infected. Women who received antibiotics in pregnancy, which are known to treat syphilis (penicillin, erythromycin, azithromycin, or ceftriaxone) were considered to be treated. Any positive rapid HIV test or documentation of a previous positive HIV test was considered to indicate HIV infection.
Outcome Definitions
Outcomes were stillbirth, preterm delivery (<37 weeks of gestational age), low birth weight (<2500 g), and in-hospital neonatal death (<28 days). Estimated gestational age was determined at the time of delivery by the government maternity nurses, using last menstrual period as reported by the mother at initial antenatal care visit and/or ultrasound scan in pregnancy when available. Fundal height estimates were rarely used in instances when last menstrual period or ultrasound were not available. Stillbirths included both fresh stillbirths and macerated stillbirths (APGAR 0,0,0). Neonatal death was defined as any death in an infant less than 28 days of life who had not been discharged out of hospital (and therefore could be captured by the in-hospital surveillance).
Statistical Analysis
Statistical analysis was performed using STATA (version 13.0). Frequencies and proportions were used to describe maternal characteristics, HIV and syphilis status, and birth outcomes. We used univariate and multivariable logistic regression techniques to describe adverse birth outcomes by HIV and syphilis status. Birth outcomes in HIV syphilis coinfection were also compared with birth outcomes in HIV infection alone and in syphilis infection alone. For the final adjusted model, we included univariate factors P < 0.2. We examined effect modification using the Mantel–Haenszel analysis to determine the potential interaction of HIV and syphilis on birth outcomes.
Ethical Approval
Study approval for the Birth Outcomes Surveillance studies were obtained from the Health and Research Development Committee in the Botswana Ministry of Health and the Office of Human Research Administration in Harvard T. H. Chan School of Public Health.
RESULTS
From 2008 to 2011 and from 2014 to 2016, a total of 76,466 women delivered at all the surveillance sites; 31,590 (41.3%) in 2008–2011, and 44,876 (58.7%) 2014–2016. In the 2 study periods, the median age of the pregnant women was 26 years (interquartile range 21–31 years) (Table 1). Most women were unmarried (88%) and were educated up to secondary school (70%). About half of the women (53%) were unemployed (Table 1). There were no significant demographic and/or geographic differences between study periods (data not shown).
Seroprevalence of HIV Infection, Syphilis, and Coinfection
HIV test results were available for 75,770 (99.1%) of the pregnant women, and 20,520 (27.1%) were HIV infected. Of the HIV-infected participants, 18,741 (91.4%) had received an antiretroviral regimen in pregnancy. Between 2008–2011 and 2014–2016, the proportion of women with HIV decreased from 29.4% to 25.4% (P < 0.00001). Syphilis test results were available for 67,290 (88.0%) women, and 697 (1.0%) had reactive RPR. Between 2008–2011 and 2014–2016, the proportion of women with syphilis remained constant (1.1% vs. 1.0%, P = 0.41).
Among only HIV-positive women, 261 (1.5%) of 17,806 tested had reactive RPR. Similarly, among 692 women with syphilis who also had an HIV test result, 261 (37.7%) were HIV infected. HIV-infected women were more likely to be infected with syphilis than HIV-uninfected women [odds ratio (OR) = 1.68; 95% confidence interval (CI): 1.44 to 1.96]. The prevalence of HIV/syphilis coinfection declined from 45% between 2008 and 2011 to 27% between 2014 and 2016 (P < 0.0001) (Table 1).
;)
TABLE 1.: Characteristics of Women Who Delivered Live Singleton Infants in the Two Study Periods, by HIV and Syphilis Status
Adverse Birth Outcomes
Table 2 shows adverse birth outcomes by maternal syphilis and HIV status. Further comparison was also performed with each individual sero-status (syphilis alone or HIV alone). We did not observe any significant effect modification by HIV status (results not shown) in the association between syphilis and birth outcomes.
TABLE 2.: Adverse Birth Outcomes by Syphilis and HIV Status
Stillbirth occurred in 5.8% of coinfected women, compared with 1.9% with no HIV/syphilis [OR = 3.09; 95% CI: 1.83 to 5.23]; 3.4% with HIV alone (OR = 1.75; 95% CI: 1.03 to 2.97); or 3.7% with syphilis alone (OR = 1.58; 95% CI: 0.77 to 3.25).
Low birth weight occurred in 24.1% of coinfected women, compared with 12.1% with no HIV/syphilis [OR 2.31; 95% CI: (1.74 to 3.08); 20% with HIV alone (OR = 1.27; 95% CI: 0.96 to 1.69]; or 14.6% with syphilis alone [OR = 1.85; 95% CI: 1.26 to 2.74].
Preterm delivery was significantly higher in HIV/syphilis coinfected women compared with women with syphilis alone in the adjusted analysis [adjusted OR (AOR) = 1.54; 95% CI: 1.05 to 2.27]. Delivery of low birth weight infants in HIV/syphilis coinfection was significantly more common compared with syphilis alone (AOR = 1.82; 95% CI: 1.21 to 2.75). Neonatal death of babies born to mothers with HIV/syphilis coinfection was not significantly different compared with babies born to women with syphilis alone (AOR = 1.71; 95% CI: 0.70 to 4.20).
Treatment
Just above half (411/697 or 59.0%) of women with a reactive RPR received antibiotic treatment during pregnancy. The most used antimicrobial was penicillin (47.9%), followed by ceftriaxone (24.7%) and erythromycin (21.1%). Of the 411 participants who were treated for syphilis in pregnancy, 172 (41.3%) had a result for a second RPR test performed during pregnancy; of these women, 101 (58.7%) had a negative RPR test.
DISCUSSION
We found that although syphilis and HIV coinfection has declined in the past decade, coinfected women had the highest risk for most adverse birth outcomes, compared with uninfected- or single-infected women. Syphilis screening in pregnancy appears to be higher in Botswana (88%) than many other African countries.16,18,34–37 The prevalence of syphilis in this population was 1%, and was constant across the 2 study periods. This is a significantly lower prevalence than reported from studies conducted in Botswana between 2005 and 2006,25,26 and lower than in other African countries in the region.6,9,38 This could be because of improved access to health care in the past few decades. These results could also indicate the benefits of HIV prevention and management strategies including early treatment of sexually transmitted infections and safer sex. There is also the possibility of HIV-related mortality having resulted in low prevalence of syphilis, because studies have confirmed a positive correlation between the 2.8,37
HIV prevalence is high in Botswana, particularly among pregnant women.31,32,39 In this study, we showed a 4% decrease in HIV prevalence among pregnant women between the 2 study eras, documenting encouraging improvements in HIV prevention (and perhaps the effect of HIV treatment) over time. The 1.3% prevalence of coinfection with syphilis is significantly lower than has been reported in other populations, where about 9.5% of HIV-infected adults are known to also have syphilis infection.40 This could be because of the general low syphilis prevalence in the country.
The prevalence of stillbirth was high among HIV and syphilis coinfected women, consistent with existing literature.8,12,19 High stillbirth despite high rates of screening could be due to delays in syphilis treatment, or to a lack of treatment effect. Furthermore, coinfected women were more likely to have preterm and low birth weight infants compared with women with syphilis only. The mechanism of the synergistic effect of HIV and syphilis infection on adverse birth outcomes is poorly understood, requiring further study.12,41–43
This study had a large sample size, making it well powered to reach conclusions despite the relatively low prevalence of syphilis and coinfection. Limitations of our study included reliance on medical record abstraction after delivery (although obstetric cards are generally complete). Treatment of syphilis may occur during pregnancy or in the postnatal period depending on when the diagnosis was made. We did not include data about syphilis treatment in the postnatal period. Furthermore, complete syphilis treatment data were only available for 2014–2016 and were not sufficient to make conclusions about the effectiveness and completeness of syphilis treatment in the antenatal period in Botswana. Neonatal deaths in this study were only recorded for infants who had not been discharged from the hospital after birth. This may under-represent neonatal deaths in the population. However, given that most women in Botswana deliver in a healthcare facility,32 most babies who are not thriving at birth are generally not discharged home. Finally, this was an observational study, so causality of adverse birth outcomes among HIV and syphilis coinfection cannot be established. Observational studies are more prone to bias and confounding,44,45 but a randomized study is not feasible for this question.
In summary, we have shown a high rate of adverse birth outcomes among HIV/syphilis coinfected mothers and overall decline in HIV/syphilis coinfection. The decline in HIV/syphilis coinfection could be because of widespread availability of antibiotics that treat syphilis, or factors associated with the changing epidemiology of HIV in Botswana. Further studies are warranted to elucidate the mechanism by which HIV and syphilis may increase the rate of adverse birth outcomes.
ACKNOWLEDGMENTS
We wish to thank the Botswana Ministry of Health and Study Staff at Botswana Harvard AIDS Institute Partnership. We also wish to thank Dr. Bhekiqiniso Shava for editing the manuscript.
REFERENCES
1. WHO. Dual Test for Syphilis and HIV 2018. Available at:
http://www.who.int/reproductivehealth/topics/rtis/syphilis/dual_test/en/. Accessed November 1, 2018.
2. Organization WH. WHO Guideline on Syphilis Screening and Treatment for Pregnant Women. Geneva, Switzerland: World Health Organization; 2017.
3. WHO. HIV/AIDS Fact Sheet. Geneva, Switzerland: World Health Organization; 2017.
4. WHO. HRP Annual Report. In: WHO, editor. WHO Department of Reproductive Health and Research, UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction (HRP); 2018.
5. Gomez GB, Kamb ML, Newman LM, et al. Untreated maternal syphilis and adverse outcomes of pregnancy: a systematic review and meta-analysis. Bull World Health Organ. 2013;91:217–226.
6. Newman L, Kamb M, Hawkes S, et al. Global estimates of syphilis in pregnancy and associated adverse outcomes: analysis of multinational antenatal surveillance data. PLoS Med. 2013;10:e1001396.
7. Gottlieb SL, Newman LM, Amin A, et al. Sexually transmitted infections and women's sexual and reproductive health. Int J Gynaecol Obstet. 2013;123:183–184.
8. Lynn WA, Lightman S. Syphilis and HIV: a dangerous combination. Lancet Infect Dis. 2004;4:456–466.
9. Lawi JD, Mirambo MM, Magoma M, et al. Sero-conversion rate of Syphilis and HIV among pregnant women attending antenatal clinic in Tanzania: a need for re-screening at delivery. BMC Pregnancy Childbirth. 2015;15:3.
10. Moura AA, de Mello MJ, Correia JB. Prevalence of syphilis, human immunodeficiency virus, hepatitis B virus, and human T-lymphotropic virus infections and coinfections during prenatal screening in an urban Northeastern Brazilian population. Int J Infect Dis. 2015;39:10–15.
11. WHO. WHO Guideline on Syphilis Screening and Treatment for Pregnant Women. Geneva, Switzerland: World Health Organization; 2017.
12. Acosta LMW, Goncalves TR, Barcellos NT. HIV and syphilis coinfection in pregnancy and vertical HIV transmission: a study based on epidemiological surveillance data [in Portuguese]. Rev Panam Salud Publica. 2016;40:435–442.
13. Mulu A, Kassu A, Tessema B, et al. Seroprevalence of syphilis and HIV-1 during pregnancy in a teaching hospital in northwest Ethiopia. Jpn J Infect Dis. 2007;60:193–195.
14. Niama RF, Loukabou Bongolo NC, Bayonne Kombo ES, et al. Syphilis and HIV infections among pregnant women attending antenatal clinics in Republic of Congo. Pan Afr Med J. 2017;28:8.
15. Watson-Jones D, Gumodoka B, Weiss H, et al. Syphilis in pregnancy in Tanzania. II. The effectiveness of antenatal syphilis screening and single-dose benzathine penicillin treatment for the prevention of adverse pregnancy outcomes. J Infect Dis. 2002;186:948–957.
16. Gloyd S, Chai S, Mercer MA. Antenatal syphilis in sub-Saharan Africa: missed opportunities for mortality reduction. Health Policy Plan. 2001;16:29–34.
17. Watson-Jones D, Changalucha J, Gumodoka B, et al. Syphilis in pregnancy in Tanzania. I. Impact of maternal syphilis on outcome of pregnancy. J Infect Dis. 2002;186:940–947.
18. Kahn JG, Jiwani A, Gomez GB, et al. The cost and cost-effectiveness of scaling up screening and treatment of syphilis in pregnancy: a model. PLoS One. 2014;9:e87510.
19. Melku M, Kebede A, Addis Z. Magnitude of HIV and syphilis seroprevalence among pregnant women in Gondar, Northwest Ethiopia: a cross-sectional study. HIV AIDS (Auckl). 2015;7:175–182.
20. Wilkinson D, Sach M, Connolly C. Epidemiology of syphilis in pregnancy in rural South Africa: opportunities for control. Trop Med Int Health. 1997;2:57–62.
21. Ratnam AV, Din SN, Hira SK, et al. Syphilis in pregnant women in Zambia. Br J Vener Dis. 1982;58:355–358.
22. Kuznik A, Lamorde M, Nyabigambo A, et al. Antenatal syphilis screening using point-of-care testing in Sub-Saharan African countries: a cost-effectiveness analysis. PLoS Med. 2013;10:e1001545.
23. Botswana UNDP. HIV and AIDS in Botswana. Botswana: United Nations Development Programme; 2018. Available at:
http://www.bw.undp.org/content/botswana/en/home/ourwork/hiv_aids/overview.html. Accessed: 01 November 2018.
24. (UNAIDS) JUNPoHA. UNAIDS Data Reference. Geneva, Switzerland: United Nations; 2017.
25. Romoren M, Rahman M. Syphilis screening in the antenatal care: a cross-sectional study from Botswana. BMC Int Health Hum Rights. 2006;6:8.
26. Creek TL, Thuku H, Kolou B, et al. Declining syphilis prevalence among pregnant women in northern Botswana: an encouraging sign for the HIV epidemic? Sex Transm Infect. 2005;81:453–455.
27. Zash R, Souda S, Chen JY, et al. Reassuring birth outcomes with Tenofovir/Emtricitabine/Efavirenz used for prevention of mother-to-child transmission of HIV in Botswana. J Acquir Immune Defic Syndr. 2016;71:428–436.
28. Caniglia EC, Zash R, Jacobson DL, et al. Emulating a target trial of antiretroviral therapy regimens started before conception and risk of adverse birth outcomes. AIDS. 2018;32:113–120.
29. Zash R, Jacobson DL, Diseko M, et al. Comparative safety of antiretroviral treatment regimens in pregnancy. JAMA Pediatr. 2017;171:e172222.
30. Zash RM, Souda S, Leidner J, et al. High proportion of deaths attributable to HIV among postpartum women in Botswana despite widespread uptake of antiretroviral therapy. AIDS Patient Care STDS. 2017;31:14–19.
31. Botswana S. Health Statistics 2009. Republic of Botswana, Southern Africa: Statistics Botswana; 2012. Report No. Available at:
http://www.statsbots.org.bw/sites/default/files/publications/Health%2520Statistics%2520Annual%2520Report_2009%5B2%5D.pdf. Accessed 31August 2018.
32. Botswana S. Botswana-maternal Mortality Ratio (MMR) 2009-2013. 2014 Contract No.: No. 2014/1; 2014. Available at:
http://www.statsbots.org.bw/sites/default/files/publications/BOTSWANA-Maternal%20Mortality%20Ration%202009-%202014.pdf. Accessed August 31, 2018.
33. Botswana MoH. Botswana National HIV & AIDS Treatment Guidelines. Gaborone, Botswana: Botswana Ministry of Health; 2012. Available at:
https://aidsfree.usaid.gov/sites/default/files/tx_botswana_2012.pdf. Accessed November 1, 2018.
34. Dionne-Odom J, Mbah R, Rembert NJ, et al. Hepatitis B, HIV, and syphilis seroprevalence in pregnant women and blood donors in Cameroon. Infect Dis Obstet Gynecol. 2016;2016:4359401.
35. Endris M, Deressa T, Belyhun Y, et al. Seroprevalence of syphilis and human immunodeficiency virus infections among pregnant women who attend the University of Gondar teaching hospital, Northwest Ethiopia: a cross sectional study. BMC Infect Dis. 2015;15:111.
36. Eticha BT, Sisay Z, Alemayehu A, et al. Seroprevalence of syphilis among HIV-infected individuals in Addis Ababa, Ethiopia: a hospital-based cross-sectional study. BMJ Open. 2013;3:1–6.
37. Kenyon CR, Osbak K, Chico RM. What underpins the decline in syphilis in Southern and Eastern Africa? An exploratory ecological analysis. Int J Infect Dis. 2014;29:54–61.
38. Bhat CJ, Ratnam AV, Chintu C, et al. Congenital syphilis in Lusaka—II: incidence at birth and potential risk among hospital deliveries. East Afr Med J. 1982;59:306–310.
39. Botswana S. Botswana AIDS Impact Survey IV; 2013.
40. Kalichman SC, Pellowski J, Turner C. Prevalence of sexually transmitted co-infections in people living with HIV/AIDS: systematic review with implications for using HIV treatments for prevention. Sex Transm Infect. 2011;87:183–190.
41. Salado-Rasmussen K. Syphilis and HIV co-infection. Epidemiology, treatment and molecular typing of Treponema pallidum. Dan Med J. 2015;62:B5176.
42. Tornatore M, Goncalves CV, Bianchi MS, et al. Co-infections associated with human immunodeficiency virus type 1 in pregnant women from southern Brazil: high rate of intraepithelial cervical lesions. Mem Inst Oswaldo Cruz. 2012;107:205–210.
43. Yeganeh N, Watts HD, Camarca M, et al. Syphilis in HIV-infected mothers and infants: results from the NICHD/HPTN 040 study. Pediatr Infect Dis J. 2015;34:e52–7.
44. Grimes DA, Schulz KF. Bias and causal associations in observational research. Lancet. 2002;359:248–252.
45. Viswanathan M, Berkman ND. Development of the RTI item bank on risk of bias and precision of observational studies. J Clin Epidemiol. 2012;65:163–178.
