Congenital syphilis is an infectious disease that is transmitted from a mother with syphilis to the fetus during pregnancy or childbirth and is caused by the bacterium Treponema pallidum (1). Globally, congenital syphilis is an infectious disease of high interest for public health but is occasionally neglected and requires collaborative actions for its control (2,3). In the Americas, congenital syphilis incidence has increased from 0.38/1,000 live births in 2009 to 0.61/1,000 live births in 2020 (4). This concerning trend underscores the importance of addressing this disease, which constitutes a major global cause of fetal loss, stillbirths, neonatal death, and congenital infection (1,5,6).
Comprehensive interventions involving diverse stakeholders in the healthcare system and community are crucial to preventing and controlling maternal and congenital syphilis, and those align with the third of the Sustainable Development Goals adopted by the United Nations Member States (2,7). Strategies used in local programs and shared globally include promoting condom use, ensuring timely access to antenatal care, early gestational syphilis detection through prompt point-of-care screening, and treating infections in a timely manner (8). Education on sexual and reproductive health, along with implementing epidemiologic surveillance, are also part of those efforts to prevent and control maternal and congenital syphilis (9). However, identifying and prioritizing populations for the specific reinforcement of those strategies in low- to middle-resource contexts is imperative.
In Colombia, a substantial increase in maternal syphilis prevalence was observed during 2017–2021; prevalence rose from 7.8 to 16.2 cases/1,000 newborns (live births and stillbirths). In 2021, maternal syphilis prevalence in Buenaventura alone was 45.8 cases/1,000 newborns, and congenital syphilis incidence also exceeded the national incidence (7.2 vs. 3.2 cases/1,000 newborns) (10), leading to the reporting of an epidemic in that area (11). Previous studies have aimed to assess the reason for this; a study conducted by Cruz et al. (11) in the same area found that only 8% of pregnant women received adequate treatment. Another study in South America revealed that congenital syphilis incidence is elevated in newborns of young Afro-American women who have lower educational attainment and women lacking prenatal care (12), elements that could be indicative of social vulnerability (13,14).
We sought to identify the characteristics (sociodemographic factors, obstetric history, and level of syphilis screening and treatment) of pregnant women enrolled in a prenatal care program (PCP) on the Pacific coast of Colombia associated with having newborns with congenital syphilis and the lack of maternal syphilis screening. In addition, we explored the characteristics of mothers with syphilis associated with having newborns with congenital syphilis.
Methods
Design
We conducted an analytical retrospective cohort study during January 2018–December 2022 in Buenaventura, Colombia. We used records of pregnant women enrolled in a PCP of a public hospital.
Study Area
Buenaventura is the main city on the Pacific coast of Colombia. The most recent population census in 2018 reported 258,445 inhabitants in the city, of whom 86.7% identify as Afro-Colombian (15). In contrast, at the national level, Afro-Colombians make up only 9.34% of Colombia’s total population and are predominantly concentrated along the Pacific and Caribbean coasts of the country (16). Afro-Colombians face elevated levels of multidimensional poverty, marked by disparities in occupation type, educational attainment, school dropout rates, literacy, and access to healthcare services (16).
Study Population
We included records of pregnant women of all ages who accessed a PCP in the main referral public hospital in Buenaventura (11) and whose babies were born in that hospital, with or without congenital syphilis. We excluded records of persons with multiple pregnancies and records with duplicate or incomplete data.
Variables and Definitions
Sociodemographic variables were age; ethnicity (Afro-Colombian or others); rural or urban area of residence; educational level, basic or lower (completed secondary school or lower) or postsecondary (technical, university, or higher); socioeconomic level, on the basis of a 1–6 scale (low-low, low, low-middle, middle, middle-high, and high) approximating the hierarchical socioeconomic difference from poverty to wealth in Colombia (17); marital status: single or with partner (de facto marriage or married); position as head of household (a person, whether single or married, who bears responsibility for providing financial or social support to their dependents); and the type of health coverage (18). We regarded the lowest categories within the outlined sociodemographic factors as characteristics of higher maternal social vulnerability factors. Other variables consisted of past pregnancy history (number of pregnancies, spontaneous abortions, and stillbirths); the gestational trimester in which prenatal care access was obtained; and whether syphilis screening had been performed and, if it was diagnosed, whether there had been a lack of treatment, defined as the lack of >1 dose of benzathine penicillin G (BPG) >30 days before delivery (9,19).
We applied the operational definitions of maternal and congenital syphilis according to the clinical practice guide issued by the Ministry of Health and Social Protection of Colombia (9). A patient with detected maternal syphilis was any mother with a diagnosis of syphilis during prenatal care, with or without clinical signs, who had a positive rapid point-of-care treponemal test accompanied by a reactive nontreponemal test at any dilution and who had not received adequate treatment or had an untreated reinfection (9).
A newborn with congenital syphilis was any live birth or stillbirth that met >1 of the following criteria: newborns of a mother with untreated syphilis or inadequate treatment (without >1 dose of BPG >30 days before delivery) to prevent congenital syphilis (9,19,20), regardless of the result of the nontreponemal test of the newborn; any newborn with nontreponemal test titers 4 times higher than the mother’s titers at the time of delivery, which is equivalent to 2 dilutions above the maternal titer; any newborn of a pregnant person whose syphilis was diagnosed during that pregnancy and with >1 clinical manifestations suggestive of congenital syphilis on physical examination, along with paraclinical tests suggestive of the infection; or any newborn with demonstrated T. pallidum in laboratory tests (9).
Data Sources
The hospital PCP provided the database, which included sociodemographic information, gestational age at the start of prenatal care, and obstetric history, as well as screening and treatment data for syphilis. Moreover, the database indicated whether the newborn was classified as a congenital syphilis case-patient. However, the database lacked details on the clinical stage of maternal syphilis and further details of the newborn, and although it reported the reactive or nonreactive result of the nontreponemal test for syphilis, it did not include the result in dilutions.
Statistical Analysis
We organized the data in Excel 365 (Microsoft, https://www.microsoft.comExternal Link) and conducted analyses using Stata 14.0 (StataCorp LLC, https://www.stata.comExternal Link). We conducted an exploratory analysis of the database to identify outliers or missing data. We reported categorical variables as frequencies and percentages and continuous variables as medians and interquartile ranges (IQRs). We determined the annual percentage of congenital syphilis cases by dividing the number of newborns with the infection by the total number of newborns for each year and multiplying by 100.
To identify maternal factors associated with having newborns with congenital syphilis, we compared pregnant women whose newborns had congenital syphilis with those whose newborns did not. Furthermore, to assess factors associated with the lack of maternal syphilis screening, we compared pregnant women who had >1 rapid point-of-care treponemal test through the PCP with those who did not. For those 2 objectives, we used 2 × 2 tables and calculated crude odds ratios with their respective 95% CIs. We assessed statistical significance using χ2 and Mann-Whitney U tests as appropriate. We conducted multivariable analyses through multiple logistic regression, and each initial or saturated model included variables with p<0.25 in the bivariate analysis, following the approach of Hosmer et al. (21), along with other variables considered to reflect social vulnerabilities. We selected the most parsimonious model by using the likelihood ratio test.
We conducted a subanalysis to identify factors in mothers with syphilis associated with having newborns with congenital syphilis. We compared patients with detected maternal syphilis during PCP and who had newborns with congenital syphilis to those who did not. We assessed the probability of having newborns with congenital syphilis using the relative risk (RR) as a measure of association. We determined RRs and corresponding 95% CIs through bivariate analysis. This research was conducted following the Declaration of Helsinki and was approved by the Human Research Ethics Committee of the Universidad Libre under protocol #010.+
