Discussion

As the prevalence of syphilis in China has risen since the 1990s, to prevent syphilis transmission from mother to child has become a priority.10 ,16 In this study, the overall prevalence of maternal syphilis was much higher than in other parts of the world, such as Europe (0.16%), the Mediterranean (0.06%) and slightly higher than the Chinese national average (0.24%).1 ,9 Additionally, the prevalence in our study was similar to those in Shenzhen and Shanghai, China, where both were estimated at 0.30%, but it was lower than those in rural parts of Southern China (0.39%), Latin America (1.1%) and India (0.57–0.78%).4 ,13–15 ,20 This largely conformed to the general population epidemiology and geographic distribution of syphilis in China.16

Although the government provided free syphilis screening to all pregnant women, 100% coverage has not been achieved for several reasons. Most of these mothers with syphilis in our study were migrants, unemployed or illiterate, which might have increased the risk of infection and the likelihood of missed active interventions. Moreover, syphilis was subjected to HIV/STD coinfection as previously reported.21 Stigma and isolation might pose obstacles to screening and antenatal care. Therefore, we found a high proportion of women (17.6%) identified at or after delivery and less than 90% received treatment. Even among those women treated during pregnancy, only 68.2% received adequate treatment. Syphilis prevention is available and treatable since the advent of penicillin; nevertheless, deferred diagnoses usually lead to delayed treatment or even lack of treatment entirely. These findings were comparable with current studies.1 ,22–24 It was estimated that globally approximately 20% of women with syphilis did not attend antenatal care.1 In some areas, over 70% of women were diagnosed at late trimester and only half of women received treatment, and in Oliveira's study, the proportion of inadequate treatment for maternal syphilis arrived at 77.6%.20 ,22 ,23 This reinforces the concept of antenatal care correlates with effective intervention.

We found herein that untreated or inadequately treated maternal syphilis remained an important risk factor for a series of adverse pregnancy outcomes, which agreed with a wide range of published studies.1–4 In a recent meta-analysis performed by Qin et al2 that involved a large number of pregnant women with and without syphilis, the prevalence of adverse results among the untreated group was 36.0% for CS, 23.2% for preterm delivery, 23.4% for LBW and 26.4% for stillbirth or fetal loss. The latest meta-analysis performed by Gomez et al3 showed 17.2% for stillbirth and fetal loss, 6.6% for neonatal death and 6.6% for prematurity or LBW among women with untreated syphilis. In general, the above figures were much higher than the untreated group in our study, which were 3.8% for CS, 14.2% for prematurity and 10.1% for LBW. Owing to variations in study design, study origin, stage of maternal infection, treatment regimens and inclusion criteria across different researches, the potential bias with regard to comparisons should be considered. In the present study, the majority of infected women were asymptomatic and had mild cases, and the data on exact disease progression in Qin and Gomez’s studies were not presented. Besides that, we only investigated the adverse pregnancy outcomes of women with syphilis with singleton births, calculated the stillbirths over 28 weeks and reported the identification of CS at 0–7 days after birth. In this way, we might have underestimated the specific adverse pregnancy outcomes. However, in Qin's study, they excluded the HIV-infected population and provided data that included all women with early fetal loss and miscarrage, and in both Qin and Gomez’s studies, prematurity and LBW and stillbirth and fetal loss were all combined.2 ,3

We observed that the treatment initiated in the first trimester reduced the risk of asphyxia, CS and LBW more than the treatment initiated in the third trimester. Conversely, no significant differences were observed in the prevalence of birth defects, preterm birth and perinatal death between women who initiated treatment in the third trimester and the first trimester. The OR values for preterm birth and perinatal death approximated the cut-off, which attracted our attention. These findings were supported by a systematic review that reported an overall OR for any adverse outcome in women treated only during the third trimester at 2.24 compared with those women treated at first and second trimesters.25 Importantly, every week of delay in treatment increased the risk for adverse pregnancy outcomes by 2.82-fold.13 The call for early treatment should be strengthened in order to promote the treatment adequacy as non-penicillin treatment exerted minimal impact on incomplete treatment in our setting.

CS is a disabling infection that can destroy multiple organ systems and present various symptoms.5–7 Among those infants with CS, 50 infants were reported with titres in the TRUST or RPR tests four or more times higher than their mothers, four infants were detected with the Treponema and others were reported to be positive for IgM. Thirteen out of 62 infants with CS showed differential clinical manifestations, such as skin pimples (eight cases), chondritis (one case), anaemia (two cases), pneumonia (three cases) and hepatosplenomegaly (three cases). Of them, four infants were reported with various clinical features. In our study, 45.2% (28/62) received no treatment for syphilis among mothers with infants with CS, which was similar to a study in USA.5 Nevertheless, among treated mothers with infants with CS, 58.8% (20/34) received inadequate treatment for syphilis, which was much higher than the data (30.0%) reported in the same study from the USA.5 Although we only presented the prevalence of CS among perinatals, CS was preventable and curable when mothers received early and adequate treatment.

Pneumonia, asphyxia, birth defects, CS, preterm birth and LBW all increased the risk of perinatal death, and the interplay between these elements was promoted to some extent. In Arnesen's study, women with gestational syphilis had increased odds of stillbirth (OR=6.87).24 Based on the report from the USA, no or inadequate treatment for maternal syphilis, infants with CS born alive at <28 weeks' gestation or born weighing <1500 g all increased the likelihood of a dead case.26 Here, we reviewed the causes of 42 new infant deaths and noticed that pneumonia, asphyxia, LBW and CS contributed greatly. Despite the unknown infection status for some of the dead infants, it was deduced that CS was an independent risk factor because the dead infants had been born to untreated or incompletely treated mothers.

There are several limitations to the present study. First, we did not provide data for all pregnant women; thus, we could not compare outcomes between women with and without syphilis. Second, we only focused on selected adverse pregnancy outcomes of women with syphilis with singleton births according to their treatment. Improved data on early fetal loss as well as complete estimate for CS and outcomes of multiple births should be addressed. Lastly, we did not analyse the mother's stage of syphilis infection and partner's infection status due to limited data, and this might negatively influence the women's health or treatment effects.

In summary, this is the first systematic investigation of syphilis prevention and pregnancy outcomes in Zhejiang Province, China. The results suggested that treatment, especially adequate and early treatment, improves pregnancy outcomes and promotes the health of offspring. This may be achievable in Zhejiang because of the integrated PMTCT services and the presence of a government-driven programme, but there are still some barriers to the complete success. Improving the coverage with regard to early screening and adequate treatment among pregnant women should be the goal for future strategies, particularly by stressing prenatal healthcare early among key segments of the population.