Lago, Eleonor G. MD, PhD; Vaccari, Alessandra MSc; Fiori, Renato M. MD, PhD
After a declining trend during the 1990s, syphilis has reemerged as a public health problem worldwide. The incidence of acquired and congenital syphilis (CS) increased dramatically in many countries during the last decade.1–3 Early screening and treatment for pregnant women with syphilis is a proven, cost-effective approach for reducing the burden of CS, which is associated with miscarriage, stillbirth, neonatal death, preterm birth, and severe disabilities.4–8 However, this disease continues to pose a major medical problem, with a great social impact, that health professionals must be prepared to face.
Important studies on CS have been done and have played an essential role in the implementation of strategies for its prevention. More recently, however, few published reports have focused on the clinical aspects of the disease. Although CS has been known for centuries and descriptions of its clinical features are widely available, it is important to assess the burden of the disease in different areas of the world. Particularly, data on long-term follow-up may constitute an important source of information and provide an additional basis for evaluation of control strategies. There are few published follow-up data on women with syphilis and their children. Thus, the present study aimed to investigate the clinical features and outcomes of children who were treated for CS in the neonatal period.
MATERIALS AND METHODS
The study was conducted in the public sector of São Lucas Hospital (SLH), a general teaching hospital belonging to the Pontifícia Universidade Católica do Rio Grande do Sul. It is located in Porto Alegre, a city of approximately 1,500,000 inhabitants and the capital of Rio Grande do Sul, the southernmost state of Brazil. Hospital delivery is a universal practice among the population seen at SLH, which is almost exclusively of urban origin. In all prenatal care facilities of Porto Alegre, syphilis testing is provided at the first visit and again at beginning of the third trimester of gestation. In a survey conducted in 1998 to 2000, 88.8% of a sample of women without syphilis who delivered at SLH received antenatal care, of which 50.9% made the first visit before 15 weeks of gestation. However, in that same period, only 67.5% of the mothers of infants with CS attended antenatal care.9
A venereal disease research laboratory test (VDRL) was performed in all delivering women at the time of admission to the obstetric ward of SLH, and titers below 1:16 were confirmed by a fluorescent treponemal antibody absorption test (FTA-ABS). All live births from May 1997 to December 2004 who met the case definition of CS were included in the study, as well as those whose mothers were diagnosed as having syphilis but received adequate treatment before delivery. The number of fetal deaths with syphilis occurring at the SLH obstetric ward over the study period was also collected but were not included in the analysis. To assess the broad spectrum of CS manifestations, neonates with CS born in other hospitals and admitted to the neonatal intensive care unit (NICU) of SLH during the enrollment period were also included in the study. Data on newborn infants and their mothers were collected throughout their stay in the maternity ward or NICU.
Congenital syphilis was defined and managed in accordance with guidelines published by the US Centers for Disease Control and Prevention (CDC)10,11 and the Brazilian Ministry of Health (BMH).12 In October 1992, BMH adopted almost entirely the 1989 guidelines of CDC for CS case definition. The only difference was that CDC included incomplete penicillin treatment in the category “no treatment,” whereas BMH considered this as “incomplete treatment.”10,12 Case definitions of CDC and BMH were also consistent with the definition of Pan American Health Organization.13
Nearly the same recommendations of CDC for diagnosis and therapy for mother and neonate were adopted by the BMH program “Congenital Syphilis Elimination,” launched in 1993.12 However, unlike 1989 CDC guidelines, Brazilian guidelines allowed the use of intramuscular benzathine penicillin to neonates with complete evaluation and no clinical abnormalities, even if the mother had untreated syphilis. In 2004, lack of partner treatment was added to the BMH criteria for inadequate maternal treatment and, consequently, to case definition of CS. However, this occurred after the period of enrollment of patients in our study.14
We defined as case of CS the infant whose mother had (1) untreated syphilis (no treatment at all); (2) treatment with a penicillin regimen inadequate for the stage of syphilis; (3) treatment administered less than or equal to 4 weeks before delivery; (4) undocumented treatment; (5) adequate treatment but nontreponemal antibody titers increased or did not decrease at least 4-fold, or there was insufficient serologic follow-up; and (6) treatment with a nonpenicillin regimen. Regardless of maternal history, the definition of CS also included an abnormal physical examination consistent with CS (described later) or a serum nontreponemal test titer 4-fold greater than the mother’s.
Our routine included a complete diagnostic evaluation (cerebrospinal fluid [CSF] examination, long bone radiographs, and complete blood count with platelets) in all infants who were defined as cases of CS. From 1997 to 2002, neonates received intravenous aqueous penicillin G for 10 days if the mother had no treatment, inadequate or undocumented treatment, treatment with a nonpenicillin regimen, or serologic evidence of relapse or reinfection after treatment; or the physical examination or any part of the evaluation of the neonate was abnormal or not done; or serum VDRL titer was 4-fold greater than the mother’s; or the CSF analysis was uninterpretable. Other cases, after a complete and normal evaluation, received a single dose of intramuscular benzathine penicillin. There was a change in the routine within the study period (in May 2002): recommendation of benzathine penicillin was expanded to infants whose mothers had received no treatment or had any type of inappropriate therapy, provided the infant had normal physical examination, normal complete laboratorial and radiologic evaluation, and assurance of outpatient follow-up.
During the enrollment period, all infants discharged after CS treatment were referred for follow-up at SLH Congenital Infections Clinic. Follow-up data were prospectively collected up to the end of 2009, when the last recruited patients completed 5 years of age. Financial support from the Brazilian National Council of Scientific and Technological Development allowed us to provide bus tickets for parents or guardians who brought children to the consultations during a part of the study period (2002–2005).
Follow-up visits and VDRL tests took place monthly until reversion to nonreactivity of the nontreponemal test. If there was a delay in presentation for follow-up and the last VDRL was positive, a control test was performed, regardless of age and time elapsed. Anthropometric parameters, general health, and psychomotor development (by the Denver Developmental Screening Test II) were evaluated at each visit. Nutritional status was assessed by the percent of the median weight for age and sex,15 using the World Health Organization weight-for-age tables as a reference. At 12 months of age, an FTA-ABS was requested and parents or guardians were invited to return with the child once a year until his or her fifth birthday.
This study was approved by the Research Ethics Committee of the Pontifícia Universidade Católica do Rio Grande do Sul, and all parents or guardians provided written informed consent for inclusion of their children in the study.
Data were entered and analyzed in Epi Info version 3.4. The χ2 test or Fisher exact test was used as indicated to test for associations. Student t test and Mann-Whitney-Wilcoxon test were used for comparison of means and of medians, respectively. The level of significance was set at P < 0.05.
We defined the following:
“Symptoms” or “Symptomatic”: physical examination abnormalities and/or symptoms consistent with CS (e.g., hepatosplenomegaly, skin rash, palmoplantar pemphigus, rhagades, hydrops, jaundice with increased conjugated bilirubin, rhinitis, pseudoparalysis, and fever)
“Laboratory/x-ray (lab/x-ray) findings”: CSF abnormalities, long bone lesions, anemia, thrombocytopenia, leukocytosis, leukopenia, or a serum VDRL titer 4-fold the maternal titer or higher. Criteria for normality/abnormality followed the literature and official standard guidelines.4,10–12 Evaluation was considered complete and diagnostic when all tests could be done. If one of these tests could not be performed or was incomplete or uninterpretable, the evaluation was considered inconclusive.
Maternal Syphilis and CS in Infants Born at the Study Hospital
Between May 1997 and December 2004, 24,920 live births took place at SLH. The mothers of 499 of these infants (490 pregnancies, 9 sets of twins) had been diagnosed as having syphilis during pregnancy or at delivery (diagnostic criteria described later, in “Causes and clinical features of CS”). Of these, 379 infants met the criteria for case definition of CS; this group was termed the CS-SLH group. The prevalence of CS in live births at SLH was 379 of 24,920, or 15.2 cases per 1000 (95% confidence interval [CI], 14–17/1000). There was no significant difference in prevalence over time, nor was there any trend toward change over the study period (data not shown).
Symptoms at birth occurred in 21 (5.5%) of the 379 liveborn infants in the CS-SLH group (95% CI, 3.5%–8.5%) and were more common in preterm infants: 21%, versus 3% in full-term infants (risk ratio [RR], 4.5; 95% CI, 2.6–7.7). Of 379 cases of CS, 292 (77%) had a conclusive lab/x-ray evaluation; of these, 85 (29%) had some physical or lab/x-ray abnormality, which were also more common in preterm (51%) than in full-term (26%) infants (RR, 1.97; 95% CI, 1.35–2.87). In all but 1 of these patients, diagnosis of CS was defined and treatment was started within the first week of life and before discharge from the SLH maternity ward.
The mothers of other 120 infants (1 set of twins) received adequate treatment of syphilis before delivery. This group of infants was termed the maternal syphilis (MS) group. Comparison between the SC-SLH and the MS groups showed that there were significantly more infants with gestational age below 34 weeks, birth weight was significantly lower, and there were more infants small for gestational age among the cases of CS. Cases with symptoms and/or lab/x-ray findings contributed more to these differences (Table 1).
During the same period, 37 fetal deaths with more than 20 weeks of gestation, attributed to CS, occurred at SLH but were not included in the analysis. Complete data on weight and gestational age were available for 25 of these stillbirth cases of CS and are described in Table 1.
Causes and Clinical Features of CS
A further 19 infants with CS, born at other hospitals and transferred to the SLH NICU were included in the study. Twelve of these patients had been discharged from maternity wards in the first hours of life in good condition, without their mothers being tested for syphilis, and the symptoms of CS appeared after a few days or weeks, leading to hospitalization at SLH, whereas 7 infants were transferred directly from other hospitals due to prematurity. This group of 19 neonates differed from the CS-SLH group for being a previously selected sample of sick infants who required admission in NICU, having 41.2% of preterm and 89.5% of symptomatic infants. Because of having a different denominator, infants from this group were not included in comparative analysis, but their clinical characteristics and outcomes were studied.
Considering both the 19 neonates from other hospitals and the 379 from the CS-SLH group, the total number of infants with CS treated during the study period was 398. This pooled group was termed the CS group. The makeup of all 3 groups is shown in Figure 1.
Among the 398 liveborn infants with CS, there were 9 twin sets, leaving 389 pregnancies. The most frequent maternal cause for case definition was (a) mother receiving no treatment at all, in 250 cases (64.3%), among which, VDRL was negative in pregnancy and positive at delivery (102), lack of prenatal care (92), pregnant woman failed to comply with diagnosis or treatment (31), treatment was not prescribed (23), or FTA-ABS was negative in pregnancy and positive at delivery (2); followed by (b) inadequate penicillin regimen, 60 cases (15.4%), among which, pregnant woman did not complete treatment (42), or doses were wrongly prescribed (18); (c) undocumented treatment, 36 cases (9.3%); (d) complete treatment but VDRL titers increased or did not decrease at least 4-fold, 22 cases (5.7%); (e) treatment administered less than 4 weeks before delivery, 14 cases (3.6%); and (f) treatment with a nonpenicillin regimen, 7 cases (1.8%).
Thirty eight neonates in the CS group had symptoms at birth or within the first 4 weeks of life, which are listed in Table 2. Risk of symptoms was higher with mothers receiving no treatment at all. Neonates were symptomatic in 32 (12.8%) of 250 cases of mothers who received no treatment at all and in 6 (4.3%) of 139 cases with other categories of inadequate maternal treatment (RR, 2.9; 95% CI, 1.3–6.9).
In 309 of 398 patients in the CS group, the specific lab/x-ray evaluation was conclusive, being more frequently abnormal in symptomatic infants. Laboratory/x-ray findings occurred in 99 cases, including 33 (91.7%) of 36 symptomatic and 66 (24.2%) of 273 asymptomatic infants. The findings were as follows: CSF abnormalities in 57 cases, anemia in 28, thrombocytopenia in 21, bone lesions in 21, VDRL titer 4-fold higher than the maternal titer in 12, elevated direct bilirubin in 11, elevated liver enzymes in 5, leukopenia in 4, and leukocytosis in 3 patients.
Among the 21 infants with long bone lesions, 15 (71%) had a VDRL titer 1:8 or higher (the highest was 1:256), and in 10 patients (47%), the titer was 4-fold that of the child’s mother. The physical examination was abnormal in 13 (62%) neonates with bone lesions, and 9 (43%) had CSF abnormalities. However, in 2 asymptomatic patients with VDRL titers of 1:1 and 1:2, respectively, osteochondritis was the only abnormality found despite full assessment.
Of the 518 patients of the sample, 508 survived to discharge and 10 died in the NICU. Five deaths were directly related to CS, and 2 were related to prematurity. Three deaths were from causes probably unrelated to CS or prematurity (2 patients with criteria for CS and 1 from the MS group). In addition, 2 deaths probably not due to CS, which occurred after the neonatal period among infants from the study sample, have come to our knowledge and are described (Table 3).
All infants defined as cases of CS were treated according to the routines described in the “Methods” section. Of 389 patients (in 9 who died, there was no time to start or complete treatment), 302 (77.6%) received intravenous aqueous crystalline penicillin G for 10 or 14 days, 85 (21.9%) received 1 dose of intramuscular benzathine penicillin G, and 2 (0.5%) received intramuscular procaine penicillin G for 10 days. Procaine penicillin therapy was found to be difficult to perform because of poor compliance and risk of discontinuation and, after a few attempts, was no longer indicated for neonates in SLH.
Follow-up at the Outpatient Clinic
Of the 508 patients who survived to discharge, 256 (50.3%) returned to SLH Congenital Infections Clinic for at least 1 follow-up visit. Of these, 202 belonged to the CS group and 54 to the MS group. Therefore, the follow-up rate was around 50% for both groups. Of these 256, 64 (25%) patients returned only up to 1 to 2 months of age, 45 (17.6%) were last examined between 3 and 7 months, 44 (17.4%) were examined between 8 and 12 months, 49 (19%) were examined between 1 and 3 years, and 54 (21%) were examined between 3 and 5 years. In 1 case, the mother refused to participate in the study.
To evaluate nutrition and the possible sequelae, only the 147 patients who could be evaluated between 8 and 60 months of age were selected. There were 27 (22.5%) of 120 in the MS group and 120 (30%) of 398 in the CS group (representing 29.5% of the infants born in SLH and 42% of those born in other hospitals, no significant difference between percentages, P = 0.367; Fig. 1.)
Weight for age was lower in the CS group when compared with the MS group. Eight children of the CS group and no children of the MS group had a very low weight for age at the last follow-up visit (percent of the median weight for age and sex <80%), a difference not statistically significant (P = 0.188). However, weight in the last visit was significantly lower in children from the CS compared with the MS group. The median of the percent of the median of the final weight was 102.5 in the CS group (interquartile range [IQR], 92–113; minimum, 52; maximum, 158) and 108 in the MS group (IQR, 103–116; minimum, 85; maximum, 125; P = 0.029). The difference remained significant even after excluding preterm infants: the median of the percent of the median of the final weight was 103 for full-term children in the CS group (IQR, 94–113.5; minimum, 52; maximum, 158) and 109 for full-term children in the MS group (IQR, 104–120; minimum, 97; maximum 125; P = 0.012). Thirty-one (30%) of the children without sequelae had symptoms and/or lab/x-ray findings in the neonatal period, which did not influenced the weight for age assessed in the last consultation.
Of the 120 children treated for CS that returned until at least 8 months, 16 had some sequela (see Late Sequalae section) and 104 were in good conditions, among which 2 had questionable development and 102 were healthy until last follow-up visit. Children with developmental disabilities were included in the group with sequelae. Among the 47 children whose mothers had received adequate treatment and were observed, 1 had questionable development and no one was considered to have a developmental delay.
Serum VDRL was positive at birth in 328 (82.5%) of the infants who met the case definition for CS, whereas 70 (17.5%) were nonreactors. In 4 neonates, VDRL was negative at birth and became positive between the second and seventh days of life. Only 2 of these infants had symptoms of CS.
Among the 398 cases of CS, 12 neonates (3%) had a serum VDRL titer 4-fold the maternal titer. In 5 (41%) of these, VDRL testing was done after the fourth day of life. In 11 (91%), symptoms were severe and they had lab/x-ray findings; 10 (83%) had bone lesions. Two of these infants did not survive to discharge. In 87 (83%) of 104 infants with symptoms and/or lab/x-ray findings in the neonatal period, the VDRL titer was less than or equal to the maternal titer.
Posttreatment VDRL reversion to nonreactivity, or persistence of negative titers, was possible to document in 173 patients in the CS group and 40 in the MS group. There was no case of increase in VDRL titers, including the patients treated with intramuscular benzathine penicillin. In 156 patients, we were able to identify the month of reversion to nonreactivity, which tended to occur earlier in infants without CS. By 3 months of age, VDRL was negative in 81.5% of the infants with CS and in 100% of the MS group (Table 4).
In only 1 patient who had bone lesions in the first month of life, VDRL remained reactive (at low titers) after 12 months of age. The infant was born in a hospital where the CS protocol was not applied, was discharged without diagnosis, and became symptomatic in the fourth week of life, when full treatment with aqueous crystalline penicillin was instituted. Because of the persistently reactive VDRL, the child was evaluated and treated again with 12 months. The only change found in the new lab/x-ray evaluation was a very high alkaline phosphatase, which persisted for years. Results from FTA-ABS were positive after 12 months (Tables 4 and 5, patient 13).
Results from FTA-ABS at 12 months of age or more were known in 87 patients with CS, being positive in 7 (8.7%). All 7 had symptoms and/or lab/x-ray findings in the neonatal period: 4 had abnormalities on physical examination; 4 had a VDRL titer 4-fold the maternal titer; 6 had major CSF abnormalities; 3 had bone lesions; and 1 had nephrotic syndrome. In 4 (57%) of the 7 patients, the diagnosis of CS was made after the first week of life, and 3 (43%) belonged to the group with late sequelae. Considering only the 32 patients with neonatal symptoms and/or lab/x-ray findings in whom FTA-ABS status after 12 months was known, it was nonreactive in 25 (78%).
Of the 120 patients in the CS group that could be evaluated between the ages of 8 and 60 months, 16 (13.3%) had 1 or more sequelae probably related to syphilis or prematurity, whereas none of the 27 patients in the MS group had sequelae (P = 0.031). Low weight for age alone was not considered a sequela. Of the 16 infants, 12 had developmental delay plus other sequelae, 2 had only developmental delay, and 2 had normal development and other types of sequelae (Table 5).
Of the 16 patients with sequelae, only 4 had an abnormal physical examination at birth, whereas in other 4, the onset of symptoms was within the first 4 weeks of life. However, 13 of the 16 neonates had lab/x-ray findings. Nine patients (56%) had CSF abnormalities, 4 (25%) had bone lesions, and 3 (19%) had a VDRL titer 4-fold the maternal titer. One patient who had arrested hydrocephalus, growth hormone deficiency, and dwarfism at the age of 5 years had persistent hypoglycemia in addition to typical clinical picture of CS in the neonatal period (Table 5, patient 9). Four patients with sequelae received their diagnoses only upon symptoms onset, after the first week of life (Table 5).
There was an association between the presence of sequelae and the presence of symptoms and/or lab/x-ray findings in the neonatal period, with a 20-fold risk of sequelae in infants with neonatal symptoms and/or lab/x-ray findings. Excluding 25 patients with inconclusive laboratory evaluation, 13 (93%) of 14 patients with sequelae and 31 (38%) of 81 without sequelae had exhibited neonatal symptoms and/or lab/x-ray findings (odds ratio, 20.9; 95% CI, 2.6–168.3). Of these 31 infants with neonatal symptoms and/or lab/x-ray findings but no late sequelae, 29 had started intravenous penicillin therapy within 7 days of life.
Much has been written about CS, but in recent years, only a few studies have addressed its clinical course. The descriptions of this disease are classic and long-standing, sporadically updated by case reports. This study, in addition to providing an update on the epidemiologic and clinical aspects of CS, presents a unique comparison between a group of newborn infants meeting a standard case definition of CS and a group of infants whose mothers were adequately treated for syphilis, drawn from the same population and monitored simultaneously.
Unfortunately, we are still far from achieving the goal of BMH and Pan American Health Organization of reducing the incidence of CS to 0.5 cases per 1000 births.12,13 The prevalence of 15.2 cases per 1000 live births was apparently higher when compared with the reported to the Brazilian official epidemiologic surveillance system, which is 1.2 cases per 1000 live births.16 However, when compared with the prevalences found in other Brazilian research studies, it was similar or lower, for example, 21.9 of 1000 in a public hospital in Niteroi City, Rio de Janeiro state,17 or 16.1 of 1000 in Manaus, Amazonas state,18 confirming the problem of underreporting of CS in Brazil.3,19
Congenital syphilis is known to be associated with preterm birth,4,5 and there was a higher proportion of very premature infants (<34 weeks’ gestational age) in the CS-SLH group compared with the MS group, which is due to the subgroup of neonates with symptoms and/or abnormal lab/x-ray findings, as we can see in Table 1. However, the proportion of infants younger than 37 weeks was not significantly different between the 2 groups, even when we include in the analysis only the symptomatic neonates. Although the sample may have had little statistical power to detect these differences, we can assume that in some pregnant women, the mechanisms whereby infection induces preterm labor may have already acted before treatment, even if the latter was adequate. In other studies, the control group consisted of infants whose mothers did not have syphilis.5,9 Data presented in Table 1, including description of stillbirth, emphasizes once more the importance of early screening and treatment for pregnant women, which has been exhaustively reported.3–8
Associations between CS, prematurity, and severity of clinical picture lead us to suppose that it would be appropriate to include preterm birth among the criteria for case definition of CS, perhaps eliminating the need for a complete lab/x-ray workup in these very frail infants. Taking this association into account could also raise awareness of the diagnosis of CS in preterm infants.
Apart from premature birth, CS causes growth restriction, both intrauterine and postnatal. This was clear in the present study upon comparison of the CS and MS groups in Table 1. Association of CS with underweight reflects the deleterious effect of infection, but clinicians should also bear in mind that other factors that affect women with syphilis during pregnancy, as already demonstrated in this and other populations, may have direct and indirect influences on fetal and child growth. In addition, the profiles of treated mothers were found to be different from the untreated ones. Usually, pregnant women who were adequately treated for syphilis had high-risk behaviors but were more attentive to health care.9,20
Some abnormality was detected by lab/x-ray evaluation in 25% of asymptomatic infants, stressing the need for a thorough clinical evaluation of infants meeting the criteria for a diagnosis of CS. Any positive finding is an indication for full treatment with aqueous crystalline penicillin, given the high probability of central nervous system invasion by Treponema pallidum in these cases.21
A VDRL titer 4-fold the maternal titer was unusual, even in the presence of symptoms and/or lab/x-ray findings (as already reported in the literature), although its occurrence should be considered indicative of clinical involvement.4,22 Ten of the 12 neonates with a VDRL titer 4 times higher than the maternal had bone lesions. The association of bone changes and a high VDRL titer seems to be explained by the fact that bone lesions occur later than most other clinical changes, when there has been enough time for VDRL titers to rise. In a recent case series, 8 of 10 patients with CS who had bone lesions were diagnosed only after the first month of life.23 In our study, 24% of infants with bone lesions received the diagnosis of CS after the first week of life. However, in 2 neonates, the only positive finding at birth was osteochondritis. This is interesting because there is a long-standing discussion about the usefulness of long bone x-ray in the CS workup.24
A VDRL test result was negative at birth in 70 infants with CS, 4 of whom converted to positive after the first day. Congenital syphilis cannot be ruled out simply because the infant is nonreactive for the nontreponemal test: these patients should be investigated and treated according to standard guidelines based on maternal status. Although this has been known for decades,25 some still equate a negative nontreponemal test result at birth with absence of CS. This clearly shows the critical role played by the strict application of standard guidelines in the management of CS. Our data suggest an association between starting treatment after the first week of life and presence of sequelae, but this aspect needs to be further studied by analyzing data on the exact number of days elapsed from birth to diagnosis.
Testing for FTA-ABS after 12 to 18 months of age has been proposed by CDC10,11 for epidemiologic surveillance purposes, as a reactive FTA-ABS after the disappearance of maternal antibodies is evidence that the child had actually been infected with T. pallidum. However, in a study published in 2001, Rawstron et al.22 showed that only 54 (27.5%) of 196 infants with CS had a reactive FTA-ABS after 12 months. Of those, 31% had exhibited symptoms and/or lab/x-ray findings during the neonatal period.22 Data were similar to those reported herein, except that all reactors in our study had neonatal symptoms and/or lab/x-ray findings. As in the study of Rawstron et al, however, FTA-ABS nonreactivity in infants who had clinical findings unquestionably due to CS showed that the test can revert to negative even in infected infants. Early treatment is likely to alter the antibody response, making FTA-ABS reactivity disappear over time.22 Symptomatic neonates, as shown by the data in this study, are more likely to have been infected for a longer time before starting treatment.
The main limitation of this study was the high rate of loss to long-term follow-up. Therefore, these results cannot be used to calculate the prevalence of sequelae in CS. Long-term monitoring of children with CS is difficult to implement because of low compliance by their mothers, who often lack motivation to seek health care. Other authors have encountered similar difficulties. In Florida, Ricci et al.26 attempted to follow 56 infants with a diagnosis of CS. At the time of publication, only 29 infants were being observed because of poor adherence to follow-up; of these, 41% had growth deficiency, 10% had chronic lung disease, and 3% had necrotizing enterocolitis.26 In the sample by Rawstron et al.22, only 17% of infants referred to follow-up after being treated for CS at Kings County Hospital Center in Brooklyn, New York, returned appropriately. In a predominantly African-Caribbean neighborhood in London, Sothinathan et al.27 started the follow-up of 63 infants whose mothers were seropositive for syphilis. Only 5 patients returned up to the age of 6 months, and 1 patient up to 12 months. Follow-up was done at the hospital, and the authors reported low adherence by mothers.27 In our study, follow-up centered at the hospital’s outpatient clinic may also have been an unfavorable factor. That we compared groups drawn from the same population, potentially suffering from the same biases, minimizes these methodological difficulties.
Even taking into account these limitations, some important conclusions emerge from this study. It showed how asymptomatic infants, when not timely treated, may become symptomatic in the first weeks of life or exhibit abnormalities only on an in-depth, specific workup. The wide range of sequelae found in this study corresponds to the pathogenic potential of CS, and in some cases, they could also be secondary to preterm birth. Almost half of all patients with detected sequelae were born preterm, and more than half had low birth weight.
Children from adverse socioeconomic strata are subjected to considerable risks because the environment in some cases may not provide all the necessary resources to enable them to achieve their potential of health, nutrition, and development.28 Thus, we must take these factors into account when interpreting the outcome of our patients. However, results found here leave no doubt of disclosing the pathogenic potential of CS. Our data are also consistent with what has been already shown: notwithstanding CS being a theoretically preventable and cost-effective intervention disease, there are circumstances that prevent its elimination. Some of these are inadequate prenatal care use by pregnant women, inadequate approaches by health care personnel, and insufficient public policies/resources.29,30 Victora et al.5 identified also research gaps that must be addressed for CS prevention.
On the other hand, we found that most infants with CS, when diagnosed and treated promptly, even when symptomatic or having lab/x-ray findings at birth, responded well to treatment and had a good outcome, provided that basic environmental conditions for growth and development were fulfilled. Therefore, the prognosis of children adequately treated for CS can be considered favorable in the absence of a very severe disease at birth and of other risk factors, especially preterm birth. Serologic and clinical follow-up revealed that intramuscular benzathine penicillin treatment for asymptomatic infants with normal clinical evaluation was safe and effective. Among the infants who were observed, none of those treated with benzathine penicillin failed to have a good response.
This study emphasizes the value of standard guidelines for management of CS according to maternal history. Criteria used to define a case of CS were recommended for epidemiologic surveillance purposes, but they proved to be useful in the clinical management of neonates and their mothers. Although CS remains an impacting disease that causes fetal and neonatal deaths, premature birth, low birth weight, and severe, irreversible sequelae in some children, case definition has been highly effective in identifying infants who had no chance of being infected and in preventing those actually infected from being discharged without treatment, therefore improving their clinical and developmental outcomes.