WARNER, LEE MPH*; ROCHAT, ROGER W. MD*†; FICHTNER, RONALD R. PhD‡; STOLL, BARBARA J. MD§; NATHAN, LAWRENCE MD⌆ AND; TOOMEY, KATHLEEN E. MD, MPH*
From the *Georgia Department of Human Resources, Division of Public Health, Atlanta, Georgia; the Centers for Disease Control and Prevention, †National Center for Chronic Disease Prevention and Health Promotion, Division of Reproductive Health, ‡National Center for HIV, STD, and TB Prevention, Prevention Informatics Office, Office of the Director, Atlanta, Georgia; and the Departments of §Pediatrics and ⌆Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, Georgia
The authors thank the following persons for their contributions to the design and implementation of the study or for review of the manuscript: Jim Bellinger, Stuart Berman, Dabo Brantley, Cassandra Bray, Stuart Brown, Sharon Clanton, Linda Webster Dicker, William Graves, Ruby Lewis Hardy, Jack Kirby, Maureen Miller, Allyn Nakashima, Tonya Ross Mitchell, Lynn Stockton, and Winston Yung.
Correspondence: Lee Warner, MPH, National Center for HIV, STD, and TB Prevention, Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, 1600 Clifton Road, Mail Stop E-46, Atlanta, GA 30333. E-mail: firstname.lastname@example.org
Reprint requests: National Center for HIV, STD, and TB Prevention, Office of Communications, Centers for Disease Control and Prevention, 1600 Clifton Road Mail Stop E-06, Atlanta, GA 30333.
Received for publication February 14, 2000,
revised June 9, 2000, and accepted June 13, 2000.
Lee Warner is currently with the National Center for HIV, STD, and TB Prevention, Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, Atlanta, GA.
Roger W. Rochat currently is with the Department of Epidemiology of the Rollins School of Public Health, Emory University, Atlanta, GA.
SYPHILIS INFECTION DURING pregnancy can result in fetal or perinatal death, stillbirth, or severe morbidity in infected infants. 1,2 The management of congenital syphilis (CS) can be costly, with estimates exceeding 35 million dollars per year in the early 1990s. 3 Since the peak of the US syphilis epidemic during the late 1980s and early 1990s when more than 50,000 cases were reported annually, the number of cases of adult syphilis and CS has declined nationwide. 4,5 Nevertheless, reports of recent syphilis outbreaks 6,7 and the historically cyclical nature of syphilis epidemics 4 suggest that prevention of CS remains an important ongoing public health challenge.
Congenital syphilis has been repeatedly associated with substance abuse and receipt of inadequate or no prenatal care. Previous studies indicate that women received no prenatal care in as many as 67% of pregnancies resulting in CS cases. 6–12 However, a significant proportion of CS cases occur among women who do receive at least some prenatal care, suggesting that opportunities for diagnosis, screening, or treatment of maternal syphilis may have been missed.
Because CS is highly preventable, national prevention guidelines have been established for the screening and treatment of syphilis in pregnant women attending prenatal clinics. 13–15 Recommendations for women in high-risk populations include a serologic test for syphilis (STS) by the first prenatal visit, a second STS during the third trimester, and a third STS at the time of delivery. Penicillin therapy administered 30 days or more before delivery with a serologic response appropriate for maternal stage of infection is considered adequate treatment.
Few studies have systematically reviewed hospital obstetric records from a high-risk population to determine factors contributing to CS among women receiving prenatal care. This study evaluated CS cases at Grady Memorial Hospital (GMH), an urban southeastern teaching hospital serving a primarily indigent population in Atlanta, GA. The study objectives were to evaluate (1) the prevalence of CS; (2) sociodemographic characteristics of women delivered of neonates with CS; (3) characteristics of prenatal care for women delivered of neonates with CS; (4) provider compliance with CS prevention guidelines; and (5) missed opportunities for CS prevention.
We reviewed medical records of deliveries at Grady Memorial Hospital for case reports that met the Centers for Disease Control and Prevention surveillance case definition of CS. 16 Hospital-specific case reports of CS were identified through record linkage of Centers for Disease Control and Prevention Congenital Syphilis Case Investigation and Report Forms (73.126) with Georgia birth and fetal death certificate data. Records of cases that occurred at GMH during a 3-year period, between July 1, 1990 (the date of implementation of CS surveillance case definition in Georgia), and June 30, 1993, were examined.
We manually abstracted maternal and infant information for CS cases from hospital medical records and from Georgia county and state health department syphilis morbidity records. Maternal information included sociodemographic characteristics, substance abuse during pregnancy, history of sexually transmitted diseases (STD) and HIV infection, onset and timing of prenatal visits, date, type, and result of screening and confirmatory tests, and treatment for syphilis before and during pregnancy. Maternal information was also obtained from state birth and fetal death certificate data to compare characteristics of women who delivered of neonates (live births or stillbirths) with CS at GMH with women who delivered of neonates (live births or stillbirths) without CS at GMH, and with the Georgia statewide obstetric population as a whole.
The first prenatal visit at GMH consists of a comprehensive medical interview and a routine battery of laboratory tests, including an initial STS. A clinical examination is not conducted until the second prenatal visit. A repeat STS is routinely performed between 34 and 36 weeks’ gestation and again at delivery. We defined the receipt of any prenatal care as the minimum completion of the first prenatal visit because an STS at this time represents the earliest possible opportunity for CS prevention.
We defined provider compliance with CS prevention guidelines as screening at the first prenatal visit (i.e., an STS before or on the date of the initial clinical examination), screening during the third trimester (i.e., an STS between 26 weeks’ gestation and 72 hours before delivery), and screening at delivery (i.e., an STS within 72 hours of delivery or immediately postpartum). For women entering prenatal care during the third trimester, an STS conducted at the first visit dually served as a third trimester STS.
We defined a case patient as a woman who delivered of a neonate (live birth or stillbirth) with CS at Grady Memorial Hospital during the specified period. Maternal infection was documented by the combined presence of a reactive nontreponemal serologic test for syphilis (rapid plasma reagin) indicative of new infection and a reactive treponemal test (fluorescent treponemal antibody absorption or microhemagglutinin-Treponema assay) for antibodies to Treponema pallidum. An obstetrician, a neonatologist, and an epidemiologist specializing in CS reviewed each case to identify factors contributing to CS occurrence.
All analyses were conducted using SAS version 6.11 (SAS Institute, Cary, NC). Tests of statistical difference between subgroups were made using chi-square tests of association. Results were considered statistically significant when P ≤ 0.05. The study protocol was reviewed and approved by the Human Investigations Committee of the Emory University School of Medicine (Atlanta, GA).
Congenital Syphilis Prevalence
From July 1, 1990 to June 30, 1993, a total of 438 deliveries statewide were classified as CS cases according to Centers for Disease Control and Prevention surveillance case definition criteria. GMH accounted for 178 (40.6%) of statewide CS cases, comprising 160 live births and 18 stillbirths. The hospital-specific prevalence of CS during the study period was 0.9% (178 cases in 19,604 deliveries). On further review, 18 of 178 (10.1%) reported CS cases from GMH did not meet surveillance case definition criteria because the mother received adequate treatment during pregnancy (n = 9) or had a serofast response to treatment administered before pregnancy (n = 9). The remaining 160 CS cases (from 157 pregnancies to case patients), showed a CS prevalence of 8.2 cases per 1000 live births. The prevalence of CS was more than 5 times higher among nonwhites (9.6 cases per 1000 live births) than among whites (1.7 cases per 1000 live births).
Birth and fetal death certificate analyses revealed that women who delivered of live or stillborn neonates with CS at GMH differed from both women who delivered of live or stillborn neonates at GMH without CS and the Georgia statewide obstetric population as a whole. Women who delivered of live or stillborn neonates with CS at GMH were more likely than those in the other groups to be 20 years or older, unmarried, nonwhite (primarily African American), multigravida, users of tobacco and alcohol during pregnancy, and without a father’s name on the infant’s birth certificate (Table 1). Additionally, these women were less likely to have prenatal care, and when prenatal care was sought were less likely to have adequate prenatal care according to the Kessner Index.
Medical record review for the 157 case patients further revealed that 94 (59.9%) had a positive toxicology screening result or a self-reported history of cocaine use, 14 (8.9%) had a positive toxicology screen or self-reported history of opiate use, and 17 (10.8%) self-reported marijuana use. Tobacco and alcohol use during pregnancy were documented in medical records for 85 (54.1%) and 66 (42.0%) case patients, respectively; reports of alcohol use were markedly higher in medical records than in birth and fetal death certificate data (28.9%) (Table 1). Thirty-three case patients (21.0%) had been sexually assaulted before pregnancy, three index pregnancies resulted from a sexual assault. Thirty-two case mothers (20.4%) had also been physically assaulted.
Sexually transmitted diseases other than syphilis (i.e., gonorrhea, chlamydia, and HIV) were diagnosed in 50 (31.8%) case patients during pregnancy; 9 of 145 (6.2%) case patients tested were HIV positive. Fifty-six case patients (35.7%) also had a history of syphilis before pregnancy. The maternal stage of syphilis infection during pregnancy was early latent (63.1%), secondary (21.0%), and late or unknown latent (15.9%). No cases of primary syphilis were reported.
Characteristics of Prenatal Care
Of the 157 case patients, 70 (44.6%) did not receive prenatal care according to medical record review. (This percentage was slightly higher than that reported on birth and fetal death certificate data). Of the remaining 87 case patients who did receive prenatal care, 9 sought care at another facility and could not be evaluated, and 14 initiated care less than 30 days before delivery, which was too late for the prevention of CS. Therefore, we restricted analyses to the 64 case patients who started prenatal care 30 days or more before delivery and who theoretically had opportunities for CS prevention.
Among these 64 women, the mean time of the first prenatal visit (i.e., registration visit) was 20 weeks’ gestation (median, 20 weeks; range, 7–33 weeks), and the mean time of the second prenatal visit (i.e., clinic examination) was 23 weeks’ gestation (median, 22.5; range, 9–40 weeks). Overall, 49 (76.6%) women initiated prenatal care during the first or second trimesters. The mean number of prenatal visits for the 64 women was 5.8 visits (median, 4 visits; range, 1–17 visits). Eight (12.5%) women never returned for prenatal care after the registration visit, and 14 (21.9%) never returned after the first clinical examination. At delivery, these 64 case patients averaged 37.9 weeks’ gestation (median, 38 weeks; range, 29–43 weeks). The number of days between the first prenatal visit and date of delivery averaged 22.2 weeks (median, 23.1 weeks; range, 6–37 weeks) during which opportunities for improved maternal screening and treatment may have existed.
Provider Compliance With Recommended CS Prevention Guidelines
For pregnancies resulting in CS, provider compliance with recommended STS screening guidelines was high at the first prenatal visit, during the third trimester, and at delivery (Table 2). Of the 64 case patients who received prenatal care 30 days or more before delivery, 59 (92.2%) had an STS by the first clinic examination.
The rate of third trimester STS screening, however, was significantly lower than the rate of first visit STS screening (76.6% versus 92.2%; chi-square (1 df), 5.93;P = 0.015) and varied by trimester of entry into prenatal care. Compared with women entering prenatal care in the first or second trimester (n = 49), those entering prenatal care during the third trimester (n = 15) were significantly more likely to receive a third trimester STS (100% versus 69.3%, P = 0.015); however, this finding is a result of the high priority assigned to STS screening for late entrants into prenatal care. The average time for third trimester STS screening was 34.1 (±4.6) weeks’ gestation (median, 33 weeks; range, 25.6–43.4 weeks), which is consistent with GMH prevention guidelines for CS. All 157 case patients received an STS at the time of delivery or immediately postpartum, regardless of prenatal care status.
Missed Opportunities for Congenital Syphilis Prevention
Several factors contributed to CS occurrence among the 64 case patients who received prenatal care 30 days or more before delivery according to a medical record review (Table 3). Four case patients (6.3%) did not receive a timely STS during pregnancy despite earlier prenatal care. Fifteen case patients (23.4%) did not have an early third trimester STS after a negative first STS result (Figure 1); of these 15 women, 9 had early third trimester visits documented before 34 weeks’ gestation, 4 discontinued prenatal care by the third trimester, and 2 delivered prematurely before a repeat STS could be conducted. An additional 17 case patients (26.6%) received no or inadequate treatment for maternal syphilis after a reactive STS (Figure 1). Of those 17 women receiving no or inadequate treatment, 12 could not be located for treatment, refused treatment, or were incorrectly assumed to have received treatment; 2 received inappropriate (erythromycin) treatment because of penicillin allergy and were not desensitized; and 3 failed to complete multicourse treatment regimens before delivery. The duration between the first reactive STS and initiation of treatment averaged 36 days (median, 39 days; range, 0–95 days).
Fifteen case patients (23.4%) had serologic evidence of treatment failure or reinfection after adequate treatment, 10 of whom experienced a fourfold increase in titer after an adequate response to treatment during pregnancy (i.e., fourfold decline in titer), suggesting possible reinfection. Twelve case patients (18.8%) were infected late in pregnancy after a negative third trimester STS result; in one clinically confirmed CS case, seroconversion of the case patient occurred after a negative STS result at delivery. In one case, apparent infection with symptomatic secondary syphilis was not detected.
Overall, 37 case patients (57.8%) with timely prenatal care had CS outcomes that could be attributed to potentially preventable causes from a provider standpoint, such as failure to provide a repeat STS or lack of treatment after receiving a positive STS result. Conversely, 27 case patients (42.2%) had CS outcomes that resulted from potentially nonpreventable causes from a provider standpoint, such as treatment failure, reinfection, or late infection. Preventable causes of CS were more likely to occur among women with fewer prenatal visits. Of 50 CS cases occurring among women with fewer than 10 prenatal visits, 33 (66%) were attributable to preventable causes. Of 14 CS cases occurring among women with 10 or more prenatal visits, 4 (28%) were attributable to preventable causes. The difference in these percentages is statistically significant from 0 (chi-square (1 df), 6.28;P = 0.012) (Table 3).
Our study provides an in-depth profile of prenatal care received by women who delivered of neonates (live births or stillbirths) with CS at GMH during the peak of the US syphilis epidemic in the early 1990s, and identifies missed opportunities for prevention of congenital infection. Marked declines in syphilis rates have been observed since the early 1990s both in Georgia and nationwide. 4 The number of early syphilis cases in Georgia declined from a peak of 4,385 cases in 1990 (67.7 cases per 100,000 population) to 740 cases in 1998 (9.9 cases per 100,000 population). Similarly, the number of CS cases in Georgia among newborn infants less than 1 year of age declined from a high of 178 cases (159.8 cases per 100,000 live births) to a low of 14 cases (12.3 cases per 100,000 live births) between 1992 and 1998. Despite these declines, the obstetric population at GMH in particular remains at risk for CS occurrence because of the number of primary and secondary syphilis cases reported in Fulton County (Atlanta, GA), which in 1998 was ranked as an area with the eighth highest concentration of syphilis cases nationally. 4,17
Our findings suggest a high level of provider compliance at GMH with CS prevention guidelines for high-risk women whose pregnancies resulted in CS. Potential opportunities for earlier screening, treatment, and diagnosis of maternal infection still existed for nearly 60% of case patients who initiated prenatal care, even for those with suboptimal use of prenatal care. Previous studies have hypothesized that CS prevention efforts may be more effective among women with more prenatal visits 12 because these women theoretically have more frequent opportunity for detection, screening, and treatment of maternal syphilis. Although these women by definition have more prenatal encounters, this would not necessarily translate into more provider-based opportunities for CS prevention. Most cases of CS occurring among women who received more prenatal care (i.e., ≥10 visits) resulted from less preventable causes from a provider standpoint, such as reinfection, treatment failure, or late infection. Conversely, CS cases among women who received less prenatal care were significantly more likely to result from more preventable causes, such as failure to provide timely screening or treatment. The finding that the majority of CS cases among women receiving prenatal care had few prenatal visits further suggests that this group is the most appropriate target for provider-based CS prevention efforts.
This finding holds important relevance for prenatal care programs because syphilis in pregnancy can be easily diagnosed and treated despite minimal clinic attendance. 10 Given the inadequate patient use of prenatal care in this study and in others, 7,8,10 we recommend that prenatal care providers serving high-risk populations manage each patient encounter as if it were the final opportunity to diagnose, screen, or treat for maternal syphilis. Efforts should continue to be made to address personal and structural barriers to early initiation and adequate use of prenatal care, such as providing patients referrals to substance abuse treatment, increasing patient awareness of the benefits of prenatal care, and assisting with patient transportation and child care needs.
Our findings also suggest that several inexpensive provider-based interventions could help reduce CS in patient populations where use of prenatal care is suboptimal. First, on-site testing and same-day treatment 1 may be effective among high-risk women who sporadically seek health care. This is particularly true for women in whom syphilis is diagnosed during the third trimester and for women who may be difficult to locate for treatment. Second, the timing of third trimester STS screening should be reevaluated to ensure that infected women receive adequate treatment 30 days or more before delivery. The GMH practice of repeat STS screening between 34 and 36 weeks’ gestation offers an advantage of detecting more syphilis infections late in pregnancy; however, this strategy misses earlier opportunities for screening and treatment and may not allow sufficient time for the fetus to be adequately treated. Recent national guidelines 15 now specify that providers serving high-risk populations should ensure that a repeat STS is conducted at the first opportunity during the third trimester (i.e., 28 weeks) as well as at delivery. Third, although most CS cases were not due to clinician management error, a small number could have been prevented had clinicians provided appropriate penicillin treatment with desensitization to infected women with penicillin allergy and properly recognized and managed symptomatic maternal infection.
Efforts should also be made to better coordinate diagnosis and treatment of syphilis in pregnant women among prenatal care clinics, hospital emergency clinics, and health department STD clinics. Improved collaboration of these facilities could help ensure correct serologic histories, timely referrals for treatment, prompt treatment of sex partners of infected women, and more accurate diagnosis and surveillance of CS. 18
Although we focused primarily on missed opportunities for CS prevention within prenatal care, we also noted that 44% of women who received prenatal care visited hospital emergency clinics during pregnancy. Even among women receiving no prenatal care, approximately one third visited hospital emergency clinics and had an STS during pregnancy, suggesting the presence of additional opportunities for CS prevention. Therefore, emergency clinics serving high-risk populations should routinely screen pregnant women for syphilis, especially women with HIV or other STD diagnoses, and refer these women to prenatal care and other needed health services.
The major strengths of our study include the use of a comprehensive medical record review and multiple data sources from a large consecutive series of CS cases from a single institution. However, these findings are subject to at least three limitations. First, levels of screening and treatment for syphilis during prenatal care reported here cannot be generalized to the entire hospital obstetric population. This evaluation, by design, focused on the small fraction of obstetric patients with untreated or inadequately treated syphilis. The exclusion of both obstetric patients who were successfully treated for syphilis during pregnancy and those who were not infected with syphilis during pregnancy may have masked effective CS prevention strategies within prenatal care and, therefore, resulted in an overestimate of missed opportunities for CS prevention. Second, our reliance on medical record review from multiple sources to assess reasons for CS outcome may be subject to bias because syphilis testing and treatment histories that were available to investigators at the time of the study may have differed from those that were available to clinicians at the time prenatal care was provided. Third, our finding that opportunities for CS prevention appeared to be strongest among case patients with fewer prenatal visits may be exaggerated. Although CS cases resulting from potentially preventable causes were more common among case patients with fewer prenatal visits, we can never be certain that CS would not have occurred in these case patients as the result of other factors (e.g., reinfection, late infection), had appropriate screening or treatment been administered.
The prevalence of HIV infection among women delivered of live or stillborn neonates with CS (6.2%) is several-fold higher than that of the hospital’s obstetric population. 19 As a result, every case of prenatal syphilis should be considered a potential case of HIV infection and a potential opportunity for HIV prevention counseling. The finding that more than one third of prenatal syphilis infections represented repeat infections also underscores the importance of client-centered HIV risk-reduction counseling to address behaviors that may place pregnant women at increased risk for acquiring STDs, including HIV infection.
This study identified CS prevention opportunities that were missed during prenatal care in a high-risk obstetric population. Improved prenatal screening and treatment practices can effectively prevent CS, and CS prevention strategies identified here should be considered in other obstetric populations at high risk for syphilis infection during pregnancy. Additionally, as demonstrated in this study, CS is an important sentinel marker of other adverse lifestyle and health conditions. Psychosocial, behavioral, and economic problems such as mental illness, substance abuse, violence, poverty, and HIV infection are characteristic of this population and undoubtedly influence patient use of healthcare services. These issues cannot be adequately addressed by provider guidelines for prevention of CS alone. Interventions for CS that directly confront such patient lifestyle and health issues must be considered in addition to provider-based interventions if the goal to eradicate syphilis, including CS, 20 is to be successful nationally.
1. Radolf JD, Sanchez PJ, Schulz KF, Murphy FK. Congenital syphilis. In: Holmes KK, Sparling PF, Mardh P-A, et al.,eds. Sexually Transmitted Diseases. 3rd ed. New York: McGraw-Hill, 1998.
2. Stoll BJ, Lee FK, Larsen S, et al. Clinical and serologic evaluation of neonates for congenital syphilis: a continuing diagnostic dilemma. J Infect Dis 1993; 167: 1093–1099.
3. Stoll BJ, Glover S, Freed G, Nahmias AJ. Cost of congenital syphilis. Pediatr Infect Dis J 1993; 12: 621.
4. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance, 1998. Atlanta, Georgia: US Department of Health and Human Services, Public Health Service, CDC, 1999.
5. Centers for Disease Control and Prevention. Congenital syphilis—United States, 1998. MMWR Morb Mortal Wkly Rep 1999; 48: 757–761.
6. Centers for Disease Control and Prevention. Epidemic of congenital syphilis—Baltimore 1996–7. MMWR Morb Mortal Wkly Rep 1998; 47: 904–907.
7. Southwick KL, Guidry HM, Weldon MM, Mertz KJ, Berman SM, Levine WC. An epidemic of congenital syphilis in Jefferson County, Texas, 1994–5: inadequate prenatal syphilis testing after an outbreak in adults. Am J Public Health 1999; 89: 557–560.
8. Ricci JM, Fojaco RM, O’Sullivan MJ. Congenital syphilis: the University of Miami/Jackson Memorial Medical Center Experience, 1986–1988. Obstet Gynecol 1989; 74: 687–693.
9. Desenclos J-CA, Scaggs M, Wroten J. Characteristics of mothers of live infants with congenital syphilis in Florida, 1987–1989. Am J Epidemiol 1992; 136: 657–661.
10. Webber MP, Lambert G, Bateman DA, Hauser WA. Maternal risk factors for congenital syphilis: a case-control study. Am J Epidemiol 1993; 137: 415–422.
11. Thompson BL, Matuszak D, Dwyer DM, Nakashima A, Pearce H, Israel E. Congenital syphilis in Maryland, 1989–1991: the effect of changing the case definition and opportunities for prevention. Sex Transm Dis 1995; 22: 364–369.
12. Mobley JA, McKeown RE, Jackson KL, et al. Risk factors for congenital syphilis in infants of women with syphilis in South Carolina. Am J Public Health 1998; 88: 597–602.
13. Centers For Disease Control. Guidelines for the prevention and control of congenital syphilis. MMWR Morb Mortal Wkly Rep 1988; 37(Suppl S-1): 1–13.
14. Centers for Disease Control and Prevention. 1993 Sexually transmitted disease treatment guidelines. MMWR Morb Mortal Wkly Rep 1993; 42: RR-14.
15. Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1998; 47(RR-1): 40.
16. Zenker P. New case definition for congenital syphilis reporting. Sex Transm Dis 1991; 18: 44–45.
17. Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 1998. MMWR Morb Mortal Wkly Rep 1999; 48: 873–878.
18. Yetman RJ, Risser WL, Barth BA, et al. Problems in physician’s classification and reporting of congenital syphilis. Int J STD AIDS 1998; 9: 765–768.
19. Lindsay MK, Peterson HB, Willis S, et al. Incidence and prevalence of human immunodeficiency virus infection in a prenatal population undergoing routine voluntary human immunodeficiency virus screening, July 1987 to June 1990. Am J Obstet Gynecol 1991; 165: 61–64.
20. St. Louis ME, Wasserheit JN. Elimination of syphilis in the United States. Science 1998; 281: 353–354.