The incidence of syphilis increased in the United States through the 1980s peaked in 1990, and then subsequently declined (Centers for Disease Control and Prevention, 2003a). The rate of syphilis in the United States in 2002 was 2.4 cases per 100,000 persons, and the rate among neonates was even lower at 1.1 cases per 100,000 population. The Centers for Disease Control and Prevention created a National Syphilis Elimination Plan with a goal of fewer than 1000 cases of primary and secondary syphilis in the country by 2005 (Mitka, 2000).

Increased rates of maternal syphilis have been linked to substance abuse, especially crack cocaine, and to inadequate prenatal care (Lago, 2004; Minkoff, 1990; Warner, 2001, and all their colleagues). Klass and associates (1994), in a study of prenatal syphilis at the Boston City Hospital during four decades, concluded that the continued prevalence of syphilis at delivery was associated with substance abuse, HIV infection, lack of prenatal care, treatment failures, and reinfection.

Antepartum syphilis can profoundly affect pregnancy outcome by causing preterm labor, fetal death, and neonatal infection by transplacental or perinatal infection (Genc and Ledger, 2000; Watson-Jones and co-workers, 2002). Fortunately, of the many congenital infections, syphilis is the most readily prevented and the most susceptible to therapy.

In the past, syphilis accounted for nearly a third of stillbirths. Today it plays a smaller but persistent role in fetal death, especially before 30 weeks (Goldenberg and Thompson, 2003; Gust and associates, 2002). Spirochetes readily cross the placenta and can result in congenital infection. Because of fetal immunocompetence prior to about 18 weeks, the fetus generally does not manifest the immunological inflammatory response characteristic of clinical disease before this time (Silverstein, 1962). Once fetal syphilis develops, there appears to be a continuum of involvement (

Fig. 59-1). Hepatic involvement is followed by anemia and thrombocytopenia and then ascites and hydrops (Hollier and colleagues, 2001).

The frequency of congenital syphilis varies with both the stage and duration of maternal infection. McFarlin and associates (1994) reported that high serological titers and unknown duration of infection were major predictors of congenital syphilis. Thus, the highest incidence is in neonates born to mothers with early syphilis — primary, secondary, and early latent — and the lowest incidence with late latent disease of greater than 1-year duration (Fiumara and colleagues, 1952; Golden and co-workers, 2003). Importantly, any stage of infectious maternal syphilis may result in fetal infection.

Figure 59-1. Hypothesized continuum of fetal syphilis infection. (VDRL = Venereal Disease Research Laboratory.)

The newborn may have jaundice with petechiae or purpuric skin lesions, lymphadenopathy, rhinitis, pneumonia, myocarditis, or nephrosis. With syphilitic infection, the placenta becomes large and pale. Microscopically, villi appear to have lost their characteristic arborescent appearance and to have become thicker and club shaped. Sheffield and colleagues (2002c) described large villi in 64 percent of placentas from 33 pregnancies with stillborn fetuses from syphilis. More than 60 percent of placentas from 18 liveborn neonates also had large villi. Blood vessels markedly diminish in number, and in advanced cases, they almost entirely disappear as a result of endarteritis and proliferation of stromal cells. Perhaps related to this, Lucas and co-workers (1991) demonstrated increased vascular resistance in uterine and umbilical arteries of infected pregnancies. In a study of 25 untreated women, Schwartz and associates (1995) reported that necrotizing funisitis was present in a third of cases. Spirochetes were detected in almost 90 percent using silver and immunofluorescent staining.

A suitable serological screening test such as the Venereal Disease Research Laboratory (VDRL) slide test or the rapid plasma reagin (RPR) test should be performed at the first prenatal visit. Testing is required by law in many states and is probably cost-effective even if the prevalence is low (Connor and colleagues, 2000; Hollier and associates, 2003). Serological tests yield positive results in the majority of women with primary syphilis and in all of those with secondary and latent syphilis. Because such reagin tests lack specificity, a treponemal test is used to confirm a positive result (Pope and Fears, 2000; Young, 2000). These include the fluorescent treponemal antibody absorption test (FTA-ABS), the microhemagglutination assay for antibodies to Treponema pallidum (MHA-TP), or the Treponema pallidum passive particle agglutination (TP-PA) test.

Cord blood screening is an insensitive test to detect early congenital syphilis. For women at high risk for syphilis, a nontreponemal screening test should be repeated in the third trimester and at the time of delivery (Centers for Disease Control and Prevention, 2002d; Lumbiganon and co-workers, 2002).

Polymerase chain reaction (PCR) is specific for detection of Treponema pallidum in amnionic fluid, and treponemal DNA has been found in 40 percent of pregnancies infected before 20 weeks (Nathan and colleagues, 1997; Wendel and associates, 1991). Fetal syphilis has also been verified by amnionic fluid darkfield examination or rabbit infectivity testing in 64 percent of a cohort of women with untreated syphilis (Hollier and co-workers, 2001). Although prenatal diagnosis can be made by funipuncture or amniocentesis, its clinical utility is not yet clear.

Because of a myriad of pathological changes in the fetus, ultrasonographic examination may be suggestive or even diagnostic of congenital infection. For example, the neonate shown in

Figure 59-2 has a large abdomen due to marked hepatosplenomegaly from congenitalsyphilis. The placenta was thickened and weighed almost as much as the neonate. The fetus may also have edema, ascites, or hydrops. Whenever possible, fetal evaluation for signs of syphilis should precede, but not delay, treatment after 20 weeks’ gestation. Barton and colleagues (1992) found, however, that delivery followed by neonatal treatment may be the best option for near-term hydropic fetuses with congenital syphilis.

Figure 59-2. Congenital syphilis. A. A 29-week-old, severely ill neonate with an enlarged abdomen caused by marked hepatosplenomegaly plus ascites. B. The large syphilitic placenta of the same neonate weighed 1200 g, almost the birthweight of the newborn.

Syphilis therapy during pregnancy is given to eradicate maternal infection and to prevent congenital syphilis. Penicillin is the treatment of choice even though it has not been evaluated rigorously. Rolfs (1995), Walker (2001), and Wendel and co-workers (2002) concluded that there are few useful comparative data available for ascertaining optimal therapy.

In retrospective analyses, benzathine penicillin G has been shown to cure early maternal infection and prevent neonatal syphilis in 98 percent of cases (Wendel and colleagues, 2002; Zenker and Rolfs, 1990). In a study of 340 pregnant women so treated, Alexander and associates (1999) reported six cases (1.8 percent) of congenital syphilis. Four of these six neonates were from a group of 75 women with secondary syphilis. The other two were identified in those delivered from a group of 102 women with early latent syphilis. Congenital syphilis was generally confined to neonates of women treated after 26 weeks and is likely related to the duration and severity of fetal infection. Sheffield and co-workers (2002b) reported that high maternal serological titers, preterm delivery, and delivery shortly after antepartum therapy are all risks of maternal treatment failing to prevent neonatal infection.

The currently recommended treatment guidelines for syphilis in pregnancy shown in

Table 59-1 are the same as for nonpregnant adults. As indicated, some authorities recommend a second dose of benzathine penicillin 1 week after the initial dose, especially for women in the third trimester or for those with secondary syphilis. Women with a history of penicillin allergy can be skin tested to confirm the risk of immunoglobulin E (IgE)-mediated anaphylaxis. If skin tests are reactive, penicillin desensitization is recommended and is followed by benzathine penicillin G treatment (Chisholm and associates, 1997; Wendel and co-workers, 1985).

bf There are no proven alternatives to penicillin therapy during pregnancy. Erythromycin may be curative for the mother, but it does not prevent all congenital syphilis, and it is not currently recommended as a penicillin alternative (Wendel, 1988). The cephalosporins, such as ceftriaxone, and the newer macrolide, azithromycin, may prove useful in adults (Augenbraun, 2002; Augenbraun and Workowski, 1999). Azithromycin therapy results in significant maternal and fetal serum drug levels (Ramsey and colleagues, 2003). Its efficacy in pregnancy, however, has not been adequately evaluated, and both resistance and treatment failures have been reported in adults (Centers for Disease Control and Prevention, 2004a; Lukehart and colleagues, 2004; Wendel and associates, 2002). Tetracyclines, including doxycycline, are effective for treatment of syphilis in the nonpregnant woman but are generally not recommended during pregnancy because of the risk of yellow-brown discoloration of fetal deciduous teeth.

In most women with primary syphilis and about half with secondary infection, penicillin treatment is followed by the Jarisch-Herxheimer reaction. Uterine contractions frequently develop with this reaction, and they may be accompanied by late fetal heart rate decelerations (Klein and co-workers, 1990). In a study of 50 pregnant women who received benzathine penicillin for syphilis, Myles and colleagues (1998) reported a 40-percent incidence of Jarisch-Herxheimer reaction. Of the 31 women monitored electronically, 42 percent developed regular uterine contractions with a median onset of 10 hours, and 39 percent developed variable decelerations with a median onset of 8 hours. All contractions resolved within 24 hours of therapy. Lucas and associates (1991) used Doppler velocimetry and demonstrated acutely increased vascular resistance during this time.

All women with syphilis should be offered counseling and testing for HIV (Koumans and co-workers, 2000). For women with concomitant HIV infection, the Centers for Disease Control and Prevention (2002d) recommended the same treatment as for HIV-negative persons (Augenbraun, 2002). Some authorities, however, recommend two additional weekly doses of benzathine penicillin G. Clinical and serological follow-up to detect treatment failures is also recommended at 3, 6, 9, 12, and 24 months in HIV-positive patients.

Lumbar Puncture. Whether to routinely analyze cerebrospinal fluid has been controversial. Many asymptomatic patients have abnormalities with primary, secondary, and early latent syphilis, but most do not develop neurosyphilis if appropriately treated. The Centers for Disease Control and Prevention (2002d) recommended lumbar puncture for latent syphilis of more than 1-year duration if there are neurological or ophthalmological symptoms, for treatment failures, for evidence of active tertiary syphilis (aortitis, gummas, or iritis), or for concomitant HIV infection.

Follow-Up. Any patient treated for syphilis needs conscientious follow-up. Sexual contacts within the last 3 months should be evaluated for syphilis and treated presumptively, even if seronegative. Maternal serological titers should be repeated in the third trimester and at delivery to confirm a serological response to treatment or to document reinfection in this high-risk group.

Treatment of Congenital Syphilis. Every newborn with suspected or proven congenital syphilis should undergo cerebrospinal fluid examination prior to treatment. After therapy, they should be followed at 2- to 3-month intervals until nontreponemal serological tests become negative or serofast. Infants born to mothers treated with erythromycin or a nonpenicillin regimen for syphilis during pregnancy should be retreated as though they have congenital syphilis.

Provided by ArmMed Media
Revision date: June 22, 2011
Last revised: by Jorge P. Ribeiro, MD