In 1897, Sir William Osier stated "know syphilis in all its manifestations and relations, and all other things clinical will be added unto you." In 1917, he observed that 20% of stillbirths and 22% of infant deaths in the United States resulted from syphilis.1 Unfortunately, in the postantibiotic era in this country, many physicians had been lulled into believing that syphilis was a waning disease. In fact, syphilis did wane after the introduction of penicillin in 1943. Although the incidence rose briefly again in the 1960s, syphilis had dropped to an all-time low by the early 1980s. While in most areas syphilis remains rare, it has become a disease of epidemic proportions among some populations. In some areas of the United States, clinicians have begun to "know syphilis in all its manifestations" once again.
Syphilis incidence in neonates and children closely parallels adult trends. Large differences in incidence exist between different racial and socioeconomic groups and in certain areas of the country. In the late 1980s, the incidence in black men and women rose dramatically, while in white women it remained constant and in white men it declined.2 There was a 66% increase in cases from 1985 to 1989. In Denver, Colorado, syphilis cases rose by 38% during 1993, with the greatest increases occurring in females, whites, and persons 25 to 34 years old. Concomitantly, rates for congenital syphilis increased also.3
Mothers of syphilitic infants have more pregnancies on the average. They are more likely to be black or Hispanic and of lower socioeconomic status. Also, a correlation has been shown between the abuse of cocaine and rising incidence of syphilis.4 The exchange of sex for drugs is thought to be an important contributing actor to this trend.5 Mothers of syphilitic infants also have had much fewer prenatal visits than their peers. Inadequate prenatal care prevents mothers from receiving routine syphilis screens and thus access to early intervention.
The Centers for Disease Control and Prevention (CDC) recommends that all women be screened serologically for syphilis in early pregnancy, and in areas of high risk, women should be retested in their third trimester and at delivery.6 Infants should never leave the hospital without the serologic status of their mothers determined. Any seropositive woman is considered to be infected unless clear documentation of treatment is obtained. Acceptable proof is a health department record or other medical record showing treatment and documentation of falling antibody titers. Pregnant women treated with antibiotics other than penicillin are considered to be untreated. A mother with untreated syphilis in the primary or secondary stage has a near 100% chance of infecting her newborn. Risk is still substantial during the first 4 years of untreated syphilis, after which transmission rates decline.7
Often, a mother does not know how long she has been ill, or even that she is infected. She may not know if she was adequately treated, making it difficult to determine the risk of transmission to the infant. Clinicians then are faced with the decision of whether and how to treat infants of infected mothers with inadequate histories.
THE DIAGNOSIS OF SYPHILIS
Syphilis is caused by the spirochete Treponema pallidum. In cases of symptomatic syphilis, dark-field microscopy of primary or secondary lesions or lymph node aspirates may demonstrate tréponèmes. This method of direct visualization of the tréponème is diagnostic of syphilis. The direct fluorescent antibody test for T pallidum (DFA-TP) also is used as a direct visualization technique. In infants who are not symptomatic, diagnosis becomes less straightforward.
Two types of tests are used for the serologic diagnosis of syphilis: treponemal antigen and nontreponemal antigen tests. Neither of these tests are specific for syphilis or completely sensitive, and biologic false positives (BFP) can occur. These occur more commonly in nontreponemal tests and can be acute or chronic reactions. Approximately two thirds of the patients with BFP tests have acute reactions lasting less than 6 months. These can be caused by acute febrile illnesses, by viral diseases, by a variety of other illnesses, and after vaccinations. The antibody titers in these cases are low. Chronic BFP reactions last more than 6 months and are caused by systemic disorders such as collagen vascular disease, chronic hepatitis, and drug addiction. In the presence of rheumatoid factor or positive antinuclear antibody, even treponemal tests may be positive. Though often reported, it is uncertain whether pregnancy itself causes a preponderance of BFPs.7
Nontreponemal tests listed in Table 1 include the microscopic Venereal Disease Research Laboratory (VDRL) test and the macroscopic rapid plasma reagin (RPR) test. They detect immunoglobulin (Ig) M and IgG antibody to certain antigens. These tests are inexpensive, easy to perform, and offer antibody titers that may be correlated to clinical disease. Rising titers indicate new infection, reinfection, reactivation of old infection, or treatment failure. Titers will fall with appropriate therapy. Sequential tests should be performed by the same laboratory, using the same serologic test each time. Although both tests are equally valid, RPR titers often are higher than those of the VDRL. A disadvantage is that these tests may be insensitive in very early or very late disease. Also, infants may show positive titers through passive antibody transfer from infected mothers, although the child is not truly infected.
Diagnostic Tests for Syphilis
Treponemal tests such as the fluorescent treponemal antibody absorbed (FTA-ABS) and the microhemagglutination assay for antibody to T pallidum (MHA-TP) are used to confirm positive nontreponemal tests. They measure IgM and IgG antibody directed against T pallidum using indirect immunofluorescence. These tests are the first to become positive and remain positive even after effective treatment. Thus, they are not useful for following treatment response. They are useful, however, for confirming nontreponemal tests and diagnosing congenital and late syphilis; MHA-TP is very specific for syphilis and FTA-ABS is relatively specific (and much more widely available).
Dark-field microscopy, DTA-TP, and VDRL of amniotic fluid are helpful in the diagnosis of congenital syphilis. Newer tests are being developed for the diagnosis of syphilis, but are either not widely available or are cost prohibitive at this time. Most lack sensitivity or specificity. These tests include FTAABS IgM, Western blot to detect IgM, and polymerase chain reaction.8
DIAGNOSTIC APPROACH TO THE INFANT WITH SUSPECTED CONGENITAL SYPHILIS
Transplacental infection may occur any time during the pregnancy. Although extremely rare, it can be acquired by contact with a mucosal lesion at birth.9 Infants should be evaluated if they are born to seropositive mothers with positive nontreponemal or treponemal tests if they meet the criteria in Table 2.6 Screening at delivery should be performed on maternal serum, not cord blood. Cord blood has many false-negative results. If syphilis was acquired late in pregnancy, the RPR will be negative in cord blood even though the infant may be infected.
Who Should Be Evaluated?*
If the mothers meet the criteria in Table 2, thorough evaluation of the newborn should take place. Evaluations should include a thorough physical examination, quantitative nontreponemal tests on infant blood, long bone radiographs, complete blood count and liver function tests, bilirubin, and cerebrospinal fluid (CSF) analysis.
Physical examination of the infant is performed to look for manifestations of early syphilis. Birthweight may be <2500 g. Hepatomegaly with or without splenomegaly also is common.10 Rhinitis, coryza, and snuffles (a profuse nasal drainage that persists longer than the common cold and can be bloody) usually appear in the first week of life, but can occur as late as 3 months. Other later manifestations of early congenital syphilis include the syphilitic rash, which appears 1 to 2 weeks after rhinitis. This typically is a maculopapular rash on the hands and feet and is followed by desquamation. Fissures around the mucous membranes of the mouth, nares, and anus, although rarely seen, are highly characteristic. Eyes should be examined carefully for evidence of chorioretinitis, glaucoma, and uveitis if clinically evident syphilis is diagnosed. Obviously, many of the physical signs of congenital syphilis may not be present in the immediate newborn period and will not help a clinician make a decision to treat.
Long bone radiographs may be more helpful initially. Approximately 90% of infected symptomatic infants and 20% of asymptomatic infants will show pathologic changes.11 Although the lesions are usually seen at birth, they may begin to appear after the first few weeks. The femur and humerus are involved most often. Typically seen are areas of irregular density producing a moth-eaten appearance in the metaphysis. Metaphyseal lucent bands are highly suggestive of syphilis. The epiphyseal margin may have a ragged appearance, secondary to calcification. Epiphyseal separation may occur, as well as periostitis. These lesions are painful and may cause pseudoparalysis.
Hematologic findings may include anemia, thrombocytopenia, leukopenia or leukocytosis, and Coombs negative hemolytic processes. Liver function tests may be elevated as well as bilirubin. A chest radiograph showing a fluffy diffuse infiltrate is consistent with pneumonia alba, a common finding in congenital syphilis. Pancreatitis and nephritis are seen sometimes.
Cerebrospinal fluid studies should include VDRL (RPR cannot be done on CSF), cell count, glucose, and protein. Because of widely varied values for the interpretation of CSF, the only specific test is the VDRL. The CDC states that infants with evidence of increased CSF white cells or protein should be treated, regardless of serology.6 It is difficult to tell if an infant has developed central nervous system disease since early VDRLs may be negative even in the presence of central nervous system involvement. The treatment for congenital syphilis therefore includes a regimen for neurosyphilis.
WHO TO TREAT
Infants should be treated if their mothers had untreated syphilis at delivery, had syphilis treated less than 30 days prior to delivery, or had evidence of relapse or reinfection after treatment. Other criteria are listed in Table 3. Infants whose follow-up is not certain should be treated.
TREATMENT OF CONGENITAL SYPHILIS
The CDC recommends treatments with aqueous crystalline penicillin G 100 000 to 150 000 U/kg/day given as 50 000 U/kg intravenously every 12 hours for the first 7 days and then every 8 hours thereafter. Treatment is for 10 to 14 days. An alternative course would be procaine penicillin G 50 000 U/kg/day intramuscularly for 10 to 14 days. If more than 1 day of treatment is missed, the entire course should be restarted.
Who Should Be Treated?*
In an infant with a normal evaluation whose mother was treated with erythromycin, treated less than 1 month before delivery, or treated adequately but has yet to show an appropriate serologic response, a shorter course of penicillin can be used. A single dose of benzathine penicillin G 50 000 U/kg may be given intramuscularly. Infants with a completely normal evaluation and assurance of good follow-up may be followed serologically without treatment.
Any seroreactive infant (or infant whose mother was seroreactive at delivery) who is not treated should be followed carefully. Examinations should be performed at 1, 2, 3, 6, and 12 months. Nontreponemal tests should decline by 3 months and be nonreactive at 6 months if infants have acquired passive antibody from their mothers. If antibody titers remain stable or increase, the infant should be reevaluated, including CSF tests, and treated. Antibodies may persist for 1 year; if present at more than 1 year; the infant should be treated.
Treated infants also should be followed very carefully at 2 to 3 months to assure falling titers. Nontreponemal tests should be nonreactive by 6 months. Infants who exhibited CSF pleocytosis initially should have CSF examined every 6 months until normal. If the pleocytosis has not resolved in 2 years, the infants should be retreated. Infants with reactive CSF VDRLs at 6 months also should be retreated.
LATE CONGENITAL SYPHILIS
Congenital syphilis is a result of hematogenous infection of the entire body. It causes an inflammatory response in perivascular tissue. AU organs of the body can be involved.
Late manifestations are a result of scarring from early untreated systemic disease. Approximately 40% of surviving infants will have late manifestations if untreated.10 Keratitis may occur even in the face of adequate therapy.
Interstitial keratitis is the most common late finding, usually appearing close to puberty. It manifests as a ground glass appearance in the cornea with vascularization of the sclera. These become bilateral and cause blindness. Eighth nerve deafness may appear at 8 to 10 years of age. Paresis is the more common form of neurosyphilis seen in congenital syphilis.
Bone changes such as frontal bossing, short maxilla, and saddle nose can be seen. Destruction of the hard palate is nearly pathognomonic of congenital syphilis. Hutchinson's teeth (notched central incisors) can be seen even in unerupted permanent teeth with the use of dental radiographs. Painless arthritis of the knees (Clutton's joint) and saber shin are rare, but can be seen. Radiographs are helpful in differentiating late congenital syphilis from acquired syphilis because the bone changes in congenital syphilis generally are not seen in acquired syphilis.
Nonspecific presentations in children after the newborn period may be manifested by failure to thrive, persistent fevers, hepatitis, or aseptic meningitis. Children also may exhibit nonspecific skin rashes.
Very little is written about acquired syphilis in children. It can be transmitted from adults to children in sexual or nonsexual ways. Transmission by casual contact is rare; intimate contact is required, usually through infected oral or genital lesions.9,12 Acquired syphilis exhibits nearly all of the features of adult syphilis, although it differs in its mode of transmission: nearly 95% is related to sexual abuse.13
Acquired syphilis in children progresses in much the same manner as adult syphilis. The initial lesion is a painless indurated chancre that can be located anywhere on the body, but is usually genital, anal, or oral. This is accompanied by local painless adenopathy. Discharge, if any, is filled with spirochetes and is highly infective. Often there is no evidence of the primary chancre.14 This phase lasts for approximately 1 to 5 weeks and heals spontaneously. The differential diagnosis is limited. Local trauma and herpes could be considered, but both are usually painful lesions, whereas the chancre is not painful.
Six to 8 weeks after the initial phase, the secondary phase begins, predominated by a generalized rash. This rash can be macular, papular, or maculopapular and usually is present on the palms and soles. It is often confused with scabies, pityriasis rosea, or tinea. Occasionally, the rash may be mild and go unnoticed. Secondary syphilis also may present with condylomata, a raised, warm, moist lesion that may resemble condyloma acuminata. In fact, in the anal region, the two may be confused.15 Mucous patches, generally white with a well-defined red border, also may be present on mucosal surfaces of the mouth or genital region. Children may present with more nonspecific findings of fever, sore throat, headache, or malaise. This phase lasts 2 to 6 weeks and resolves spontaneously.
Latent syphilis has no clinical symptoms, but the serologic tests are positive. It is classified as early latent if the syphilis is present less than 1 year and as late latent if it is present over 1 year. Two thirds of patients remain latent for life. Three to 10 years after secondary syphilis, patients may develop tertiary syphilis. This is characterized by local lesions of dermal elements known as gummas. Late neurosyphilis may develop, resembling virtually any other neurologic disease. Ten to 40 years after primary disease, cardiovascular involvement of the great vessels may occur.
Because 95% of syphilis in children is acquired as a result of sexual abuse, some experts recommend routine testing of all sexually abused children.6,12,16,17 Although syphilis can be transmitted nonsexually, this is extremely rare except for transplacental passage. Because of the potential trauma to the child and the costs of testing for sexually transmitted diseases, the Section on Child Abuse of the American Academy of Pediatrics states that routine testing of all sexually abused children for syphilis and other STDs is not indicated. In most cases, testing should be done only on those children with signs and symptoms of an STD. Any case of noncongenital syphilis in children is considered to be sexual abuse until proven otherwise and should be reported to appropriate child protective services.18
Diagnosis of Acquired Syphilis
Diagnosis of syphilis for children is the same as for adults. Dark-field examination of primary or secondary lesions is diagnostic. Alternately, nontreponemal tests confirmed by treponemal tests can be used. Children diagnosed with syphilis after the neonatal period should have both their own and their mothers' medical records carefully examined for evidence of congenital syphilis.
Treatment of Acquired Syphilis
Children presenting with syphilis beyond the newborn period should have a cerebrospinal fluid VDRL test and routine cerebrospinal fluid studies to rule out neurosyphilis. If eye symptoms are present, ophthalmology consults should be obtained.
Treatment for primary and secondary syphilis is benzathine penicillin G 50 000 U/kg intramuscularly up to a maximum dose of 2.4 million units (maximum adult dose) in one dose. Early latent syphilis is treated the same. The treatment for late latent syphilis is 50 000 U/kg intramuscularly for three doses, up to the adult dose of 7.2 million units total.
In penicillin-allergic children, skin testing should be performed to determine true allergy, followed by desensitization if necessary. Although tetracycline and doxycycline may be used in adults without evidence of neurosyphilis, they are not recommended for pregnant women and children. Erythromycin is known to be less effective than other regimens, and experience with ceftriaxone has been too brief to recommend it.6
Children with acquired syphilis beyond the newborn period should be examined at 3 months and again at 6 months for falling titers. Patients with recurrent symptoms, failure of titers to decrease fourfold, or rising titers should be retreated. For retreatment, three weekly injections of intramuscular benzathine penicillin G is recommended.
Syphilis is a disease very much on the rise in certain populations in this country. It has reached epidemic proportions in some areas. This trend should be a concern to physicians caring for children in the United States. Rates of congenital syphilis are continuing to rise. Certainly, the poor prenatal care received by many women is inhibiting our ability to prevent congenital syphilis.
All infants should have a maternal test for syphilis clearly documented before hospital discharge. In the event that maternal history of length of disease or treatment cannot be documented, the infant should be treated with a full course of antibiotics. In addition, all symptomatic infants should be treated and followed to ascertain falling antibody titers.
Acquired syphilis should be suspected in any child with either a positive test for syphilis or symptomatic syphilis in whom evidence for congenital syphilis cannot be documented. Sexual abuse must be strongly suspected even in the absence of disclosure, and the case must be reported to local child protective services. Adequate treatment and follow-up must be assured.
Clinicians should refamiliarize themselves with the many and varied presentations of syphilis. Many practitioners will be faced with diagnostic and treatment decisions while the infants are still in the newborn nursery. Occasionally, a child will escape diagnosis and present with symptomatic syphilis later. This disease should not be overlooked in the differential diagnoses of many seemingly simple childhood complaints.
1. Schub KF, Murphy FK, Patamasucon AZ. Congenital syphilis. In: Holmes KK, Mardh PA, Sparling PF, et al, eds. Sexually Transmitted Diseases. New York, NY: McGraw Hill; 1990:821-842.
2. Rolfs RT, Nakashima AK. Epidemiology of primary and secondary syphilis in the United States, 1981 through 1989. JAMA. 1990;264:1432-1437.
3. Colorado Department of Health. Colorado Disease Bulletin. 1994;21(3):l-2.
4. Webber MP, Lambert G, Bateman DA, Hauser WA. Matemal risk factors for congenital syphilis: a case-control study. AmJ Epidemiol. 1993;137:415-422.
5. Centers for Disease Control. Congenital syphilis- New York City, 1986-1988. MMWR. 1989;38:825-829.
6. Centers for Disease Control and Prevention. 1993 sexually transmitted diseases treatment guidelines. MMWR. 1993;42:21-46.
7. Zenker PN, Berman SM. Congenital syphilis: trends and recommendations for evaluation and management. Pediatr infect Dis). 1991;10:516-522.
8. van der Sluis JJ. Laboratory techniques in the diagnosis of syphilis: a review. Genitourin Med. 1992;68:413-419.
9. Rawstron SA, Bromberg K, Hammerschlag MR. STD in children: syphilis and gonorrhea. Genitourin Med. 1993;69:66-75.
10. Gutman LT. Syphilis. In: Feigin RD Cherry JD eds. Textbook of Pediatric Infectious Diseases. 3rd ed. Philadelphia, Pa: WB Saunders Co; 1992:553-563.
11. Brion LP, Manuli M, Rai B, Kresch MJ, Pavlov H, Glaser J. Long-bone radiologic abnormalities as a sign of active congenital syphilis in asymptomatic newborns, ftdiotrics. 1991;88:1037-1040.
12. Neinstein LS, Goldenring J, Carpenter S. Nonsexual transmission of sexually transmitted diseases: an infrequent occurrence. Pediatrics 1984;74:67-76.
13. Ginsberg CM. Acquired syphilis in prepubertal children, ftdiatr infect Dis ]. 1983;2:232-234.
14. Lowy G. Sexually transmitted diseases in children. ftdiacr Dermatol. 1992;9:329-334.
15. Goldenring JM. Secondary syphilis in a prepubertal child: differentiating condyloma lata from condyloma acuminata. NY State J Med. 1989;89:180-181.
16. White ST, Loda FA, Ingram DL, Pearson A. Sexually transmitted diseases in sexually abused children. Pediatrics. 1983;72:16-20.
17. Lande MB, Richardson AC, White KC. The role of syphilis serology in the evaluation of suspected sexual abuse. Pediatr infect Dis ). 1992;1 1:125-127.
18. Committee on Child Abuse and Neglect, American Academy of Pediatrics. Guidelines for the evaluation of sexual abuse of children, ftdiatrics. 1992;87:254260.
Diagnostic Tests for Syphilis
Who Should Be Evaluated?*
Who Should Be Treated?*