With syphilis at an all-time low in the United States—occurring at a rate of 3.2 cases per 100,000 population¹—the future holds lation—the future holds hope for eradication of this disease. The rate of syphilis declined 84 percent from 1990 to 1997¹; before further reduction can occur, primary care clinicians must become familiar with the clinical presentation and treatment of this disease.

The most common dilemmas posed by diagnosis and treatment of syphilis are how to interpret serologic studies, when to perform a lumbar puncture and how to assess the patient's response to therapy. In addition, the presence of human immunodeficiency virus (HIV) infection and acquired immunodeficiency syndrome (AIDS) has raised other concerns about the diagnosis and treatment of this disease.

Clinical Manifestations

Syphilis is often called “the great imitator” because of its wide variety of signs and symptoms, but its characteristic manifestations are usually clear-cut (Table 1).

PRIMARY SYPHILIS

The symptoms of primary syphilis become evident about three weeks after infection, although the onset of symptoms may range from three to 90 days.² One or more characteristic chancres erupt at the site of inoculation. Syphilitic chancres are classically described as painless, indurated, clean-based ulcers, unlike chancroid ulcers, which are deep, undermined and purulent, and herpetic ulcers, which are generally multiple, shallow and tender.³ Nonulcerative lesions occasionally occur. Although regional lymphadenopathy is common in primary syphilis, it is not an essential component in the diagnosis. Without treatment, the chancre usually resolves in three to six weeks.

SECONDARY SYPHILIS

If untreated, primary syphilis progresses to secondary syphilis between two weeks and two months after the appearance of the chancre. While secondary syphilis has many manifestations, a rash is the most common. Virtually any kind of rash, except a vesicular one, can occur in secondary syphilis. A maculopapular rash on the palms and soles is especially suggestive of syphilis. Ulcers and flattened eroded lesions on the mucous membranes (which are often referred to as “mucous patches”) can occur in the mouth or throat. Fever, pharyngitis, arthralgias or generalized lymphadenopathy develop in most patients with secondary syphilis.

Systemic involvement can occur in secondary syphilis and anytime thereafter. Syphilitic meningitis is usually a slowly developing, mild to moderate meningitis characterized by stiffness of the neck, headache, nausea and vomiting, unlike the rapidly evolving febrile presentation seen in bacterial meningitis.^4,5 Cranial nerve abnormalities, especially of the seventh and eighth nerves, are common in syphilitic meningitis. Other types of systemic involvement include hepatitis, uveitis and nephritis.

LATENT SYPHILIS

Latent syphilis is asymptomatic, detected only by abnormal serologic tests. Early latent syphilis, as defined by the U.S. Public Health Service, is the stage of disease that occurs up to one year after inoculation, as documented by seroconversion, clear symptoms of primary or secondary syphilis or sexual exposure to a person with primary, secondary or early latent syphilis within the preceding year.⁶

Late latent syphilis is also asymptomatic and is defined as infection for more than one year after inoculation. This distinction is important because therapies for early latent and late latent syphilis are different. In addition, relapse of infectious secondary syphilis is most likely to occur in the early latent period, whereas syphilis is unlikely to be infectious in patients who have had the disease for longer than one year.

TERTIARY SYPHILIS

Tertiary syphilis develops in 8 to 40 percent of untreated patients.⁷ It most commonly presents as neurosyphilis, which can be asymptomatic or can produce a variety of clinical syndromes, most classically tabes dorsalis and general paresis. Because differentiation of these syndromes from nonsyphilitic causes of neurologic disease can be difficult, particularly in the demented patient, familiarity with the well-described clinical syndromes is essential. Tabes dorsalis, with its characteristic wide-based “steppage” gait, signifies destruction of the dorsal roots of the spinal column. Lightning pains of the extremities, decreased peripheral reflexes and urinary and fecal incontinence can also occur. General paresis is chiefly manifested by chronic dementia but can be accompanied by other neurologic signs as well.

This constellation of symptoms is described by the mnemonic “PARESIS”: personality disturbances, affect abnormalities, reflex hyperactivity, eye abnormality, sensorium changes, intellectual impairment and slurred speech.² The ocular abnormality is manifested as Argyll-Robertson pupils (small, irregular pupils bilaterally, nonreactive to light but reactive to accommodation). While syphilitic meningitis can develop in tertiary syphilis, it more commonly occurs in earlier stages of the disease.

Cardiovascular manifestations of tertiary syphilis are uncommon. They include aortic dilatation and regurgitation as well as non-atherosclerotic coronary artery disease caused by obliterative endarteritis at the coronary ostia.² Gummatous syphilis, also uncommon, is characterized by granulomatous-like lesions of the liver, bone, skin, brain and other organs.

Laboratory Diagnosis

SERUM EVALUATION

Initial screening for syphilis is performed with one of the nontreponemal antibody tests, the VDRL test or the reactive plasma reagin (RPR) test. These tests are quite sensitive yet not very specific for syphilis. The VDRL and RPR, respectively, are reactive in 78 percent and 86 percent of patients with primary syphilis.⁸ They become positive within approximately four to six weeks after infection or one to three weeks after the appearance of the primary lesion. Thus, these tests can be negative in early syphilis, when patients have lesions. Conversely, in late syphilis, about one fourth of untreated patients have negative VDRL results.⁸ Therefore, the nontreponemal tests cannot be relied on for diagnosis when the disease is in its very early or late stage.

False-positive nontreponemal tests have a variety of causes (Table 2). ^5,8 Given the frequency of both false-positive and false-negative test results, all positive tests, even in asymptomatic patients, as well as negative tests in patients in whom syphilis is strongly suspected clinically, should be verified by a specific treponemal test.

The nontreponemal tests are quite useful for monitoring the patient's response to treatment, because the titers reflect disease activity. When these tests are used for this purpose, it is important to use the same test (either VDRL or RPR) for serial measurements because the two tests can differ significantly in their titers. When possible, it is also recommended that the same laboratory be used.

The specific treponemal tests are the micro-hemagglutination assay for Treponema pallidum (MHA-TP) and the fluorescent treponemal antibody absorption (FTA-ABS) test. The MHA-TP is positive in 76 percent of patients with primary syphilis, and the FTA-ABS is positive in 84 percent.⁸ Compared with nontreponemal tests, treponemal tests may become positive earlier in the course of infection.⁵ Titers of the treponemal tests do not correlate with disease activity and cannot be used to follow the patient's response to treatment.

VISUALIZATION OF THE SPIROCHETE

A final method of detection of T. pallidum is direct visualization of the organism by using one of several techniques. The presence of characteristic organisms in a specimen obtained from a typical lesion of very early syphilis is diagnostic, even in the absence of positive serologic tests. Dark-field microscopy of material from a genital lesion demonstrates corkscrew-like organisms moving with a spiraling motion. Oral and anal lesions cannot be used as sources for specimens because of the presence of nonpathogenic treponemes. An alternative to this technique is immunofluorescence or immunoperoxidase staining by means of specific treponemal antibodies.⁹ Because this technique is immunologic, it can distinguish pathogenic from nonpathogenic treponemes with a specificity approaching 100 percent. However, interpretation of these tests requires considerable expertise. Repeated attempts at detection are necessary before the tests can be declared truly negative.

No effective method exists for culturing T. pallidum in the clinical setting. The experimental model for detection through propagation in rabbits is impractical, expensive and ethically unacceptable, and requires too much time to be clinically useful.

CEREBROSPINAL FLUID EXAMINATION

A common dilemma in diagnostic testing for syphilis is the role of a cerebrospinal fluid (CSF) evaluation in diagnosing neurosyphilis.¹⁰ In patients with primary or secondary syphilis but no neurologic signs or symptoms, CSF evaluation is not required. Such patients are not at significant risk of neurologic complications or other recurrences when they receive the standard therapy for early syphilis. Conversely, in patients with neurologic symptoms or signs potentially attributable to syphilis, CSF examination is generally required.

The question of whether to evaluate CSF in patients with latent syphilis is a frequent clinical problem. Patients who are certain they were infected within the preceding year are considered to have early latent syphilis and generally do not require a CSF evaluation. However, patients often cannot identify the time of infection because the lesions of primary syphilis are painless and often remain unnoticed. Similarly, the rash of secondary syphilis may not be recognized as a sign of secondary syphilis. Both the lesion and rash resolve without treatment.

Thus, patients with late latent syphilis or syphilis of unknown duration present a common management problem. Those with any neurologic symptoms should generally undergo a lumbar puncture and CSF evaluation to assess whether neurosyphilis is present. Occasional exceptions occur, the most common of which is a patient with chronic dementia, in whom the diagnosis and treatment of neurosyphilis would not be expected to change the overall outcome.

Other indications for lumbar puncture in patients with latent disease are listed in Table 3.⁶ The necessity of performing a lumbar puncture in asymptomatic immunocompetent patients with late latent syphilis or syphilis of uncertain duration remains a clinical dilemma. Asymptomatic neurosyphilis, defined as CSF cellular and protein abnormalities, may be found in 10 to 30 percent of patients with latent syphilis.⁵Although relatively uncommon, this state can progress to symptomatic disease and irreversible neurologic damage. Because there is no absolute indication for lumbar puncture under these circumstances, patients should be advised of the risk of disease progression and should participate in the decision-making process.⁶

When CSF evaluation is performed, it should include determination of protein and glucose levels, cell count and VDRL. The RPR is not valid in a CSF evaluation because of a high false-positive rate. The CSF-VDRL is specific but not sensitive, so false-negative results can occur. It should not be used to exclude neurosyphilis.¹¹ Sometimes the only indication of neurosyphilis is an elevated CSF leukocyte count (more than 5 white blood cells per mm³) or a CSF protein measurement greater than 40 mg per dL (40 mg per L).¹² Low CSF glucose levels are also consistent with neurosyphilis.

The CSF should not be tested routinely for treponema-specific antibodies (MHA-TP or FTA-ABS), because false-positive tests occur frequently and, although not as common, false-negative tests occur as well.¹⁰ However, some authorities would use a negative FTA-ABS as a criterion for excluding neurosyphilis, since false-negative results are very rare.¹³

Treatment

Once the stage of syphilis is determined, choosing an appropriate treatment regimen is usually straightforward (Table 4).⁶ In patients with primary, secondary or early latent syphilis, the recommended treatment is a single intramuscular injection of benzathine penicillin G, in a dosage of 2.4 million U. In patients with late latent syphilis or tertiary syphilis that is not considered to be neurosyphilis, the recommended treatment is intramuscular benzathine penicillin G, in a dosage of 2.4 million U once weekly for three weeks. In patients with neurosyphilis, either symptomatic or asymptomatic, the recommended treatment is 2 to 4 million U of aqueous penicillin G every four hours for 10 to 14 days; an alternative regimen consists of daily procaine penicillin G injections.

PENICILLIN ALLERGY

Approximately 10 percent of adults report a history of allergy to penicillin.¹ Because penicillin is the best treatment for syphilis, the allergy history should be carefully assessed to ensure that penicillin therapy is administered whenever possible. Persons with definite penicillin allergy and primary or secondary syphilis can be treated with an alternative regimen of doxycycline, in a dosage of 100 mg orally twice daily for 10 days. This regimen is generally better tolerated than an equivalent tetracycline regimen. Erythromycin, in a dosage of 500 mg orally four times daily for two weeks, is a second alternative, although it is not as effective as doxycycline. Evidence for the effectiveness of ceftriaxone (Rocephin) is limited,^14,15 but a regimen of 1 g daily, given intramuscularly for eight to 10 days, may maintain adequate treponemacidal levels.⁶ Careful monitoring of the serologic response is necessary to ensure eradication of the organism whenever an alternative regimen is used.

Penicillin is the only antibiotic with proven efficacy in neurosyphilis and syphilis in pregnancy. Allergic patients should undergo oral desensitization followed by treatment with penicillin. Consultation with an expert is advised when serious complications are a concern. An alternative to desensitization in all patients who report a history of penicillin allergy is skin testing. Some authorities assert that penicillin is also the only appropriate treatment for patients with HIV infection and syphilis, and that penicillin-allergic patients with both diseases should be desensitized and treated with penicillin. In the absence of pregnancy or HIV infection, however, patients with latent syphilis and penicillin allergy should be treated with doxycycline or tetracycline. The duration of therapy is two weeks for early latent syphilis and four weeks for syphilis of unknown duration or late latent syphilis.⁶

MONITORING RESPONSE TO TREATMENT

All of the signs and symptoms of primary or secondary syphilis usually resolve with or without treatment. Following proper treatment, patients with primary, secondary and early latent disease have less than a 10 percent risk of progression to late complications.¹⁶ Predicting the response to treatment of neurosyphilis is somewhat more complicated. Symptoms of syphilitic meningitis and, to a lesser extent, meningovascular syphilis usually resolve with treatment. However, symptoms of tabes dorsalis or general paresis are unlikely to abate. The purpose of treatment in such cases is to arrest further disease progression.

Appropriate therapy can precipitate the Jarisch-Herxheimer reaction, particularly in patients with early syphilis.¹⁷ This reaction, which consists of fever, headache, hypotension and myalgias, is thought to be the result of an inflammatory response to the destruction of treponemes. It can be treated symptomatically with antipyretics, but there is no known preventive therapy. Patients should be advised that this reaction is not a manifestation of penicillin allergy.

Laboratory follow-up after treatment of primary or secondary syphilis consists of evaluating the VDRL or RPR titers at six and 12 months after antibiotic therapy, or more frequently if compliance with follow-up is uncertain.⁶ No studies have established the absolute laboratory criteria for successful therapy, but a large epidemiologic study indicates that patients who have primary or secondary syphilis should show a fourfold (two tube) decrease in the titers by six months, and those with early latent syphilis should have a fourfold decrease by one year.¹⁸

Successful treatment of late latent and tertiary syphilis is usually followed by changes in nontreponemal test titers. Serum VDRL or RPR titers should be rechecked at six- and 12-month intervals in these patients. It is not clear how far these titers should be expected to fall, but if they increase or if a high titer does not decrease, treatment should be considered a failure.

Whether the VDRL and RPR revert to negative after treatment depends on the stage of disease. In a 1991 study, only 63 percent of patients eventually had negative RPR results. These persons were predominantly those with a first episode of primary syphilis, in whom 72 percent had a response, or secondary syphilis, in whom 56 percent had a response.¹⁸ Most patients continue to have positive treponemal antibody tests (MHA-TP or FTA) for life.

In patients with neurosyphilis, repeat lumbar punctures should be performed at six-month intervals until the CSF pleocytosis resolves. Lack of resolution or an increase in the cell count at any time during follow-up should initiate consideration of repeat treatment. Other parameters, including VDRL results, can remain positive for more than a year after treatment of neurosyphilis.

Syphilis and HIV Infection

The clinical and laboratory presentations of syphilis in HIV–infected persons are generally similar to those in persons who are not infected with the virus. Compared with syphilis in HIV–negative patients, the disease in HIV–positive patients more commonly is associated with chancres and other ulcerating lesions.^19,20 Although standard serologic tests for syphilis are considered a relatively reliable method of diagnosis in patients with HIV infection, false-negative treponemal and nontreponemal tests occur somewhat more frequently in patients who have HIV infection.^5,21 If suspicious lesions are present in a patient with HIV infection, but serologic tests are negative, biopsy of the lesion should be performed for microscopic examination to identify spirochetes.

Although current treatment recommendations from the Centers for Disease Control and Prevention (CDC) are considered adequate,²² careful follow-up with serial serologic tests, clinical assessment and CSF examination (in the case of neurosyphilis) is necessary. There are reports of HIV–infected patients who have persistent or recurrent syphilis despite appropriate antisyphilitic therapy.^21,23 In addition, the serologic response to therapy is often less predictable in HIV–positive patients than in HIV–negative patients, particularly those with primary or secondary syphilis. This altered serologic response, however, does not necessarily indicate a poorer clinical response.

A continuing dilemma in the management of syphilis in an HIV–positive patient is when to perform a lumbar puncture. Lumbar puncture, of course, should be performed in all patients with neurologic symptoms. Current CDC recommendations are that patients with late latent syphilis or syphilis of unknown duration should undergo a diagnostic lumbar puncture. When CSF findings are consistent with syphilis, intravenous antibiotic therapy for neurosyphilis should be administered.

Syphilis and Pregnancy

A VDRL or RPR test should be obtained at least once in all pregnant patients. In areas of high prevalence, repeat screening should be performed in the third trimester and also at the time of delivery. Pregnant patients with positive screening tests should be evaluated to determine the stage of disease and treated accordingly.

Penicillin is the only acceptable treatment regimen in the pregnant patient: it treats the infected fetus as well as the mother. The Jarisch-Herxheimer reaction, although not more common in pregnancy, can cause premature uterine contractions and fetal distress.²⁴ Because the consequences of congenital syphilis, both early (stillbirth, prematurity, anemia, thrombocytopenia)²⁵ and late (deafness, mental retardation, seizures, bony deformities),²⁶ are severe, these prenatal risks should not alter the decision to treat the patient, but careful observation is necessary.

Public Health Concerns

All identified sexual contacts of patients with infectious syphilis should be thoroughly evaluated.^5,6,17 Even if no sign or serologic evidence of active disease is found, contacts of patients with primary, secondary or early-latent disease should receive prompt treatment for primary syphilis, without waiting for clinical manifestations or positive serologic test results. Patients who have had syphilis of unknown duration and who have high (greater than 1:32) nontreponemal serologic test titers are considered to be infected with early syphilis. Serologic testing should be performed in contacts of patients with late latent or tertiary syphilis, but these persons do not require treatment if the tests are non-reactive. In patients with primary syphilis, contacts should be traced back three months before the first signs or symptoms of syphilis. In patients with secondary syphilis, contacts should be traced back six months, and in patients with early latent syphilis, contacts should be traced back one year.