University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Management Recommendations > Syphilis

Ward 86 Management Recommendations

Diagnosis and Treatment of Syphilis

updated April 2016

Contributors: Mark Jacobson, MD
Sulggi Lee, MD, PhD

How to Diagnose Syphilis

  1. Primary syphilis can present clinically as a minimally painful ulcer in the anogenital area (usually solitary with heaped borders), as a mouth ulcer that may be painful, or as proctitis. Resolution of symptoms from primary chancre and lymphadenopathy is approximately 3-6 weeks in the absence of treatment. The diagnosis can be made with a positive rapid plasma reagin (RPR) or Venereal Disease Research Laboratory (VDRL) serum titer that is confirmed by a positive serum direct test result (see below). However, 25% of patients with primary syphilis are diagnosed clinically before their RPR or VDRL results become positive. Dark-field microscopy is the confirmatory test of choice for genital lesions in primary syphilis (since treponemal or non-treponemal serologies are only 75% sensitive during primary syphilis), but few clinical laboratories have the capability to perform this procedure. When the RPR is negative in primary syphilis, the diagnosis can be confirmed in almost all cases by repeating an RPR test 2 weeks later.

  2. Secondary syphilis can present with protean findings. The classic round macules with a scaly collarette on palms and soles are often absent. Rash morphology is highly variable, or rash may be absent. Patients also may have mucosal lesions (gray plaques or ulcers) or wart-like lesions that develop in moist areas (condyloma lata). Secondary syphilis also can present with or without rash as hepatitis, ocular, or cranial nerve abnormalities (including tinnitus or hearing loss), meningitis, or nonspecific fatigue or malaise. (See section on Management of Neurosyphilis for more information on how to diagnose ocular and cranial nerve syphilis as well as syphilitic meningitis.) Many other manifestations of secondary syphilis have been reported. Some patients may have clinical findings of both primary and secondary disease at the same time.

  3. Early latent syphilis is defined as serologic evidence of syphilis infection in an asymptomatic patient who has:

    1. a newly positive RPR or VDRL result (confirmed by a positive direct treponemal result--see below) in the setting of a prior negative result that was obtained within the past 12 months, OR

    2. a 4-fold increase in RPR or VDRL titer compared with the most recent titer obtained within the past 12 months

  4. Late latent syphilis is defined as syphilis infection in an asymptomatic patient who has a positive RPR or VDRL result (confirmed by a direct treponemal test) and:

    1. has never been previously tested, OR

    2. whose most recent test was negative but was obtained >12 months ago, OR

    3. has a 4-fold increase in RPR or VDRL compared with a most recent positive titer obtained >12 months previously

  5. Neurologic manifestations can occur at ANY STAGE of infection (see section on Management of Neurosyphilis).

How to Interpret Serologic Tests

  1. The nontreponemal RPR and VDRL assays are indirect serologic tests that measure antibodies to cardiolipin. To make a syphilis diagnosis, positive RPR or VDRL titers must be confirmed by a positive result with a direct serologic treponemal test (eg, FTA-ABS, TPPA, TPHA, or MHA-TP) that detects antibodies to Treponema pallidum surface antigens. Many laboratories automatically run a confirmatory direct treponemal test whenever an indirect test result is positive.

  2. Some U.S. clinical laboratories that receive high volumes of syphilis serology tests use an automated direct treponemal test to screen for antibodies to recombinant T pallidum antigens and reverse the testing sequence.(1) They first screen submitted serum specimens with the automated direct treponemal antibody test and then recheck any reactive specimens with a nontreponemal test (ie, RPR or VDRL). This approach sometimes identifies persons who are reactive with the automated direct treponemal test but nonreactive with the nontreponemal test. In this situation, most experts recommend that a second, traditional, direct treponemal test (eg, FTA-ABS, TPPA, TPHA, or MHA-TP) be performed and that those who are positive by this second direct treponemal test be treated for late latent syphilis while those who are negative be considered false positives. This test may have limited clinical utility in populations with a high prevalence of past syphilis.

  3. HIV-infected patients, even those with long-term undetectable HIV RNA, often have persistent B-cell activation and high IgG production. Thus, they respond to treatment for syphilis with a slower decline in RPR or VDRL titer than HIV-uninfected persons. Many remain serofast (ie, have a persistently positive, low-titer RPR or VDRL result that does not change in titer by >1 dilution over time). The serofast state does not indicate treatment failure. It is compatible with cure if there was a ≥4-fold decrease in RPR or VDRL titer after treatment and the sustained serofast titer is ≤1:16. Higher sustained titers are suspicious for ongoing infection or reinfection.

  4. Comparisons of RPR and VDRL titer (e.g., to assess treatment success or determine whether there is new infection in a serofast patient) are accurate only when the same test, RPR or VDRL, is performed in the same clinical laboratory. Thus, in general, RPR titers should not be compared with prior VDRL titers, or vice versa; and titer results obtained from different clinical laboratories should not be compared.

  5. RPR or VDRL results may be negative in primary syphilis (in up to 25% of cases) but in secondary syphilis are virtually always positive and are typically >1:8. However, a false-negative RPR or VDRL can result from the prozone phenomenon, which may occur when the true titer is very high (>1:512) and an overabundance of anticardiolipin antibodies prevents antigen-antibody cross-linking. When a patient with risk factors and clinical findings consistent with secondary syphilis has a negative RPR or VDRL result, the clinical laboratory should be requested to repeat the assay with serial dilutions of the serum sample.

  6. Treponemal test results are almost always positive in late syphilis, irrespective of treatment history. However, over time, even without treatment, the RPR can yield a negative result (though a direct treponemal test result will remain positive).

  7. False-positive RPR and VDRL titers may occur with older age, pregnancy, autoimmune diseases, and viral infections (e.g., HIV). Direct treponemal test results may be falsely positive in Lyme disease and with rare endemic treponemal diseases (eg, yaws, pinta, bejel).

How to Treat

There is no evidence to suggest that treatment for syphilis in HIV coinfected patients should differ from treatment for syphilis in those without HIV. Only two adequately powered, randomized treatment trials of therapy for syphilis have even included substantial numbers of HIV-coinfected patients. One, a comparison of single-dose oral azithromycin (2 gm) vs the standard intramuscular (IM) dose of benzathine penicillin (2.4 million units) conducted in Tanzania included 170 HIV-infected patients.(2) There was no difference in outcomes between HIV-infected and noninfected patients and no difference between the treatment arms. However, T pallidum resistance to azithromycin in the United States makes azithromycin an inappropriate treatment option. The other trial, a multicenter comparison of high-dose amoxicillin plus probenecid administered 3 times daily for 10 days compared with a single standard IM benzathine penicillin dose, included 101 HIV-infected patients among a total of 541 subjects. Like the previous study, it demonstrated no difference in rates of treatment failure between HIV-infected and noninfected patients or between the two treatment arms.(3)

  1. Syphilis is highly sensitive to penicillin. No strains of T pallidum have been reported to be resistant to this drug. Penicillin always should be used for treatment of syphilis whenever it is feasible and safe.

  2. Patients with lesions suspicious for primary or secondary syphilis should be treated immediately and a serum sample should be sent at the same time for RPR or VDRL assay (this is the day of treatment titer; see How to Assess Response to Treatment, below).

  3. Treatment for primary, secondary, or early latent syphilis is a single IM dose of 2.4 million units of benzathine penicillin. An alternative for penicillin-allergic patients is doxycycline 100 mg taken orally twice daily for 14 days. There are limited data suggesting efficacy of ceftriaxone 0.25 g daily or 0.5 g every other day for early syphilis.(4) For pregnant patients, only parenteral penicillin G should be used (with desensitization if the patient is penicillin allergic). (For treatment of ocular, cranial nerve, and central nervous system complications of syphilis, see Management of Neurosyphilis recommendations).

  4. Late latent syphilis should be treated with 3 consecutive weekly IM doses of 2.4 million units of benzathine penicillin (doxycycline 100 mg orally twice daily for 4 weeks if penicillin allergic).

  5. A Jarisch-Herxheimer reaction (fever, headache, myalgias) may occur within 6 hours after initiation of therapy for early syphilis (mainly secondary syphilis) and may induce labor in pregnant patients. This reaction is treated with antipyretics.

How to Assess Response to Treatment

  1. A serum RPR or VDRL titer must be obtained at the time of treatment initiation. Otherwise, the titer test of treatment success at 6 months cannot be determined accurately. In the natural history of untreated early syphilis, titers initially rise, then peak, then subsequently drop. Thus, at the time of treatment, the titer may be on the way up or down, and a baseline titer obtained more than 2-3 days before or after treatment is administered may not be accurate.

  2. Treatment success is defined by a ≥4-fold drop in serum RPR or VDRL titer at 6 months posttreatment. In our experience, the vast majority of apparent "treatment failures" turn out to be cases of reinfection. For patients who do not have a 4-fold drop in titer at 6 months after treatment, a thorough sexual history should be taken to rule out reinfection. If reinfection is a possible explanation, the patient should be re-treated for early syphilis and monitored for serologic response.

  3. In patients who have responded clinically to treatment for primary or secondary syphilis but fail to have a 4-fold drop in titer at 6 months and in whom reinfection can be excluded, we re-treat as for late latent syphilis with 3 weekly doses of IM benzathine penicillin (see below). If this treatment fails, lumbar puncture to rule out occult neurosyphilis may be indicated. In such cases, clinicians should consult with a local public health or U.S. Centers for Disease Control and Prevention (CDC) sexually transmitted disease (STD) expert.

  4. Of note, in primary, secondary, and early latent syphilis, T pallidum often invades the central nervous system. Indirect evidence of this phenomenon (eg, elevated cerebrospinal fluid [CSF] white blood cell count or protein or a positive CSF VDRL result) can be detected in up to half of patients with early syphilis. In one study, 26% of patients with secondary syphilis who had no neurologic symptoms had T pallidum grow from culture of their CSF. Despite this high frequency of central nervous system invasion in early syphilis, it is extremely rare for patients who are treated with single-dose IM benzathine penicillin to develop tertiary neurosyphilis. The randomized trial of high-dose amoxicillin with probenecid for early syphilis (a regimen that can achieve treponemocidal CSF levels) showed no evidence of benefit.(3) This issue is discussed in more detail in the the Management of Neurosyphilis recommendations.

  5. Symptomatic late syphilis (ie, tertiary syphilis) is covered in the Management of Neurosyphilis recommendations. Other manifestations of tertiary syphilis (eg, cardiac) are exceedingly rare now in the United States; if diagnosed, clinicians should consult with local public health or CDC STD experts.


  1. Early syphilis is highly contagious; late syphilis is not. Sexual contacts of individuals diagnosed with early syphilis should be tested serologically and treated presumptively for early syphilis (even if their serologic test results are negative) as described above. We screen all patients who have risk factors for STDs (eg, condomless sex with multiple partners or with a nonmonogamous partner) by RPR at 3- to 6-month intervals. We also test patients whenever they tell us they are concerned about a recent contact, and we treat them empirically whenever a sexual contact has told the patient that the contact has been diagnosed with syphilis. We advise all patients we diagnose with syphilis to advise their sexual contacts to be tested and treated.


  1. Hoover KW, Radolf JD. Serodiagnosis of syphilis in the recombinant era: reversal of fortune. J Infect Dis. 2011 Nov;204(9):1295-6.
  2. Riedner G, Rusizoka M, Todd J, et al. Single-dose azithromycin versus penicillin G benzathine for the treatment of early syphilis. N Engl J Med. 2005 Sep 22;353(12):1236-44.
  3. Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. The Syphilis and HIV Study Group. N Engl J Med. 1997 Jul 31;337(5):307-14.
  4. Hook EW 3rd, Roddy RE, Handsfield HH. Ceftriaxone therapy for incubating and early syphilis. J Infect Dis. 1988 Oct;158(4):881-4.