A Randomized, Comparative Pilot Study of Azithromycin Versus Benzathine Penicillin G for Treatment of Early Syphilis : Sexually Transmitted Diseases

Secondary Logo

Journal Logo


A Randomized, Comparative Pilot Study of Azithromycin Versus Benzathine Penicillin G for Treatment of Early Syphilis


Author Information
Sexually Transmitted Diseases: August 2002 - Volume 29 - Issue 8 - p 486-490
  • Free

DESPITE A MORE THAN 80% DECREASE in cases since 1990, 1 in the United States early (primary, secondary, and early latent) syphilis continues to be among the top 10 reportable infectious diseases and is more common than in most other developed nations. In October 1999, the Centers for Disease Control and Prevention (Atlanta, GA) announced a campaign to eliminate syphilis transmission in the United States by the year 2005. 2,3 Key elements of this campaign include enhanced disease surveillance, efforts to strengthen partnerships with involved communities, rapid interventional responses to new and continuing outbreaks, health promotion, and expanded clinical and laboratory services.

New, easy to use, effective therapeutic agents could also assist in syphilis control efforts. At present, only intramuscular injections of benzathine penicillin G or (for patients who are allergic to penicillin) a 14-day course of oral doxycycline or tetracycline is recommended for treatment of early syphilis. 4,5 The recommended penicillin regimen has the advantage of being single-dose therapy, but it requires injection of a relatively large volume (4 ml) of medication and may not be used for the nearly 10% of patients who report penicillin allergy. 6,7 For penicillin-allergic patients the doxycycline and tetracycline regimens may be less effective than penicillin, create concerns about compliance with the medication, and are not recommended for treatment of pregnant patients.

Azithromycin is an azalide antimicrobial agent with a long (68-hour) half-life that has been used as oral single-dose therapy for sexually transmitted diseases (STDs) including chlamydial infections, 8 nongonococcal urethritis, 9 chancroid, 10 and gonorrhea. 11 The drug has been shown to be effective in multiple doses for treatment of experimental syphilis in rabbits 12 and in a small number of patients with early syphilis. 13 In addition, a pilot study suggests that a single 1.0 g oral dose may be effective for syphilis prevention in persons recently exposed to infected sex partners (incubating syphilis). 14

To begin to address the continuing need for alternatives to currently recommended therapy, we performed a pilot study evaluating 2.0 g of azithromycin, administered orally as either a single dose or as two doses one week apart for treatment of patients with early syphilis.


Patients with early syphilis at STD clinics in Birmingham, Alabama, and New Orleans, Louisiana, were referred to the investigators for evaluation and possible enrollment in the study. They were eligible for study participation if they had early (primary, secondary, or early latent) syphilis.

For this study, primary syphilis was defined on the basis of positive dark-field microscopy for Treponema pallidum on lesion exudate. Secondary syphilis was defined as a clinically typical or dark-field-positive cutaneous eruption and rapid plasma reagin (RPR) titer ≥1:8, and a reactive microhemagglutination T pallidum (MHA-TP)test or fluorescent treponemal antibody–absorption (FTA-ABS) test. Early latent syphilis was defined as a reactive RPR test, a reactive MHA-TP or FTA-ABS test, and either a clear history of clinical manifestations typical of primary or secondary syphilis within the past year, health department documentation of exposure to a known case of primary or secondary syphilis diagnosed within the past year, or a negative syphilis serological test within the past year. 4 Additional enrollment criteria included age of ≥18 years and willingness to return for study-related follow-up visits for 1 year.

Exclusion criteria for the study were pregnancy; breast-feeding; allergy to β-lactam or macrolide antibiotics; history of intravenous drug abuse; history of use of antibiotics active against T pallidum or use of an investigational drug in the 30 days preceding enrollment (use of quinolone, sulfonamide, trimethoprim, metronidazole, and spectinomycin antibiotics was allowed); known or suspected coexistent STDs requiring treatment with drugs effective against T pallidum; advanced HIV infection manifested by a history of opportunistic infection; known severe liver or renal disease; and unreliability (as considered likely by the investigators) for participating in the study procedures and prompt follow-up.

This study was reviewed and approved by the institutional review boards of the University of Alabama at Birmingham and Louisiana State University.

Treatment and Follow-Up

Following provision of written informed consent, participants were randomly allocated by means of a computer-generated randomization code to receive one of three treatment regimens: azithromycin, 2.0 g administered as a single oral dose; two 2.0-g oral doses of azithromycin, administered 6 to 8 days apart; or benzathine penicillin G, administered as either 2.4 million units intramuscularly once in Birmingham or twice, 7 days apart, in New Orleans (the latter was the standard treatment for early syphilis in the New Orleans Health Department STD clinics at the time of this study).

Participants were seen at 7 and 14 days after initiation of treatment and then 1, 3, 6, 9, and 12 months after initiation of therapy. At each visit patients provided an interval history of sexual exposure, were clinically evaluated for persistent or recurrent syphilis, and had a serum specimen obtained for syphilis serological testing. Consenting patients who were initially HIV-seronegative were retested for HIV at 6 and 12 months. Urine pregnancy tests were performed for all women at the initial visit.

After therapy, individual participants underwent locally performed serological testing for syphilis, which provided results soon after each visit. For study-related evaluation of therapeutic response, sera were stored frozen and shipped to Birmingham, where all sera for each patient were tested at the same time by a technician masked to all clinical data in order to minimize any potential effect of day-to-day variation in serological test performance. Participants received payment at the time of each study visit to reimburse them for their efforts related to study participation.


For evaluation of response to therapy, aliquots of sera were stored, and all serum specimens obtained from each patient over the duration of the study were tested at a single time using RPR (Macro-Vue; Becton Dickinson, Cockeysville, MD) and FTA-ABS tests according to standard procedures. 15 The highest titer measured on day 0, 7, or 14 was used as the baseline titer for evaluation of response to therapy. Response to therapy was determined with use of RPR titer data from the visits in month 3, 6, 9, and 12.

Therapeutic response was defined as follows. Cure was resolution of all signs and symptoms of syphilis, including all suspicious lesions present at baseline, and either a negative RPR titer or a ≥4-fold decrease in RPR titer. Clinical response/serological nonresponse (serofast status) was resolution of signs and symptoms of syphilis present at baseline and either no change in RPR titer or a twofold (one-dilution) decrease or increase in RPR titer. Failure was the appearance of new diagnostic (i.e., dark-field-positive) lesions or a greater than fourfold (two-dilution) increase in RPR titer, without a definite history of interval exposure to an infected sexual partner.

Data were analyzed using SAS (version 8; SAS Institute, Cary, NC), and dichotomous variables were compared using Fisher exact test. For evaluation of differences in the distribution of baseline characteristics or side effects between groups, P values were computed as two-tailed. For evaluation of response to therapy, participants in each treatment group were dichotomously classified as response/nonresponse, and relative risks and 95% CIs were reported as measures of association, with evaluation of the difference between treatment regimens. One-sided P values were calculated for these analyses. For all calculations, P values ≤0.05 were considered significant.


From October 1995 through December 1997, 74 patients with early syphilis were enrolled in the study (Table 1). The average age of participants was 30 years (range, 18–56). Forty-one participants (55%) were male and 69 (93%) were black. At enrollment, 30 patients (41%) had primary syphilis, 24 (32%) had secondary-stage syphilis, and 20 (27%) had early latent syphilis. Seventy-three (97%) of 74 participants had reactive RPR and FTA-ABS tests; a single patient whose RPR/FTA-ABS tests were nonreactive and who was positive for primary syphilis by dark-field microscopy was enrolled. There were no significant differences among the patients in each of the three treatment groups in terms of age, sex, race/ethnicity, and stage of syphilis. Only three study participants (4%) were HIV-seropositive, two of whom received benzathine penicillin G therapy and one of whom received two 2.0-g doses of azithromycin 1 week apart.

Characteristics of the Patients at Enrollment

Sixty patients (81%) were followed up for ≥3 months and could be evaluated for response to therapy (benzathine penicillin, 14; azithromycin, single dose, 17; azithromycin, two doses, 29). Seven (33%) of 21 patients treated with benzathine penicillin were nonevaluable because they did not return for follow-up (4 patients) or because of intercurrent use of antibiotics between the time of enrollment and the 3-month follow-up visit (3 patients). Seven (13%) of 53 patients treated with one of the azithromycin regimens were withdrawn from the study before the 3-month follow-up visit: 4 (19%) of 21 in the single-dose azithromycin group and 3 (9%) of 32 in the two-dose azithromycin group. Of these seven azithromycin recipients who were dropped before the 3-month follow-up visit, six were dropped because of failure to keep scheduled follow-up appointments and one was dropped because the tentative diagnosis of early latent syphilis at enrollment could not be verified.

In general, both therapeutic agents were well tolerated. Jarisch–Herxheimer reactions were reported by five (24%) of the penicillin-treated participants and nine (17%) of the azithromycin-treated participants (P = 0.52). Gastrointestinal adverse events were more common in the azithromycin treatment groups. Among participants who returned for at least one follow-up visit, 1 vomited about 45 minutes after his single-dose treatment with azithromycin (this patient remained in the study and was followed for 3 months after therapy, responding with a 5-dilution decrease in RPR titer), and nausea was reported by 7 (13%) of 52 patients who returned for 1- or 2-week follow-up visits; self-limited diarrhea was reported by 5 (10%) of these 52 patients. In contrast, only 1 (5%) of 19 participants who were treated with penicillin and returned for 1- or 2-week follow-up visits reported a gastrointestinal adverse event (nausea). Although patients receiving azithromycin were nearly five times more likely than penicillin recipients to report gastrointestinal side effects, the differences were not significant (relative risk [RR] = 4.75; 95% CI, 0.67–33.9;P = 0.09). In all instances, these adverse events were classified by participants as mild to moderate in severity.

All patients with clinical manifestations of syphilis such as chancres, rashes, or condylomata lata demonstrated clear improvement of lesions at the first follow-up visit, 1 week after therapy. No participant experienced the onset of new or recurrent syphilitic lesions or rashes after therapy. No seroconversions to HIV-1 were noted.

The proportion of patients with serological response to therapy was similar in all three treatment groups. Serological response rates for evaluable participants at each specified follow-up visit are shown in Table 2. When overall response rates for the three groups were tabulated, using the last evaluable date to define response, serological responses were also similar; the majority of patients who did not respond to therapy were classified in the clinical response/serological nonresponse category. Thus, using the last evaluable clinic visits, the overall response rate among benzathine penicillin–treated patients was 86% (12 of 14). One patient dropped from the study after 3 months of follow-up had not achieved a serological response at the time he was dropped, and another patient had a two-dilution increase in RPR titer at the 6-month follow-up visit and was classified as a treatment failure.

Serological Response to Therapy at 3, 6, 9, and 12 Months Following Therapy With Benzathine Penicillin G or Azithromycin

Of the participants treated with azithromycin (2.0 g) on a single occasion, 94% (16 of 17) were found to have attained a serological response to therapy at the time of their last follow-up, a rate not significantly different from the cure rate for the benzathine penicillin treatment group (RR = 0.97; 95% CI, 0.74–1.27;P = 0.75). One participant evaluable through 6 months of therapy did not have a serological response to therapy. Among participants treated with two 2.0-g doses of azithromycin separated by a 1-week interval, 83% (24 of 29) had serologically responded to therapy at the time of their last study follow-up visit (RR for cumulative cure rate in comparison with that for benzathine penicillin: 0.88; 95% CI, 0.72–1.08;P = 0.95).

Of the five participants who did not respond serologically to treatment, four (whose last study visits were in months 3, 6, and 12 [two patients] after initiation of therapy) were classified in the serological nonresponse category, and the fifth was classified in the treatment failure category on the basis of a two-dilution increase in the RPR titer. Both patients classified as treatment failures were retreated with benzathine penicillin G when the serological treatment failure was noted.


Penicillin has been the therapy of choice for syphilis since its introduction, and for >30 years the administration of long-acting benzathine penicillin has been the preferred therapy for early syphilis. 5 Single administrations of benzathine penicillin G overcome potential problems related to poor adherence with multiple-dose therapy. In addition, the drug is readily affordable, and it can be utilized for treatment of pregnant women. The discomfort associated with the relatively large-volume, deep intramuscular injection and the relatively high prevalence of self-reported penicillin allergy 6,7 may compromise its use in some settings. The tetracycline and doxycycline regimens recommended for penicillin-allergic patients are somewhat less effective. 4 There is no recommended alternative to penicillin for treatment of syphilis in pregnant women. 4

This pilot study suggests that single-dose oral treatment with 2.0 g of azithromycin may be as effective as benzathine penicillin injections for treatment of early syphilis. In relatively few patients, the outcome of treatment with a single 2.0-g dose of azithromycin or two 2.0-g doses given 1 week apart was not significantly different from the outcome of treatment with benzathine penicillin G in a comparable group of patients. In addition, the serological responses seen in this study were similar to those in a large multicenter syphilis treatment trial in which the same dose of benzathine penicillin was used. 16 Furthermore, there was no significant difference in response rates between patients treated with single doses of azithromycin and those who received two doses of medication a week apart.

Our treatment results are also consistent with those in previous animal studies, 12 in a small study of multiple-dose azithromycin therapy for early syphilis, 13 and in a small randomized trial of azithromycin versus benzathine penicillin for prevention of syphilis in exposed patients. 14 Taken in combination, these results suggest that further evaluation of single doses of azithromycin for syphilis treatment is warranted.

In this study azithromycin therapy was also relatively well tolerated: although 13% of participants noted nausea and 10% reported loose bowel movements following therapy, these difficulties were classified by patients as mild to moderate in severity. Furthermore, none of the 32 participants randomized to receive two 2.0-g doses of azithromycin was hesitant to take the second dose of medication because of previously experienced side effects. Given anecdotal reports that the gastrointestinal side effects associated with azithromycin vary with the formulation of drug ingested (i.e., greater side effects with the sachet formulation than with capsules, which in turn are less-well-tolerated than the currently available tablet formulation), both the rate of gastrointestinal side effects and formulation of drug utilized should be further evaluated in future studies.

Although our results are encouraging, a number of limitations of this pilot study should be acknowledged and considered in future studies. In a study utilizing a rabbit model of early syphilis, 12 similar cure rates were noted for penicillin- and azithromycin-treated rabbits, but experimental syphilitic chancres resolved somewhat more slowly in the azithromycin-treated rabbits. Lesion-resolution rates were only subjectively evaluated in this study and appeared comparable. Nonetheless, careful measurement of lesion resolution should be included in future studies. In addition, although syphilis is associated with increased HIV seroprevalence in the United States, only 3 (4%) of 74 participants in this trial were known to be HIV-coinfected. Thus, further evaluation of azithromycin for treatment of syphilis in patients with HIV infection is also warranted.

It might also be argued that the patients receiving treatment with the different benzathine penicillin regimens used in Birmingham and New Orleans during the time the study was conducted should be considered two comparison groups rather than one. Because no significant differences in penicillin response rates were noted between the two cities (data not shown), we chose to combine the results to increase the size of the comparison group.

Finally, definition of cure is difficult in any study of the treatment of syphilis in humans because of imprecise monitoring of serological response to therapy. Overall, 40 (87%) of the 46 patients in this trial treated with azithromycin had a two-dilution decrease in the syphilis serological test (RPR) titer, the level utilized by the CDC to define therapeutic response. 4 This response rate is comparable to that noted by Rolfs et al. 16 in the only large evaluation of benzathine penicillin as therapy for syphilis in the past 30 years.

That approximately 15% of treated patients do not have meaningful changes in serological test titers for up to 1 year after treatment is troublesome to clinicians. In addition, this relatively high nonresponse rate suggests the need for larger sample sizes in studies designed to have sufficient power to demonstrate that therapeutic regimens are comparable. Although rising serological test titers likely indicate treatment failure, the clinical significance of serological test titers that neither rise nor fall but remain unchanged is unclear and is worthy of further study.

In summary, in this pilot study the response to treatment of early syphilis with azithromycin, 2.0 g, given by mouth as either a single dose or two doses 1 week apart, appeared similar to response rates with recommended doses of benzathine penicillin G. If its efficacy is verified by further study, azithromycin may be the first new agent effective for single-dose treatment of syphilis in >30 years. As such, it has the potential to contribute to the ongoing efforts to eliminate endemic syphilis within the United States.


1. Centers for Disease Control and Prevention. Primary and secondary syphilis: United States, 1998. MMWR Morb Mortal Wkly Rep 1999; 48: 873–878.
2. Hook EW III. Is elimination of endemic syphilis transmission a realistic goal for the USA? Lancet 1998; 351: 19–21.
3. St. Louis ME, Wasserheit JN. Elimination of syphilis in the United States. Science 1998; 281: 35–36.
4. Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted diseases. MMWR Morb Mortal Wkly Rep 1998; 47 (RR-1):1–116.
5. Hook EW III, Marra CM. Acquired syphilis in adults. N Engl J Med 1992; 326: 1060–1069.
6. Lentz JW, Nicholas L. Penicilloyl-polylysine intradermal testing for penicillin hypersensitivity. Br J Vener Dis 1970; 46: 457–460.
7. Gadde J, Spence M, Wheeler B, Adkinson NF. Clinical experience with penicillin skin testing in a large inner-city STD clinic. JAMA 1993; 270: 2456–2463.
8. Martin DH, Mroczkowski TF, Dalu ZA, et al. A controlled trial of a single dose of azithromycin for the treatment of chlamydial urethritis and cervicitis. N Engl J Med 1992; 327: 921–925.
9. Stamm WE, Hicks CB, Martin DH, et al. Azithromycin for empirical treatment of the nongonococcal urethritis syndrome in men: a randomized double-blinded study. JAMA 1995; 274: 545–549.
10. Martin DH, Sargent SJ, Wendel GD, McCormack WM, Spier NA, Johnson RB. Comparison of azithromycin and ceftriaxone for the treatment of chancroid. Clin Infect Dis 1995; 20: 409–414.
11. Handsfield HH, Dalu AZ, Martin DH, et al. Multicenter trial of single-dose azithromycin vs. ceftriaxone in the treatment of uncomplicated gonorrhea. Sex Transm Dis 1994; 21: 107–111.
12. Lukehart SA, Fohn MJ, Baker-Zander SA. Efficacy of azithromycin for therapy of active syphilis in the rabbit model. J Antimicrob Chemother 1990; 25: 91–99.
13. Verdon MS, Handsfield HH, Johnson RB. Pilot study of azithromycin for treatment of primary and secondary syphilis. Clin Infect Dis 1994; 19: 486–488.
14. Hook EW III, Stephens J, Ennis DM. Azithromycin compared with penicillin G benzathine for treatment of incubating syphilis. Ann Intern Med 1999; 131: 434–437.
15. Larsen SA, Pope V, Johnson RE, Kennedy EJ, eds. A Manual of Tests for Syphilis. 9th ed. Washington, DC: American Public Health Association, 1998.
16. 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. N Engl J Med 1997; 337: 307–314.
© Copyright 2002 American Sexually Transmitted Diseases Association