Primary Syphilis: Serological Treatment Response to Doxycycline/Tetracycline versus Benzathine Penicillin
Article Outline
Abstract
Background
Benzathine penicillin G is the treatment of choice for infectious syphilis, but tetracycline and doxycycline are believed to be effective second-line treatments. The objective of this study was to assess the serological response from treatment of primary syphilis with benzathine penicillin compared with doxycycline or tetracycline.
Methods
We examined rapid plasma reagin serological test results of all first-time primary syphilis patients in Alberta, Canada from 1980 to 2001 and compared treatment with single dose of penicillin with 14-day course of oral doxycycline (100 mg twice a day) or oral tetracycline (500 mg 4 times a day). Serological treatment success was defined as a minimum 4-fold decrease in baseline rapid plasma reagin test antibody titer within 6 months, or ≥8-fold decrease within 12 months, or ≥16-fold decrease by 24 months. The median time to successful response was estimated, and factors associated with treatment success were identified by unadjusted logistic regression.
Results
Of the 445 primary syphilis cases with available treatment outcome data, 420 (94.4%) received penicillin and 25 (5.6%) received doxycycline/tetracycline. The serological treatment success rate was 97.4% in the penicillin group (409/420) and 100% in the doxycycline/tetracycline group (25/25), and not significantly different. The estimated median time to serological treatment success was 72.0 days (mean
=101.7, range 10-603) in penicillin and 43.0 days (mean=78.6, range 15-334) in doxycycline/tetracycline-treated patients; however, this difference was not statistically significant (P=0.16).
Conclusion
Doxycycline/tetracycline had a similarly high serological treatment success rate when compared with penicillin in the treatment of primary syphilis.
Keywords: Canada, Doxycycline, Penicillin, Syphilis, Treatment response
The rate of infectious syphilis (primary, secondary, and early latent of <1 year duration) has increased substantially in recent years in Canada and elsewhere. In 1997, the reported rate of infectious syphilis was 0.4 per 100,000 but climbed to 5.1 per 100,000 by 2006.1 Controlling syphilis involves the early recognition and adequate treatment of patients with clinical and serological evidence of infection and management of all sexual partners. Benzathine penicillin G has long been the treatment of choice for infectious syphilis (primary, secondary, and early latent syphilis) in North America. In the absence of contraindications, tetracycline or doxycycline are considered second-line treatments in patients unable to tolerate penicillin allergies.2, 3 Doxycycline is a tetracycline derivative with better oral bioavailability, convenient twice a day dosing, and fewer gastrointestinal side effects. Doxycycline at a dosage of 100 mg every 12 hours is pharmacologically equivalent to tetracycline at 500 mg every 6 hours.4 The mean inhibitory and bactericidal concentrations against treponemes are similar for tetracycline and doxycline.5, 6 The major advantages of long-acting penicillin treatment are its safety, effectiveness, and the favorable adherence to the weekly dosing schedule of 1 to 3 weeks. Disadvantages of this regimen include the cost, pain associated with intramuscular injection, and the incidence of penicillin allergy in some patients.
Several characteristics of doxycycline or tetracycline make them attractive alternatives to penicillin, including their oral route of administration and their pharmacologic and pharmacokinetic properties. However, to date there have been limited studies on the use of these second-line agents for the treatment of primary syphilis.7, 8, 9, 10, 11, 12 A recent small study using a retrospective chart review showed good response to doxycycline among patients with early syphilis from 2 sexually transmitted disease clinics in Baltimore.9 In that study, serological failure was noted among 4 of 73 (5%) patients treated with penicillin, while no treatment failures were found among 34 patients treated with doxycycline.
The effectiveness of syphilis treatment can be monitored using serial changes in antibody titer of the nontreponemal test.13, 14, 15, 16, 17 A minimum 4-fold decrease in titer by 6 months following treatment has been used as a measure of adequate serological response.2, 3 Conversely, an increase in titer without reinfection is indicative of treatment failure.
We conducted a retrospective cohort study to compare the serological response rates of patients with primary syphilis treated with penicillin and doxycycline/tetracycline.
Methods
Syphilis Notification and Treatment in Alberta
Alberta is a Western Canadian province with a population of 3.4 million, which constitutes 10.5% of the total population in Canada.18 In the context of universal health insurance and the provincial public health program, diagnosis and treatment for sexually transmitted infections are provided free in the province. Laboratories and health care providers are legally required to report all positive cases of syphilis to the provincial Ministry of Health, Alberta Health and Wellness, using a standard Sexually Transmitted Infection Notification form. The Sexually Transmitted Infection Notification documents a client's demographic information, clinical and laboratory findings, diagnoses, and treatment based on national Sexually Transmitted Infections Guidelines.2 A positive syphilis case is diagnosed based on clinical and laboratory test results. Patients with syphilis are staged on the basis of available darkfield microscopy, and clinical and serological findings. When a nontreponemal rapid plasma reagin test was performed, a positive test result was confirmed with either the microhemaglutination test for Treponema pallidum or the fluorescent treponemal antibody absorption test.
Inclusion and Exclusion Criteria
Eligible subjects during the study period of January 1, 1980 to December 31, 2001 were all patients with a first-time diagnosis of primary syphilis who had at least 2 serological titers within 12 months (1 titer at or around the date of treatment [ie, baseline titer] and at least 1 follow-up post-treatment test). Subjects were included if they were treated with penicillin (2.4 million units in a single dose intramuscularly) or doxycycline (100 mg orally, twice daily for 14 days) or tetracycline (500 mg orally, 4 times daily for 14 days). Patients were excluded if:
Records from patients whose HIV status was undocumented were not excluded as they were expected to constitute a small proportion of study subjects, because the prevalence of HIV in Alberta is low.
Of the 863 primary syphilis cases reported during the study period, 445 cases were included in the final study sample.
Variables of Interest and Definitions
Demographics, clinical, testing, and treatment information from Sexually Transmitted Infection Notification forms were extracted. Variables of interest were demographic characteristics (age, sex, ethnicity), sexual practice, year of initial rapid plasma reagin serology, and baseline rapid plasma reagin test antibody titer (ie, titer at treatment). Ethnicity was recorded in the surveillance data as reported by patients or attributed by health care provider. The category of Canadian Aboriginal was defined as North American Indian, Métis, or Inuit. Sexual practice was inferred from patient reports about the sex of their sexual partners.
The outcome of interest was serological treatment success, defined as a decrease in the baseline rapid plasma reagin test titer since treatment initiation:
If rapid plasma reagin test titer was 1:4, 1:2, or 1:1 at baseline, a serofast stable serology (ie, ±1 dilution from baseline) between 6 months and 12 months after treatment initiation also was considered treatment success. Treatment was considered to have failed if none of the above conditions were met, including if the rapid plasma reagin test titer increased from baseline by at least 4-fold between 1 month and 6 months. Time to serological treatment success was calculated as the period between the treatment initiation date and the earliest recorded date post-treatment when the serological treatment success, as defined above, was reached.
Statistical Analysis
Pearson's chi-squared test was used to compare differences in categorical variables and Student's t-test for normally distributed continuous variables. Results were considered statistically significant at P <0.05 (2-tailed). We compared the estimated time to successful serological response between patients treated with penicillin and doxycycline/tetracycline using Kaplan-Meier curves and the log-rank test.
Results
Of the 445 primary syphilis cases with available treatment outcome data, 420 (94.4%) received penicillin and 25 (5.6%) received doxycycline/tetracycline. Overall, the median age of the study sample was 27.9 years (interquartile range [IQR]=15.3), subjects were mostly male (73.3%) and Caucasian (52.8%). There was a similar distribution of patient characteristics in each treatment group (Table).
Table. Baseline Characteristics of Primary Syphilis Patients by Treatment Group, Alberta, Canada, 1980-2001
| Benzathine Penicillin G (n=420) | Doxycycline or Tetracycline (n=25) | P Value⁎† | Excluded (n=418) | ||||
|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | ||
| Age (years) | |||||||
| Median (IQR) | 27.8(15.2) | 29.8(15.1) | .34 | 30.3(17.5) | |||
| 10-19 | 54 | 12.9 | 2 | 8.0 | .96 | 32 | 7.7 |
| 20-29 | 183 | 43.6 | 12 | 48.0 | 174 | 41.6 | |
| 30-39 | 99 | 23.6 | 6 | 24.0 | 99 | 23.7 | |
| 40+ | 84 | 20.0 | 5 | 20.0 | 113 | 27.0 | |
| Sex | |||||||
| Male | 310 | 73.8 | 16 | 64.0 | .30 | 343 | 82.1 |
| Female | 110 | 26.2 | 9 | 36.0 | 75 | 17.9 | |
| Ethnicity | |||||||
| Caucasian | 176 | 53.2 | 9 | 47.4 | .74 | 205 | 61.7 |
| Aboriginal | 110 | 33.2 | 7 | 36.8 | 79 | 23.8 | |
| Black | 11 | 3.3 | 2 | 10.5 | 7 | 2.1 | |
| Asian/South Asian | 34 | 10.3 | 1 | 5.3 | 41 | 12.3 | |
| Sexual practice | |||||||
| Heterosexual | 339 | 84.3 | 17 | 70.8 | .08 | 345 | 89.1 |
| Homo/bisexual | 63 | 15.7 | 7 | 29.2 | 42 | 10.9 | |
| Year of initial rapid plasma reagin serology and diagnosis of primary syphilis | |||||||
| 1980-1984 | 303 | 72.1 | 21 | 84.0 | .72 | 333 | 79.7 |
| 1985-1989 | 98 | 23.3 | 4 | 16.0 | 58 | 13.9 | |
| 1990-1994 | 7 | 1.7 | 0 | 0 | 6 | 1.4 | |
| 1995-1999 | 6 | 1.4 | 0 | 0 | 7 | 1.7 | |
| 2000-2001 | 6 | 1.4 | 0 | 0 | 14 | 3.3 | |
| Initial rapid plasma reagin antibody titer | |||||||
| ≤1:4 | 80 | 19.7 | 6 | 24.0 | .59 | 90 | 43.5 |
| 1:8 | 58 | 14.3 | 2 | 8.0 | 15 | 7.2 | |
| 1:16 | 59 | 14.5 | 6 | 24.0 | 24 | 11.6 | |
| 1:32 | 67 | 16.5 | 4 | 16.0 | 25 | 12.1 | |
| ≥1:64 | 143 | 35.1 | 7 | 28.0 | 53 | 25.6 | |
⁎P value associated with Pearson chi-squared test or Kruskal-Wallis test as appropriate. |
†P value comparing benzathine penicillin G and doxycycline/tetracycline cases. |
A statistically similar serological treatment success rate was observed in 97.4% (409/420) of penicillin-treated patients when compared with 100% (25/25) of patients treated with doxycycline/tetracycline (Figure 1). The median follow-up time was similar in the 2 groups (penicillin: 120.5 days [IQR=193] vs doxycycline/tetracycline: 139 days [IQR=164]; P=0.65). Of the 11 patients with serological treatment failure, all were among those treated with penicillin, and had been followed between 167 days and 1561 days. Two of these 11 patients were HIV-negative and the HIV status of the others was unknown. Re-infection could not be ruled out for one individual treated with penicillin and who had no follow-up rapid plasma reagin test except for one at 18 months post-treatment (baseline rapid plasma reagin test titer at 1:32 and 18-month post-treatment rapid plasma reagin test titer at 1:128).
Figure 1.
Case selection for analysis and treatment protocol for all first-time primary syphilis cases reported in Alberta, Canada, 1980-2001.
Figure 2 shows the time to earliest response to treatment stratified by type of treatment. The estimated median time to treatment response was 72.0 days (mean=101.7; range 10-603) in penicillin patients and 43.0 days (mean=78.6; range 15-334) in those treated with doxycycline/tetracycline (P=0.16).
Figure 2.
Serological treatment success according to treatment with benzathine penicillin G or doxycycline/tetracycline. Note: Serological treatment success was defined as a decrease in the baseline rapid plasma reagin test titer since treatment initiation by ≥4-fold by 6 months, or ≥8-fold decrease within 12 months, or ≥16-fold decrease within 24 months. If rapid plasma reagin test titer was 1:4, 1:2, or 1:1 at baseline, a serofast stable serology (ie, ±1 dilution from baseline) between 6 months and 12 months (184 days to 365 days) of treatment initiation also was considered treatment success. Treatment was considered to have failed if none of the above conditions were met, including if the rapid plasma reagin test titer increased from baseline by at least 4-fold between 1 month and 6 months (31 days to 183 days).
Discussion
Serological Treatment Success Rates
One of the key findings of our study is that serological treatment success was high in both penicillin- (97.1%) and doxycycline/tetracycline-treated patients (100.0%). The serological failure rate of 2.9% among patients treated with penicillin is consistent with the 2%-12% range of treatment failure reported in other studies,9, 19 such as a previous study in Alberta that reported a treatment success rate of over 95%.17 In another example in Tanzania, Riedner et al. used a similar 4-fold decrease in rapid plasma reagin test titer to define serological success and found serological success rates of approximately 80% with penicillin at 6 months and 94% among 328 patients with high baseline rapid plasma reagin test titer (≥1:64).20 The absence of treatment failure with doxycycline/tetracycline echoes a recent study of 1558 patients from 2 sexually transmitted infection clinics in Baltimore, which found 94.5% serologic treatment success with penicillin and 100% among 34 patients treated with doxycycline.7, 9
Time to Serological Treatment Success
The time to serological treatment success was similar in primary syphilis cases treated with doxycycline/tetracycline compared with penicillin (median=43 days vs 72 days, respectively). In the Baltimore study, successful treatment for primary, secondary, and early latent syphilis (defined as a minimum 4-fold decrease in baseline rapid plasma reagin test titer by 9-13 months) was reported within 106 days for doxycycline, compared with 137 days for penicillin.9 As with the current study, the small number of patients (n=34) treated with doxycycline likely did not allow sufficient power to demonstrate a faster serological response than penicillin.
Limitations
The results must be considered in light of several limitations. First, this study assessed only serological response rather than clinical response. Moreover, there are no data on treatment adherence to ensure that full-dose treatment with a 14-day course of oral doxycycline/tetracycline was followed by all patients. It is possible that patients who were compliant with oral treatment were more likely to return for follow-up testing. Second, because this was a retrospective study and treatment was not randomly allocated, residual confounding is an issue. However, baseline characteristics of the two treatment arms were comparable on the demographic and clinical characteristics examined. In addition, because of the record-based nature of the analysis, the measured times to serological treatment success may be an overestimate of actual times to achieve treatment response. Our calculated time to treatment response depended on the frequency of patient follow-up and quality of serology data. For individuals with inadequate serological response, it is not possible to know if they were indeed treatment failures or re-infections. Fourth, our surveillance data had limited information about HIV status, which can slow response to syphilis treatment. In fact, only 27 of the 863 primary syphilis cases had HIV status information, and of those, only 3 were known to be HIV-positive. Finally, the small number of patients treated with doxycycline/tetracycline warrants future studies with larger sample sizes. More importantly, the lack of statistical significance in treatment does not imply that clinical response to each regimen is the same.
Implications
The Canadian, US, and European guidelines for the treatment of syphilis list doxycycline and tetracycline as the second-line therapies for infectious syphilis.2, 3, 21 Although the efficacy of penicillin is established for the treatment of infectious syphilis, there is suggestion that doxycycline/tetracycline may perform just as well. The time to adequate serological response observed for doxycycline/tetracycline appears to be similar, if not faster, than penicillin. There are many other additional benefits to using doxycycline or tetracycline, including the ability to dose these drugs orally, their lower cost per course of treatment relative to penicillin and their ready availability. Fortunately, there have been no reports of resistance thus far in T. pallidum to either tetracyclines or penicillin, which would preclude the use of these drug classes from routine use for the treatment of syphilis. Perhaps the biggest concern revolves around the use of a 14-day oral agent versus the virtually 100% compliance that can be assured with a single injectable dose of penicillin. Our data should ideally be confirmed by prospective, randomized trials of doxycycline and tetracycline in treating infectious syphilis using clinical and serological outcomes, in addition to studies to evaluate the compliance level with the optimal duration of oral doxycycline treatment.
Conclusion
In summary, doxycycline/tetracycline had a similarly high serological treatment success rate when compared with penicillin in the treatment of primary syphilis.
Acknowledgments
The authors thank the staff of Alberta Health and Wellness STI Services, in particular Ruth Sutherland and Karen Sutherland for assistance with data extraction.
References
- . Reported cases of STI from January 1 to June 30, 2006 and January 1 to June 30, 2005 and corresponding rates for January 1 to December 31, 2005 and 2006. http://www.phac-aspc.gc.ca/std-mts/stdcases-casmts/cases-cas-07_e.htmAccessed June 12, 2007
- . Canadian Guidelines on Sexually Transmitted Infections 2006. Ottawa, ON: Public Health Agency of Canada; 2006;
- . Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep. 2006;55(RR-11):1–94
- . Tetracyclines and chloramphenicol. In: Mandell GL, Douglas RG, Bennett JE editor. Priniciples and Practice of Infectious Diseases. 6th edition. Philadelphia, PA: Elsevier Churchill Livingstone; 2005;p. 356–373
- . Inhibition of growth of treponemes by antimicrobial agents. Br J Vener Dis. 1971;47:407–412
- . Bactericidal activity of antimicrobial agents for treponemes. Br J Vener Dis. 1971;47:413–418
- . Doxycycline in early syphilis: a long term follow up. Indian J Dermatol. 1982;27:119–124
- . Therapeutic effect of oral doxycycline on syphilis. Br J Vener Dis. 1979;55:110–115
- . Doxycycline compared with benzathine penicillin for the treatment of early syphilis. Clin Infect Dis. 2006;42:e45–e49
- . Doxycycline. Ther Drug Monit. 1982;4:115–135
- . Tetracycline in venereal diseases. Antiseptic Madras. 1956;53:9–19
- . Treatment for early syphilis and reactivity of serologic tests. JAMA. 1972;221:471–476
- . Reinfection annular secondary syphilis: a case report. Sex Transm Dis. 1978;5:160–162
- . Treatment of primary and secondary syphilis (Serological response). JAMA. 1980;243:2500–2502
- . Serological response to syphilis treatment (A new analysis of old data). JAMA. 1985;253:1296–1299
- . Laboratory diagnosis and interpretation of tests for syphilis. Clin Microbiol Rev. 1995;8:1–21
- Serologic response to treatment of infectious syphilis. Ann Intern Med. 1991;114:1005–1009
- . Canadian Population Census, 2006. Ottawa, ON: Statistics Canada; 2007;
- . Treatment of syphilis, 1993. Clin Infect Dis. 1995;20(Suppl 1):S23–S38
- Single-dose azithromycin versus penicillin G benzathine for the treatment of early syphilis. N Engl J Med. 2005;353:1236–1244
- . European guideline for the management of syphilis. Int J STD AIDS. 2001;12(Suppl 3):14–26
This study was supported by funding from Alberta Health and Wellness and from the Public Health Agency of Canada, and approved by the University of Alberta Health Research and Ethics Board.
PII: S0002-9343(08)00610-4
doi:10.1016/j.amjmed.2008.04.042
© 2008 Elsevier Inc. All rights reserved.
