Guidelines for treatment of syphilis in patients with HIV have very poor evidence base

Current guidelines for the treatment of syphilis in patients with HIV are based on limited clinical data, investigators show in the online edition of Sexually Transmitted Infections.

Even though their systematic review had broad inclusion criteria, they were only able to identify 23 studies examining the outcomes of HIV-positive patients treated for syphilis. Only two of these studies were rated as “high quality” by the investigators.

Rates of treatment failure varied considerably, and were as high as 31% for latent syphilis. However, the investigators believe that confounding factors rather than the poor efficacy of treatment are the likely explanation for this.

Since its development the 1940s, penicillin has remained the favoured treatment for syphilis. Current UK and US guidelines recommend that the infection should be treated with intramuscular injection of long-lasting benzathine penicillin G. The oral antibiotic doxycycline is an alternative therapy for patients with allergy to penicillin.

Azithromycin is another oral antibiotic that is also active against syphilis. However, its routine use is not recommended because the Street 14 strain of syphilis is naturally resistant to this antibiotic. For patients with neurosyphilis, high-dose intravenous aqueous crystalline penicillin if the preferred therapy.

Current US guidelines recommend the same therapies for HIV-positive patients, but with more intensive follow-up.

Nevertheless, there is uncertainty about the best treatment for syphilis for patients with HIV. Indeed, a survey of US doctors revealed that two-thirds used non-standard treatments for this group of patients. There are also anecdotal reports from the UK of HIV-positive patients with high CD4 cell counts and early syphilis being treated with 14 days of intramuscular injections, a therapy that is likely to involve considerable inconvenience and discomfort for patients.

Because of this uncertainty, investigators from Baltimore systematically reviewed the literature regarding syphilis treatment for patients with HIV.

They set broad inclusion criteria:

·      Syphilis diagnosis made on the basis of serology or microscopy.

·      Studies involved ten or more patients, at least one of which was HIV-positive.

·      The HIV status of patients was known at or near (within a year) of the syphilis diagnosis.

·      Both the type and duration of antibiotic therapy were documented.

·      Outcomes were reported at least six months after treatment for early syphilis and twelve months after the completion of therapy for late latent and neurosyphilis.

Their initial literature search identified 1380 studies. However, only 23 met their inclusion criteria, and just two of these studies were deemed to be of good quality.

“Our study entry criteria were not overly restrictive. In fact, we chose criteria that would allow for the least restrictive clinically meaningful interpretation of the data”, comment the investigators.

They continue, “even high-quality randomised trials did not evaluate the efficacy of syphilis treatment in HIV-positive patients as their primary outcome.”

The failure rates of treatment for early syphilis ranged from 7% to 22%. The failure rate of therapy for latent syphilis ranged from 19% to 31%, and between 27% and 28% of patients did not respond to treatment for neurosyphilis.  

“Of the studies summarised…only a few had HIV-uninfected controls”, note the authors. Although these studies HIV-positive patients were more likely to experience treatment failure than HIV-negative patients, “most comparisons were not statistically significant owing to small numbers.”

Historically, failure rates of penicillin treatment for early and latent syphilis in HIV-negative patients range between 3% and 10%.

Although the investigators consider the possibility that this treatment is less effective in patients with HIV, they believe that there are other explanations for the high failure rates seen in the studies.

Follow-up in most of the studies was below the 24 months recommended in guidelines, and “this might have inflated the rates of seriological failure.”

Underlying cerebrospinal fluid abnormalities unrelated to syphilis may also have complicated the interpretation of results for the outcome of neurosyphilis. “The main clinical measure of syphilis disease activity, serological titres, are not specific for treponemes and may not reflect the underlying microbiology”, note the researchers.

The authors conclude that the best treatment for syphilis for HIV-positive patients is unknown. Therefore, “any guideline recommendations in this population are ultimately based on limited objective data.”

Blank LJ et al. Treatment of syphilis in HIV infected subjects: a systematic review of the literature. Sex Transm Infect, online edition, 10.1136/sti.2010.043893, 2010 (click here for abstract).