Editorial commentary: The optimal dose of penicillin when treating syphilis in HIV-infected persons: enough, already?

Despite unambiguous and prescriptive recommendations in national guidelines [1], clinicians are still unsure of the optimal antibiotic regimen to treat human immunodeficiency virus (HIV)-infected persons with early syphilis. The recommendation to use a single 2.4-millionunit (MU) dose of long-acting benzathine penicillin G (BPG) to treat all persons with early syphilis has been viewed with suspicion, given case reports of early neurological complications of syphilis occurring in HIV-infected patients presenting after receiving this (and other) regimens [2]. In this issue of Clinical Infectious Diseases, Ganesan et al attempt to answer an important question: Are multiple doses of 2.4 MU BPG better than a single dose when treating HIV-infected persons with early syphilis? This question came about as a result of several observations. Biologically, Treponema pallidum disseminates into the central nervous system (CNS) early after infection. Lukehart et al found that 3 of 3 HIV-infected patients with early syphilis and no neurological symptoms who were documented to have T. pallidum in their cerebrospinal fluid (CSF) and were treated with a single dose of 2.4 MU of BPG failed to clear the organism from their CSF [3]; the only HIV-infected patient who cleared T. pallidum from the CSF received >1 dose of BPG. In contrast, all HIV-uninfected persons cleared T. pallidum from the CSF whether they were treated with a single dose of 2.4 MU BPG or multiple doses. The numbers were small, but this study suggested that a single dose of BPG may be insufficient to eradicate T. pallidum from the CNS inHIV-infected persons. Clinically, ≥2 doses of BPG were used to treat symptomatic neurosyphilis in the pre-HIV antibiotic era, with good clinical responses [4]. Indeed, 3 doses of 2.4 MU BPG was a recommended alternate regimen for the treatment of neurosyphilis until 1982 [5]. Its use was abandoned in the early 1980s after studies demonstrated a lack of consistent treponemicidal penicillin concentrations in the CSF of persons treated with this regimen [6]. Whether CSF treponemicidal concentrations of penicillin are necessary to cure neurosyphilis is not known; the decades of successful clinical experience with this regimen might argue against it. Nonetheless, the use of BPG for the treatment of neurosyphilis fell out of favor. Given these biological and clinical observations, there was great interest in using enhanced antibiotic regimens to treat early syphilis in the HIV era. A randomized controlled trial (RCT) consisting of oral amoxicillin and probenecid in addition to single-dose BPG therapy for early syphilis was conducted in the 1990s, but the study was underpowered to address that question in HIV-infected individuals, as only a small number of immunocompromised persons were recruited and completed follow-up [7]. Moreover, the enhanced therapy used in that study had never been a standard regimen used to treat neurosyphilis. Consequently, the questionofwhetherenhanced BPG therapy would improve outcomes remained unanswered. Ganesan et al performed a retrospective analysis of 478 cases (provided by 350 subjects) of early syphilis in HIVinfected patients followed in the US Military HIV Natural History study from 1986 to 2013. These patients were treated with 1 dose of 2.4 MU BPG (29%), ≥2 doses of 2.4 MU BPG (53%), or other regimens including non-penicillin-based regimens (18%). Early syphilis cases were identified if there was a documented seroconversion from a negative to a positive nontreponemal test in the preceding Received 15 October 2014; accepted 22 October 2014; electronically published 10 November 2014. Correspondence: Khalil G. Ghanem, MD, PhD, Johns Hopkins University School of Medicine, Division of Infectious Diseases, 5200 Eastern Ave, MFL Center Tower #378, Baltimore, MD 21224 ([email protected]). Clinical Infectious Diseases 2015;60(4):661–3 © The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals. [email protected]. DOI: 10.1093/cid/ciu890