Until now only non-treponemal tests (e.g. rapid plasma reagin [RPR]) have been used to monitor syphilis activity (e.g. distinguishing between treated, untreated and repeat disease) and efficacy of treatment. However, they usually require manual operation and are less specific than treponemal tests. The aim of the current study was to evaluate the use of the antitreponemal IgM testing in the diagnosis of early and repeat syphilis in HIV-infected and non-infected patients. One hundred and seventeen patients with early syphilis were included in this prospective study. RPR and anti- Treponema pallidum-IgM (TP-IgM) tests were conducted at onset and at three-month intervals during 24-month follow-up after initial treatment. In 31 of 117 syphilitic patients the co-occurrence of HIV infection was diagnosed. A positive TP-IgM test was present in 78.6% of patients with newly-diagnosed primary syphilis, 95.8% with secondary and 57.9% with early latent syphilis, but only in 38.5% patients with syphilis reinfection. There was a significant correlation between primary and secondary syphilis, higher baseline RPR titre and the pre-treatment IgM test reactivity. Regardless of the syphilis stage, HIV-seropositive individuals were more frequently positive for TP-IgM, both during the first onset of the disease (90.3%), and reinfection (71.4%), as compared to the HIV-seronegative group (71.4% and 0%, respectively, P < 0.03). TP-IgM seroreversion was observed in 115 out of 117 patients studied (98.3%) during follow-up (mean time to seroreversion 6.9 months). The time to TP-IgM seroreversion after treatment was significantly shorter in patients with early symptomatic syphilis (mean 4.9 months) when compared to early latent syphilis (7.7 months, P < 0.05). A negative TP-IgM test was found in approximately 20% and 40% of individuals with primary and early latent syphilis, respectively. The value of IgM testing in the diagnosis of syphilis reinfection is doubtful.