Oral mucosal lesions that are associated with HIV infection can play an important role in guiding the decision to initiate antiretroviral therapy (ART). The incidence of these lesions relative to the timing of ART initiation has not been well characterized. A randomized controlled clinical trial was conducted at the GHESKIO Center in Port-au-Prince, Haiti between 2004 and 2009. 816 HIV-infected ART-naïve participants with CD4 T cell counts between 200 and 350 cells/mm3 were randomized to either immediate ART initiation (early group; N = 408), or initiation when CD4 T cell count was less than or equal 200 cells/mm3 or with the development of an AIDS-defining condition (delayed group; N = 408). Every 3 months, all participants underwent an oral examination. The incidence of oral lesions was 4.10 in the early group and 17.85 in the delayed group (p-value <0.01). In comparison to the early group, there was a significantly higher incidence of candidiasis, hairy leukoplakia, herpes labialis, and recurrent herpes simplex in the delayed group. The incidence of oral warts in delayed group was 0.97 before therapy and 4.27 post-ART initiation (p-value <0.01). In the delayed group the incidence of oral warts post-ART initiation was significantly higher than that seen in the early group (4.27 versus 1.09; p-value <0.01). The incidence of oral warts increased after ART was initiated, and relative to the early group there was a four-fold increase in oral warts if ART was initiated following an AIDS diagnosis. Based upon our findings, candidiasis, hairy leukoplakia, herpes labialis, and recurrent herpes simplex indicate immune suppression and the need to start ART. In contrast, oral warts are a sign of immune reconstitution following ART initiation.