To assess potential antagonist wear and survival probability of silica-infiltrated zirconia compared to glass-graded, glazed, and polished zirconia. Table top restorations made of 3Y-TZP (3Y), 5Y-PSZ (5Y), and lithium disilicate (LD) were bonded onto epoxy resin preparations. Each zirconia was divided into five groups according to the surface treatment: polishing; glaze; polishing-glaze; glass infiltration; and silica infiltration. The LD restorations received a glaze layer. Specimens were subjected to sliding fatigue wear using a steatite antagonist (1.25×106 cycles, 200N). The presence of cracks, fractures, and/or debonding was checked every one/third of the total number of cycles was completed. Roughness, microstructural, Scanning electron microscopy, wear and residual stress analyses were conducted. Kaplan-Meier, Mantel-Cox (log-rank) and ANOVA tests were performed for statistical analyses. The survival probability was different among the groups. Silica infiltration and polishing-glaze led to lower volume loss than glaze and glass-infiltration. Difference was observed for roughness among the zirconia and surface treatment, while lithium disilicate presented similar roughness compared to both glazed zirconia. Scanning electron microscopy revealed the removal of the surface treatment after sliding fatigue wear in all groups. Compressive stress was detected on 3Y surfaces, while tensile stress was observed on 5Y. 3Y and 5Y zirconia behaved similarly regarding antagonist wear, presenting higher antagonist wear than the glass ceramic. Silica-infiltrated and polished-glazed zirconia produced lower antagonist volume loss than glazed and glass-infiltrated zirconia. Silica-infiltrated 3Y and lithium disilicate restorations were the only groups to show survival probabilities lower than 85%.