The majority of total knee replacements currently implanted present an articulation composed of two metal parts, femoral and tibial components, between which there is a polyethylene insert serving as a bearing surface. The finishing surface of the metal components is a very important factor in minimizing the polyethylene wear rate and, later, the gradual production of metal and plastic debris. Considering the role of surface roughness on volumetric wear rates, the purpose of this study is to develop a protocol for the roughness characterization of total knee prosthesis (TKP) metal components, taking into consideration a limited number of points on each surface. Six mobile TKP of different sizes (three size 2 and three size 6) were tested on a knee joint simulator to compare the wear behavior of each group. After 2 million cycles the weight loss by the polyethylene inserts was measured with gravimetric method and the surface roughness of the metallic components was assessed in terms of average surface roughness, Ra, skewness, Rsk, and total roughness, Rt. Roughness measurement involved 29 points on each femoral condyle and 26 points on each metal tibial plate. The data collected has shown an increased roughness upon wear testing for both the investigated TKP sizes. No statistical differences were observed between the two groups for both the parameters Ra, Rsk, and Rt. The surface of all metallic components became more negatively skewed, indicating diminishing peaks. The various parameters were correlated to the volumetric loss using a linear regression analysis.