Gradients in surface nanotopography were prepared by adsorbing gold nanoparticles on smooth gold substrates using diffusion technique. Following a sintering procedure the particle binding chemistry was removed, and integration of the particles into the underlying gold substrate was achieved, leaving a nanostructured surface with uniform surface chemistry. After pre-adsorption of human fibrinogen, the effect of surface nanotopography on platelets was studied. The use of a gradient in nanotopography allowed for platelet adhesion and activation to be studied as a function of nanoparticle coverage on one single substrate. A peak in platelet adhesion was found at 23% nanoparticle surface coverage. The highest number of activated platelets was found on the smooth control part of the surface, and did not coincide with the number of adhered platelets. Activation correlated inversely with particle coverage, hence the lowest fraction of activated platelets was found at high particle coverage. Hydrophobization of the gradient surface lowered the total number of adhering cells, but not the ratio of activated cells. Little or no effect was seen on gradients with 36. nm particles, suggesting the existence of a lower limit for sensing of surface nano-roughness in platelets. These results demonstrate that parameters such as ratio between size and inter-particle distance can be more relevant for cell response than wettability on nanostructured surfaces. The minor effect of hydrophobicity, the generally reduced activation on nanostructured surfaces and the presence of a cut-off in activation of human platelets as a function of nanoparticle size could have implications for the design of future blood-contacting biomaterials.