Two approaches (surface doping with bactericidal element and loading of antibiotic into specially formed surface microcontainers) to the fabrication of antibacterial yet biocompatible and bioactive surfaces are described. A network structure with square-shaped blind pores of 2.6 +/- 0.6 x 10(-3) mm(3) for drug loading was obtained by selective laser sintering (SLS). The SLS-fabricated samples were loaded with 0.03, 0.3, 2.4, and 4 mg/cm(2) of co-amoxiclav (amoxicillin and clavulanic acid). Ag-doped TiCaPCON films with 0.4, 1.2, and 4.0 at.% of Ag were obtained by co-sputtering of composite TiC0.5-Ca-3(PO4)(2) and metallic Ag targets. The surface structure of SLS-prepared samples and cross-sectional morphology of TiCaPCON-Ag films were studied by scanning electron microscopy. The through-thickness of Ag distribution in the TiCaPCON-Ag films was obtained by glow discharge optical emission spectroscopy. The kinetics of Ag ion release in normal saline solution was studied using inductively coupled plasma mass spectrometry. Bacterial activity of the samples was evaluated against S. epidermidis, S. aureus, and K. pneum. ozaenae using the agar diffusion test and photometric method by controlling the variation of optical density of the bacterial suspension over time. Cytocompatibility of the Ag-doped TiCaPCON films was observed in vitro using chondrocytic and MC3T3-E1 osteoblastic cells. The viability and proliferation of chondrocytic cells were determined using the MTS assay and PicoGreen assay tests, respectively. The alkaline phosphatase (ALP) activity of the SLS-fabricated samples loaded with co-amoxiclav was also studied. The obtained results showed that the moderate bacteriostatic effect of the Ag-doped TiCaPCON films is mainly manifested in the change of bacterial colony morphology and optical densities of bacteria suspensions. In contrast, the SLS-prepared samples showed a very rapid initial drug release resulting in strong bactericidal effect just from the start of the test and for as long as several days. Cytocompatibility and ALP activity tests demonstrated that it is possible to achieve a pronounced antibacterial effect compatible or even higher than that in the control sample ( standard disk loaded with the amoxicillin/clavulanic acid mixture (30 vg)) without compromising the material biocompatibility and bioactivity. (C) 2014 Elsevier B.V. All rights reserved.