The objective of this study was to determine the best performance of an anaerobic sequencing batch biofilm reactor (AnSBBR) based on the use of four different bed materials as support for biomass immobilization. The bed materials utilized were polyurethane foam (PU), vegetal carbon (VC), synthetic pumice (SP), and recycled low-density polyethylene (PE). The AnSBBR, with a total volume of 7.2L, was operated in 8-h batch cycles over 10 months, and fed with domestic sewage with an average influent chemical oxygen demand (COD) of 358+/-110 mg/L. The average effluent COD values were 121+/-31, 208+/-54, 233+/-52, and 227+/-51 mg/L, for PU, VC, SP, and PE, respectively. A modified first-order kinetic model was adjusted to temporal profiles of COD during a batch cycle, and the apparent kinetic constants were 0.52+/-0.05, 0.37+/-0.05, 0.80+/-0.04, and 0.30+/-0.02 h(-1) for PU, VC, SP, and PE, respectively. Specific substrate utilization rates of 1.08, 0.11, and 0.86 mg COD/mg VS day were obtained for PU, VC, and PE, respectively. Although SP yielded the highest kinetic coefficient, PU was considered the best support, since SP presented loss of chemical constituents during the reactor's operational phase. In addition, findings on the microbial community were associated with the reactor's performance data. Although PE did not show a satisfactory performance, an interesting microbial diversity was found on its surface. Based on the morphology and denaturing gradient gel electrophoresis (DGGE) results, PE showed the best capacity for promoting the attachment of methanogenic organisms, and is therefore a material that merits further analysis. PU was considered the most suitable material showing the best performance in terms of efficiency of solids and COD removal.