The efficacy of combinations of membranes and autogenous bone grafts at immediate implants were compared in a prospective study. Sixty-two consecutively treated patients each received an immediate implant for a single tooth replacement at a maxillary anterior or premolar site. Dimensions of the peri-implant defect at the implant collar were measured as follows: vertical defect height (VDH), horizontal defect depth (HDD) and horizontal defect width (HDW). Each implant randomly received one of five augmentation treatments and were submerged with connective tissue grafts: Group 1 (n=12)--expanded polytetrafluoroethylene membrane only, Group 2 (n=11)--resorbable polylactide/polyglycolide copolymer membrane only, Group 3 (n=13)--resorbable membrane and autogenous bone graft; Group 4 (n=14)--autogenous bone graft only, and Group 5 (n=12)--no membrane and no bone graft control. At re-entry, all groups showed significant reduction in VDH, HDD and HDW. Comparisons between groups showed no significant differences for VDH (mean 75.4%) and HDD (mean 77%) reduction. Significant differences were observed between groups for HDW reduction (range, 34.1-67.3%), with membrane-treated Groups 1, 2 and 3 showing the greatest reduction. In the presence of dehiscence defects of the labial plate, HDW reduction of 66.6% was achieved with membrane use compared with 37.7% without membranes. Over 50% more labial plate resorption occurred in the presence of a dehiscence defect irrespective of the augmentation treatment used. The results indicate that VDH and HDD reduction at defects adjacent to immediate implants may be achieved without the use of membranes and/or bone grafts.