The overall aim of this thesis was to investigate different therapeutic strategies in treatment of the edentulous maxilla with dental implants and their importance for treatment outcome. The introduction of one-stage surgery, in place of two-stage surgery, was a paradigm shift in the area of implant treatment since submerged implant healing underneath the mucosa was considered a prerequisite for healing in the original concept. The advantages of a one-stage method are that a second surgery is unnecessary, costs are lower, and patients complain less about the surgical procedures. The development of implant treatment, regardless of whether it is performed in the mandible or the maxilla, strives to shorten the period from implant placement to implant loading. For the edentulous patient--due to esthetic, economical, or psychological reasons--shortening this time and thus avoiding a long period of wearing a transitional removable prosthesis is advantageous. Use of conventional one-stage surgery makes possible and is a prerequisite for immediate loading of implants. Successful treatment outcome has been demonstrated for immediate loading of implants in the mandible, but documentation of the method in the maxilla is still sparse. Two prospective clinical studies compared (i) one- and two-stage surgery and (ii) immediate and conventional loading in patients consecutively treated in the edentulous maxilla with implant-supported fixed prostheses. The first study found that the cumulative survival rate (CSR) after one-stage surgery performed according to a conventional protocol was consistent with two-stage protocol CSRs reported in previous studies. The second study evaluated an immediate loading protocol that provided patients with interim fixed prostheses within 24 hours after implant placement. A comparison of the studies found no significant difference in CSRs. But it was found that when a conventional protocol was used, transitional removable prostheses could traumatize the bone-implant interface during healing by adverse loading on the implants, which pierced the mucosa. Moreover, splinting the implants immediately after surgery with an interim fixed prosthesis might protect them from adverse loading. In a finite element analysis comparing uncoupled and splinted implants--imitations of the clinical situations in the two studies--splinted implants drastically reduced stresses in the bone tissue surrounding the implant, which might facilitate bone healing. Two factors considered important for a successful treatment outcome, especially when loading implants immediately, are (i) jawbone quality and (ii) primary implant stability at placement. In implant literature, bone quality is generally equivalent to bone density. Results of the third clinical study in this thesis indicate that use of computed tomography with calculations of bone mineral density can be a useful tool in bone tissue evaluation before implant placement. After 1 year of loading, changes in marginal bone level, compared to baseline, did not differ between implants that were stable and implants that were not stable at placement. The results of this thesis do not strengthen earlier recommendations that immediate and early loading is a treatment alternative that can be considered only in jaws with good bone quality. In conclusion, immediate loading with interim fixed prostheses in the edentulous maxilla is a viable treatment alternative. Splinting of implants seems to be important in immediate loading, especially when bone density is low.