Simulation of the expansion process. At the end of the expansion process, a minimal snap back is realised. (MP4 9407 kb)

Planning of the implantological treatment followed usual clinical and radiological examination and, concerning the position and number of implants, the recommended categories from the German consensus conference [35]. The drilling sequence, condensing preparation (where necessary) and manual implant insertion as well as expansion are described in detail in Table 2. Participants were instructed not to wear their denture 1 week after surgery. Afterwards, the conventional dentures were relined with soft material (Visco-gel, Dentsply, Salzburg, Austria). In this study, conventional periods of submerged healing were chosen: 3 months in the mandible and 6 months in the maxilla. During the re-entry surgery, a minimum of 2-mm keratinised periimplant soft tissue mucosa was considered.

All prosthetic treatments were provided at the University School of Dental Medicine, Department of Prosthetic Dentistry. At the earliest, 2 weeks after surgical re-entry, prosthetic treatment was started. All treatment steps were performed as described in detail in Table 3. The abutment screws were fixed with a torque of 15 N cm. Wherever possible, adjacent implants were primarily splinted (crowns, bar) and extra-axial loading during dynamic occlusion was avoided. In other cases, eccentric group guidance was achieved. To reduce overloading in the periimplant bone and implant-abutment connection, the occlusal surface was designed smaller [20, 21, 30, 36]. Patients were instructed about optimal oral hygiene, and the use of a dental water jet was recommended.

All treatments were provided by two experienced maxillofacial surgeons (WR, CH) and two experienced prosthodontists (RS, JH) to minimise bias.

The first clinical follow-up was arranged at the latest 4 weeks after prosthetic treatment was completed. Further follow-ups were scheduled quarterly in the first year and later every 6 months. Patients were screened clinically and radiologically (yearly) for biological and technical complications. The authors applied the abovementioned success criteria according to Buser et al. [32]. Crestal bone changes were evaluated on digital radiograms (SIDEXIS imaging software, Sirona, Bensheim, Germany). The distance between the implant shoulder and first bone-implant contact at the mesial and distal aspect of each implant was measured (implant length as reference) by the first author (WR), and the mean values per implant were calculated [37] 1 and 2 years after loading.