Page 72 - Peri-implant health: the effect of implant design and surgical procedure on bone and soft tissue stability

CHAPTER 4
The list of generated articles was obtained through elimination based on the title and detailed information given in the abstract. Further evaluation and refining of the selected papers was performed by reading the papers and registering the described results in the data set. In case of disagreement over the inclusion/ exclusion, both evaluators (RD & VC) discussed and reached a consensus or, in case of further doubt, a third evaluator was consulted (HDB).
To be included in the study, the papers had to be published in English, report on bone loss compared to a baseline measurement and include at least 10 patients after a minimal mean follow-up time of 5 years. Only studies discussing straightforward implant treatment in systemically healthy patients, as the test strategies were included. Hence, studies describing implant treatments in tumor- resected areas, studies involving extensive bone grafts or zygomatic implants were excluded. Exceptional, experimental or uncommon implant designs, as well as implants with unknown surface topography, were excluded. Studies were additionally rejected when statistical evaluation was hampered because of incomplete data reporting, such as lacking failure rate, implant or patient numbers, bone loss and standard deviation calculated on implant level. However, studies lacking standard deviation on bone loss were included for calculating implant survival.
Table 2 gives an overview of exclusion criteria and the references of the excluded papers. The review did not exclude studies with smokers, patients with periodontal history, controlled diabetes, or implants in sinus lifted bone. These procedures are today considered part of daily good clinical practice. To avoid a biased selection and to ensure that the papers reflected the daily clinical situation, no distinction was made based on study design (prospective or retrospective) or surgical or prosthetic treatment protocol.
Statistical Analysis
For each study the mean bone loss was used together with the number of implants to calculate the weight of the study in the overall statistical analysis of bone loss and to estimate the proportion of implants with bone loss above 1, 2, and 3 mm. Analysis was performed additionally per implant surface roughness group and qualified as minimally rough, moderately rough, rough, or mixed/ unknown. The latter included studies with unspecified implant surface roughness
70