Diederik Hentenaar

132 Chapter 6 Clinical assessment Clinical evaluation of the peri-implant soft tissue was performed 1 month (baseline, T1) and 5 years (T60) after placement of the final implant crown. The sulcus bleeding index (24), the gingival/mucosal index (25,26) and the probing depth using a manual periodontal probe (Williams Color Coded Probe; Hu-Friedy, Chicago, IL, USA), were clinically recorded. At four points around the implant (mesiobuccal, distobuccal, mesiolingual, distoligual), probing depths were assessed to the nearest millimeter. Statistical analysis Sample size calculation Since this is the first study evaluating the influence of cervical crown contour on marginal bone loss we could not use an effect size from the literature to perform a proper sample size calculation. Therefore we used a pilot study design in which all eligible implants (122 implants in 96 patients) from two previous studies were used (21,22). Taken into account the eligibility criteria a total of 67 implants and 64 patients were included. A significance level in statistical analysis of p < 0.05 was chosen. Inter-rater and intra-rater reliability To assess the accuracy and reliability of the new protocol for crown contour measurement, an intra-class correlation coefficient for the crown angles was calculated. Twelve radiographs (from the current dataset) were randomly selected and assessed twice by two independent examiners (DFMH&HJAM) and twice by one and the same examiner (DFMH) for the inter-rater and intra-rater reliability assessment. Pearson correlation Pearson correlation coefficients were calculated using Statistical Package for Social Sciences (version 25; SPSS Inc, Chicago, IL, USA) to determine whether the observed changes in peri-implant bone level at the mesial and distal aspects of the implant (change between one month and five years) were correlated with the measured crown angles on respectively the first 1 mm, 2 mm and 3 mm of crown height. RESULTS Patients A combined total of 96 patients (37 patients from study Telleman et al. 2014 (22) and 59 patients from study Guljé et al. 2013 (21)) with 122 bone-level implants were available for evaluation. Fifty-five implants were excluded, of which 40 implants were placed adjacent to another implant and 15 implants were not fully depicted on a peri-apical radiograph. A total of 64 unique patients with 67 bone-level implants were included