Diederik Hentenaar

35 Biomarkers in crevicular peri-implant fluid at 12000 rpm (13.800 g ) at 4°C. All sampled were washed 3 times to yield a total elution volume of 70µl. The samples were processed according to the manufacturer’s recommendations (ThermoFisher Scientific Inc., Bleiswijk, The Netherlands), in duplicate on a 96-well plate, including a standard line on all runs. The results were analysed using the MAGPIX® (with xPONENT® software) fluorescent detection system. Total biomarker concentration levels (Luminex ™ output) were determined in the elution buffer as pg/mL. A customized, highly sensitive bead-based multiplex immunoassay (Invitrogen ProcartaPlex Human 10-plex Luminex ™ panel) was used to simultaneous analyse the following 10 biomarkers: interleukin 1β (IL-1β), interleukin 6 (IL-6), tumour necrosis factor alpha (TNF-α), monocyte chemoattractant protein 1 (MCP-1/CCL2), macrophage inflammatory protein 1 alpha (MIP-1α/CCL3), interferon gamma (IFN-γ), matrix metalloproteinase 8 (MMP-8), soluble receptor activator of nuclear factor kappa-Β ligand (sRANKL), osteoprotegerin (OPG) and granulocyte-colony stimulating factor (G- CSF). According to the manufacturer’s instructions, the selected biomarkers’ lower limits of detection were: 1.62 pg/mL for IL-1β; 8.01 pg/mL for IL-6; 9.96 pg/mL for TNF-α; 3.56 pg/mL for MCP-1; 1.87 pg/mL for MIP-1α/CCL3; 14.40 pg/mL for INF-γ; 35.91 pg/mL for MMP-8; 9.11 pg/mL for OPG; 7.40 pg/mL for sRANKL; 12.72 pg/mL for G-CSF. Clinical and radiographic examination Peri-implant and full mouth clinical parameters were assessed, including bleeding on probing (BoP), suppuration on probing (SoP), probing pocket depth (PPD) and plaque index (PI). Peri-implantitis implants were assessed at baseline (T0) and 3 months (T3) after therapy. Healthy implants were only assessed at baseline (T0). All examinations were undertaken by the same researcher (DFMH). Peri-apical radiographs were taken (Planmeca Intra X-ray unit; Planmeca, Helsiniki, Finland) using a paralleling technique and an individualized X-ray holder (Meijndert et al. 2004). Peri-implant bone loss was assessed at the mesial and distal implant site using DICOM software (DicomWorks 1.5, UMCG, Groningen) by two examiners (DFMH and HJAM) showing an almost perfect observer agreement (Viera et al. 2005). Within group (peri-implantitis before and after treatment) bone loss differences were examined. Sample size calculation and statistical analysis The findings of priorly performed recoverability experiments were used to perform a sample size calculation. A group size of 18 implants per group was determined with an average effect size (Cohen’s D) of 0.9 and a power of 80%. To correct for a possible 10% drop-out (of implants), a total of 20 implants per group were required as sufficient amount to reach a reliable statistical significant difference using a significant (α) level 2