Christoph A. Ramseier* Sigrun Eick* €nnimann Claudia Bro Daniel Buser €gger Urs Bra Giovanni E. Salvi
Authors’ affiliations: Christoph A. Ramseier, Sigrun Eick, Claudia Br€ onnimann, Giovanni E. Salvi, Department of Periodontology, School of Dental Medicine, University of Bern, Bern, Switzerland Daniel Buser, Department of Oral Surgery and Stomatology, School of Dental Medicine, University of Bern, Bern, Switzerland Urs Br€ agger, Department of Reconstructive Dentistry and Gerodontology, School of Dental Medicine, University of Bern, Bern, Switzerland
Host-derived biomarkers at teeth and implants in partially edentulous patients. A 10-year retrospective study
Key words: biomarkers, dental implants, gingival crevicular fluid, host response, peri-implant
mucositis, peri-implant sulcus fluid, peri-implantitis Abstract Objectives: To assess a selection of host-derived biomarkers in peri-implant sulcus fluid (PISF) and gingival crevicular fluid (GCF) from adjacent teeth 10 years following implant placement. Material and methods: Peri-implant sulcus fluid and GCF samples obtained from the deepest sites of 504 implants and 493 adjacent teeth were analysed for levels of interleukin (IL)-1b, matrix metalloproteinase (MMP)-3, MMP-8, MMP-1, and MMP-1 bound to tissue inhibitor of MMP (TIMP)-
Corresponding author: PD Dr Sigrun Eick Laboratory of Oral Microbiology, Department of Periodontology, School of Dental Medicine, University of Bern, Freiburgstrasse 7, CH-3010 Bern, Switzerland Tel.: +41 31 632 25 42 Fax: +41 31 632 86 08 e-mail: [email protected]
1 (MMP-1/TIMP-1) by enzyme-linked immunosorbent assay (ELISA) technique. Results: Overall, MMP-8 was detected in 90% of the sites. In more than 50% of the sites, IL-1b was identified while in 30% of the sites MMP-1, MMP-1/TIMP-1 and MMP-3 were found over the detection level. Increased biomarkers levels from PISF and GCF were positively correlated (r = 0.375–0.702; P < 0.001). However, no qualitative and quantitative differences were found between PISF and GCF. The levels of MMP-1 were negatively correlated with those of MMP-1/TIMP1 at implants (r =
0.644; P < 0.001). Median MMP-1 levels at implants were high (5.17 pg/site) in
subjects with severe chronic periodontitis and low in patients with mild-to-moderate chronic periodontitis (0 pg/site; P = 0.026) or gingivitis (0 pg/site; P = 0.034). Levels of IL-1b were found to be different in GCF according to the periodontal conditions (P = 0.001) with the highest level found in mild-to-moderate periodontitis (6.2 pg/site). Clinical attachment levels at implants demonstrated an inverse correlation with MMP-1/TIMP-1 (r =
0.147; P = 0.001).
Conclusions: Increased levels of MMP-8 and IL-1b in PISF or GCF may be associated with inflammation around teeth and implants while lower levels of MMP-1/TIMP-1 may be an indicator of disease progression around implants.
*These authors contributed equally to the manuscript. Date: Accepted 14 January 2015 To cite this article: Ramseier CA, Eick S, Br€ onnimann C, Buser D, Br€ agger U, Salvi GE. Host-derived biomarkers at teeth and implants in partially edentulous patients. A 10-year retrospective study. Clin. Oral Impl. Res. 00, 2015, 1–7 doi: 10.1111/clr.12566
Although oral implant rehabilitation has consistently become more successful and predictable over the past 20 years, an increasing number of implant failures caused by periimplant diseases still cause issues in everyday professional dental care. Peri-implant mucositis is defined as an inflammatory lesion of the oral mucosa surrounding a dental implant, while peri-implantitis is accompanied by loss of the implant supporting bone (Lindhe & Meyle 2008). Prevalence of inflammatory peri-implant diseases is high and increasing. In Brazil, 212 patients with 578 implants were examined; 64.6% of the subjects and 62.6% of the implants presented with peri-implant mucositis while in 8.9% of the patients and 7.44% of the implants, periimplantitis was diagnosed (Ferreira et al. 2006). In Sweden, Roos-Jansaker et al. (2006)
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
analysed 218 subjects with 994 implants; 48% of the implants and 76.6% of the patients demonstrated probing depths ≥4 mm along with bleeding on probing (Roos-Jansaker et al. 2006). Smoking, periodontitis, diabetes and poor oral hygiene have been associated with the development of peri-implantitis indicating the presence of microbes and diminished host response as risk factors (Ferreira et al. 2006). Consequently, both prevention and treatment of peri-implant diseases need progressively more attention. Peri-implant sulcus fluid has been analysed for different molecules. The spectrum covered a number of biomarkers associated with bone turnover (e.g. osteocalcin) (Murata et al. 2002), pro-inflammatory cytokines (e.g. tumour necrosis factor (TNF)a) (de Mendonca et al. 2009), and proteinases (e.g. neutrophil
1
Ramseier et al Host-derived biomarkers at teeth and implants
elastase) (Ataoglu et al. 2002). Increased levels of inflammatory mediators in the periimplant sulcus fluid associated with signs of peri-implant inflammation indicate a potential value for monitoring host response in implant therapy (Salvi & Lang 2004). Among others biomarkers, MMPs and IL1b were used for monitoring peri-implant conditions (Nomura et al. 2000; Cosgarea et al. 2012). IL-1b belongs to the interleukin1 cytokines, which are induced by proinflammatory mediators such as lipopolysaccharide, TNF and IL-1b itself. As reviewed by Barksby et al. (2007), it accounts for neutrophil infiltration and tissue turnover via inducing matrix metalloproteinase (MMP) cytokines (Barksby et al. 2007). Moreover, as reviewed by Sorsa et al. (2004), MMPs as a family of enzymes, which degrade extracellular matrix and basement membrane components, are mostly produced in a latent form but are activated through tissue proteinases, bacterial proteinases and oxidative stress (Sorsa et al. 2004). The activities of MMPs are controlled by tissue inhibitors of matrix metalloproteinases (TIMPs) (Sorsa et al. 2004). MMP-8 (neutrophil collagenase) and MMP-1 (tissue collagenase) target native collagen while MMP-3 (stromelysin) uses aggrecan, laminin, fibronectin as substrates (Uitto et al. 2003). A long-term retrospective evaluation was made of both the survival and the complication rates of implants with a sandblasted and acid-etched surface (SLA) (Straumannâ Dental Implant System; Institut Straumann AG, Basel, Switzerland) 10 years after placement in partially edentulous patients. Clinical results have recently been published (Buser et al. 2012). One part of the present evaluation was focusing on the levels of selected host-derived biomarkers at implants. Therefore, its main aims were (1) to compare levels of biomarkers at implants with those at adjacent teeth, (2) to find associations of these biomarkers with periodontal disease, (3) to detect differences in the release of these biomarkers in relation to the peri-implant inflammation and (4) to determine the influence of the smoking status on the concentration levels of IL-1b and a selection of MMPs.
Material and methods Study design
The study design was recently published in detail (Buser et al. 2012). Briefly, a sample of patients from the three Departments of Oral Surgery, Periodontology, and Fixed Prostho-
2 |
Clin. Oral Impl. Res. 0, 2015 / 1–7
dontics at the University of Bern, Switzerland, having received at least one dental implant were recruited for a comprehensive evaluation 10 years after implant installation. Overall, the patients had been treated according to a comprehensive care treatment strategy prior to the installation of the implant(s) and incorporation of the reconstruction(s). Thereafter, supportive periodontal therapy and maintenance care at intervals of 3, 4 or 6 months were offered to all the patients. All the implants installed were solid screws with a diameter of 3.3, 4.1 or 4.8 mm, respectively, a length between 6–14 mm, and a sandblasted and acid-etched surface (SLA) (Straumannâ Dental Implant System; Institut Straumann AG). They were placed according to the manufacturer’s instructions (Buser et al. 2000). The reconstructions were either cemented or screw-retained and consisted of single-unit crowns or short-span fixed partial dentures with or without cantilever extension(s). At the re-examination, both medical history and tobacco use history were assessed with all the patients. Reconstructions were examined for any technical and mechanical complications (Wittneben et al. 2014). Periodontitis was classified according to the American Academy of Periodontology criteria (Page & Eke 2007). The implant and the mesial adjacent tooth were examined in more detail for the presence or absence of plaque biofilm recorded as plaque-index (PlI) (Silness & L€ oe 1964), gingival or peri-implant mucosal tissue inflammation recorded as gingivalindex (GI) (L€ oe & Silness 1963) as modified for oral implants (mGI) (Mombelli et al. 1987), bleeding on probing (BoP) (Lang et al. 1986), probing depth (PD), and clinical attachment level (CAL) were assessed (Buser et al. 1990, 2012). The study was conducted according to the ethical principles, including the World Medical Association Declaration of Helsinki (version 2008). Sampling of gingival crevicular fluid and periimplant sulcus fluid
Samples were obtained from the deepest site of the implant and the adjacent tooth, respectively. The area was isolated with a cotton roll and a saliva ejector and air-dried. Gingival crevicular fluid (GCF) and peri-implant sulcus fluid (PISF) were collected by means of sterile paper strips (Periopaper; Oraflow Inc., Smithtown, NY, USA). Paper strips were placed at the entrance of the crevice and left in place for 30 s. Following the sampling, the paper strips were transferred to tubes and
placed in ice. Samples were eluted at 4°C overnight into 750 ll phosphate-buffered saline containing proteinase inhibitors (SigmaAldrich, St. Louis, MO, USA). Subsequently, the eluates were centrifuged at 3000 g for 10 min. The supernatants of the eluates were stored at 80°C until further analysis. Assessment of host-derived biomarkers
From the eluates, the levels of IL-1b, MMP-3, MMP-8, MMP-1 and MMP-1/TIMP-1 were determined using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems Europe Ltd, Abingdon, UK) according to the manufacturer’s instruction. The kits’ detection levels were 3.75 pg/site for IL-1b, MMP-1, MMP-1/TIMP-1, 7.5 pg/site for MMP-3, and 37.5 pg/site for MMP-8, respectively. The procedure by discarding the sediments included only the molecules which were neither located inside nor attached to host cells, for example neutrophils. Data analysis
Nonparametric tests were used for the statistical analysis. By means of Wilcoxon-tests comparisons of implants with their adjacent teeth were made. Kruskal–Wallis tests and Mann–Whitney tests were used to analyse differences between independent groups. Further, Spearman’s correlation was applied to find associations between variables. Values of P < 0.05 were considered to be significant.
Results Overall, 504 implants and 493 adjacent teeth were evaluated resulting in 19 implants with missing adjacent teeth. Host-derived biomarkers at implants and the adjacent teeth
From all the samples in this study, about 90% were tested positive for MMP-8, more than 50% for IL-1b and each about 30% over the detection level for MMP-1, MMP-1/ TIMP-1 and MMP-3 (Table 1). No statistically significant differences were identified between implants and adjacent teeth either qualitatively or quantitatively. There were strong correlations of the implants’ concentration with those at the adjacent teeth (each P < 0.001). The correlation factors r ranged between 0.375 (MMP-8) and 0.702 (IL-1b). The levels of MMP-1 correlated negatively with those of MMP-1/TIMP-1 at implants (r = 0.150; P = 0.001) and at teeth (r = 0.140; P = 0.002). If the cases with negative values for both MMP-1 and MMP-1/
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Ramseier et al Host-derived biomarkers at teeth and implants
Table 1. Total interleukin-1b, MMP-1, MMP-1/TIMP1, MMP-8 and MMP-3 (samples over the detection level, minimum, maximum, quartiles) at implants and teeth Correlation implant– tooth Biomarker MMP-3 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-8 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-1 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-1/TIMP-1 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) IL-1b Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site)
Implants (total n = 504)
Teeth (total n = 493)
Difference implant–tooth P
r
P
Association of host-derived biomarkers with clinical data
150 (29.8) 0.00 0.00 0.00 8.90 186.70
165 (33.5) 0.00 0.00 0.00 9.75 294.23
0.130
0.537
Authors’ affiliations: Christoph A. Ramseier, Sigrun Eick, Claudia Br€ onnimann, Giovanni E. Salvi, Department of Periodontology, School of Dental Medicine, University of Bern, Bern, Switzerland Daniel Buser, Department of Oral Surgery and Stomatology, School of Dental Medicine, University of Bern, Bern, Switzerland Urs Br€ agger, Department of Reconstructive Dentistry and Gerodontology, School of Dental Medicine, University of Bern, Bern, Switzerland
Host-derived biomarkers at teeth and implants in partially edentulous patients. A 10-year retrospective study
Key words: biomarkers, dental implants, gingival crevicular fluid, host response, peri-implant
mucositis, peri-implant sulcus fluid, peri-implantitis Abstract Objectives: To assess a selection of host-derived biomarkers in peri-implant sulcus fluid (PISF) and gingival crevicular fluid (GCF) from adjacent teeth 10 years following implant placement. Material and methods: Peri-implant sulcus fluid and GCF samples obtained from the deepest sites of 504 implants and 493 adjacent teeth were analysed for levels of interleukin (IL)-1b, matrix metalloproteinase (MMP)-3, MMP-8, MMP-1, and MMP-1 bound to tissue inhibitor of MMP (TIMP)-
Corresponding author: PD Dr Sigrun Eick Laboratory of Oral Microbiology, Department of Periodontology, School of Dental Medicine, University of Bern, Freiburgstrasse 7, CH-3010 Bern, Switzerland Tel.: +41 31 632 25 42 Fax: +41 31 632 86 08 e-mail: [email protected]
1 (MMP-1/TIMP-1) by enzyme-linked immunosorbent assay (ELISA) technique. Results: Overall, MMP-8 was detected in 90% of the sites. In more than 50% of the sites, IL-1b was identified while in 30% of the sites MMP-1, MMP-1/TIMP-1 and MMP-3 were found over the detection level. Increased biomarkers levels from PISF and GCF were positively correlated (r = 0.375–0.702; P < 0.001). However, no qualitative and quantitative differences were found between PISF and GCF. The levels of MMP-1 were negatively correlated with those of MMP-1/TIMP1 at implants (r =
0.644; P < 0.001). Median MMP-1 levels at implants were high (5.17 pg/site) in
subjects with severe chronic periodontitis and low in patients with mild-to-moderate chronic periodontitis (0 pg/site; P = 0.026) or gingivitis (0 pg/site; P = 0.034). Levels of IL-1b were found to be different in GCF according to the periodontal conditions (P = 0.001) with the highest level found in mild-to-moderate periodontitis (6.2 pg/site). Clinical attachment levels at implants demonstrated an inverse correlation with MMP-1/TIMP-1 (r =
0.147; P = 0.001).
Conclusions: Increased levels of MMP-8 and IL-1b in PISF or GCF may be associated with inflammation around teeth and implants while lower levels of MMP-1/TIMP-1 may be an indicator of disease progression around implants.
*These authors contributed equally to the manuscript. Date: Accepted 14 January 2015 To cite this article: Ramseier CA, Eick S, Br€ onnimann C, Buser D, Br€ agger U, Salvi GE. Host-derived biomarkers at teeth and implants in partially edentulous patients. A 10-year retrospective study. Clin. Oral Impl. Res. 00, 2015, 1–7 doi: 10.1111/clr.12566
Although oral implant rehabilitation has consistently become more successful and predictable over the past 20 years, an increasing number of implant failures caused by periimplant diseases still cause issues in everyday professional dental care. Peri-implant mucositis is defined as an inflammatory lesion of the oral mucosa surrounding a dental implant, while peri-implantitis is accompanied by loss of the implant supporting bone (Lindhe & Meyle 2008). Prevalence of inflammatory peri-implant diseases is high and increasing. In Brazil, 212 patients with 578 implants were examined; 64.6% of the subjects and 62.6% of the implants presented with peri-implant mucositis while in 8.9% of the patients and 7.44% of the implants, periimplantitis was diagnosed (Ferreira et al. 2006). In Sweden, Roos-Jansaker et al. (2006)
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
analysed 218 subjects with 994 implants; 48% of the implants and 76.6% of the patients demonstrated probing depths ≥4 mm along with bleeding on probing (Roos-Jansaker et al. 2006). Smoking, periodontitis, diabetes and poor oral hygiene have been associated with the development of peri-implantitis indicating the presence of microbes and diminished host response as risk factors (Ferreira et al. 2006). Consequently, both prevention and treatment of peri-implant diseases need progressively more attention. Peri-implant sulcus fluid has been analysed for different molecules. The spectrum covered a number of biomarkers associated with bone turnover (e.g. osteocalcin) (Murata et al. 2002), pro-inflammatory cytokines (e.g. tumour necrosis factor (TNF)a) (de Mendonca et al. 2009), and proteinases (e.g. neutrophil
1
Ramseier et al Host-derived biomarkers at teeth and implants
elastase) (Ataoglu et al. 2002). Increased levels of inflammatory mediators in the periimplant sulcus fluid associated with signs of peri-implant inflammation indicate a potential value for monitoring host response in implant therapy (Salvi & Lang 2004). Among others biomarkers, MMPs and IL1b were used for monitoring peri-implant conditions (Nomura et al. 2000; Cosgarea et al. 2012). IL-1b belongs to the interleukin1 cytokines, which are induced by proinflammatory mediators such as lipopolysaccharide, TNF and IL-1b itself. As reviewed by Barksby et al. (2007), it accounts for neutrophil infiltration and tissue turnover via inducing matrix metalloproteinase (MMP) cytokines (Barksby et al. 2007). Moreover, as reviewed by Sorsa et al. (2004), MMPs as a family of enzymes, which degrade extracellular matrix and basement membrane components, are mostly produced in a latent form but are activated through tissue proteinases, bacterial proteinases and oxidative stress (Sorsa et al. 2004). The activities of MMPs are controlled by tissue inhibitors of matrix metalloproteinases (TIMPs) (Sorsa et al. 2004). MMP-8 (neutrophil collagenase) and MMP-1 (tissue collagenase) target native collagen while MMP-3 (stromelysin) uses aggrecan, laminin, fibronectin as substrates (Uitto et al. 2003). A long-term retrospective evaluation was made of both the survival and the complication rates of implants with a sandblasted and acid-etched surface (SLA) (Straumannâ Dental Implant System; Institut Straumann AG, Basel, Switzerland) 10 years after placement in partially edentulous patients. Clinical results have recently been published (Buser et al. 2012). One part of the present evaluation was focusing on the levels of selected host-derived biomarkers at implants. Therefore, its main aims were (1) to compare levels of biomarkers at implants with those at adjacent teeth, (2) to find associations of these biomarkers with periodontal disease, (3) to detect differences in the release of these biomarkers in relation to the peri-implant inflammation and (4) to determine the influence of the smoking status on the concentration levels of IL-1b and a selection of MMPs.
Material and methods Study design
The study design was recently published in detail (Buser et al. 2012). Briefly, a sample of patients from the three Departments of Oral Surgery, Periodontology, and Fixed Prostho-
2 |
Clin. Oral Impl. Res. 0, 2015 / 1–7
dontics at the University of Bern, Switzerland, having received at least one dental implant were recruited for a comprehensive evaluation 10 years after implant installation. Overall, the patients had been treated according to a comprehensive care treatment strategy prior to the installation of the implant(s) and incorporation of the reconstruction(s). Thereafter, supportive periodontal therapy and maintenance care at intervals of 3, 4 or 6 months were offered to all the patients. All the implants installed were solid screws with a diameter of 3.3, 4.1 or 4.8 mm, respectively, a length between 6–14 mm, and a sandblasted and acid-etched surface (SLA) (Straumannâ Dental Implant System; Institut Straumann AG). They were placed according to the manufacturer’s instructions (Buser et al. 2000). The reconstructions were either cemented or screw-retained and consisted of single-unit crowns or short-span fixed partial dentures with or without cantilever extension(s). At the re-examination, both medical history and tobacco use history were assessed with all the patients. Reconstructions were examined for any technical and mechanical complications (Wittneben et al. 2014). Periodontitis was classified according to the American Academy of Periodontology criteria (Page & Eke 2007). The implant and the mesial adjacent tooth were examined in more detail for the presence or absence of plaque biofilm recorded as plaque-index (PlI) (Silness & L€ oe 1964), gingival or peri-implant mucosal tissue inflammation recorded as gingivalindex (GI) (L€ oe & Silness 1963) as modified for oral implants (mGI) (Mombelli et al. 1987), bleeding on probing (BoP) (Lang et al. 1986), probing depth (PD), and clinical attachment level (CAL) were assessed (Buser et al. 1990, 2012). The study was conducted according to the ethical principles, including the World Medical Association Declaration of Helsinki (version 2008). Sampling of gingival crevicular fluid and periimplant sulcus fluid
Samples were obtained from the deepest site of the implant and the adjacent tooth, respectively. The area was isolated with a cotton roll and a saliva ejector and air-dried. Gingival crevicular fluid (GCF) and peri-implant sulcus fluid (PISF) were collected by means of sterile paper strips (Periopaper; Oraflow Inc., Smithtown, NY, USA). Paper strips were placed at the entrance of the crevice and left in place for 30 s. Following the sampling, the paper strips were transferred to tubes and
placed in ice. Samples were eluted at 4°C overnight into 750 ll phosphate-buffered saline containing proteinase inhibitors (SigmaAldrich, St. Louis, MO, USA). Subsequently, the eluates were centrifuged at 3000 g for 10 min. The supernatants of the eluates were stored at 80°C until further analysis. Assessment of host-derived biomarkers
From the eluates, the levels of IL-1b, MMP-3, MMP-8, MMP-1 and MMP-1/TIMP-1 were determined using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems Europe Ltd, Abingdon, UK) according to the manufacturer’s instruction. The kits’ detection levels were 3.75 pg/site for IL-1b, MMP-1, MMP-1/TIMP-1, 7.5 pg/site for MMP-3, and 37.5 pg/site for MMP-8, respectively. The procedure by discarding the sediments included only the molecules which were neither located inside nor attached to host cells, for example neutrophils. Data analysis
Nonparametric tests were used for the statistical analysis. By means of Wilcoxon-tests comparisons of implants with their adjacent teeth were made. Kruskal–Wallis tests and Mann–Whitney tests were used to analyse differences between independent groups. Further, Spearman’s correlation was applied to find associations between variables. Values of P < 0.05 were considered to be significant.
Results Overall, 504 implants and 493 adjacent teeth were evaluated resulting in 19 implants with missing adjacent teeth. Host-derived biomarkers at implants and the adjacent teeth
From all the samples in this study, about 90% were tested positive for MMP-8, more than 50% for IL-1b and each about 30% over the detection level for MMP-1, MMP-1/ TIMP-1 and MMP-3 (Table 1). No statistically significant differences were identified between implants and adjacent teeth either qualitatively or quantitatively. There were strong correlations of the implants’ concentration with those at the adjacent teeth (each P < 0.001). The correlation factors r ranged between 0.375 (MMP-8) and 0.702 (IL-1b). The levels of MMP-1 correlated negatively with those of MMP-1/TIMP-1 at implants (r = 0.150; P = 0.001) and at teeth (r = 0.140; P = 0.002). If the cases with negative values for both MMP-1 and MMP-1/
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Ramseier et al Host-derived biomarkers at teeth and implants
Table 1. Total interleukin-1b, MMP-1, MMP-1/TIMP1, MMP-8 and MMP-3 (samples over the detection level, minimum, maximum, quartiles) at implants and teeth Correlation implant– tooth Biomarker MMP-3 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-8 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-1 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) MMP-1/TIMP-1 Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site) IL-1b Positive n (%) Minimum (pg/site) Percentile 25 (pg/site) Median (pg/site) Percentile 75 (pg/site) Maximum (pg/site)
Implants (total n = 504)
Teeth (total n = 493)
Difference implant–tooth P
r
P
Association of host-derived biomarkers with clinical data
150 (29.8) 0.00 0.00 0.00 8.90 186.70
165 (33.5) 0.00 0.00 0.00 9.75 294.23
0.130
0.537
