This study evaluates diagnostic markers to predict titanium implant failure. Retrospectively, implant outcome was scored in 109 subjects who had undergone titanium implant surgery, IL1A −889 C/T (rs1800587), IL1B +3954 C/T (rs1143634), IL1RN +2018 T/C (rs419598) and TNFA −308 G/A (rs1800629) genotyping, in vitro IL-1β/TNF-α release assays and lymphocyte transformation tests during treatment. TNF-α and IL-1β release on titanium stimulation were significantly higher among patients with implant loss (TNF-α: 256.89 pg/ml vs. 81.4 pg/ml; p < 0.0001; IL-1β: 159.96 pg/ml vs. 54.01 pg/ml; p < 0.0001). The minor alleles of the studied polymorphisms showed increased prevalence in the implant failure group (IL1A: 61% vs. 42.6% in controls, IL1B: 53.7% vs. 39.7% in controls, TNFA: 46.3% vs. 30.9% in controls, IL1RN: 58.5% vs. 52.9% in controls). Increasing numbers of risk genotypes of the studied polymorphisms were associated with an increasing risk of implant loss, suggesting an additive effect. Multiple logistic regression analysis showed positive IL-1β/TNF-α release assay scores ( p < 0.0001, OR = 12.01) and number of risk genotypes ( p < 0.046, OR = 1.57–6.01) being significantly and independently associated with titanium implant failure. IL-1/IL1RN/TNFA genotyping and cytokine release assay scores provide prognostic markers for titanium implant outcome and may present new tools for individual risk assessment.
Dental implants have become important therapeutic tools during the last decades. While success rates are 85–95% for all implant systems implant failure occurs despite adequate surgical conditions. The most common cause of failure is wear, debris-mediated implant loosening, a process referred to as osteolysis. Additional risk factors are smoking, one- or two-step surgery, medical pre-conditions and poor bone quality.
Studies have demonstrated that implant material is a major determinant of treatment outcome. Particles shed from titanium implants have been shown to stimulate macrophages more strongly than particles from other materials used in implant restoration. Macrophages release pro-inflammatory cytokines, such as interleukin 1 (IL-1) and tumour necrosis factor alpha (TNF-α) mediating the inflammatory and osteolytic process of peri-implantitis.
The fact that titanium particles induce inflammation and osseodisintegration only in a minority of implant recipients, points to a significant role of host factors, in particular the immune response to titanium particles. IL-1, TNF-α and the anti-inflammatory IL-1 receptor antagonist (IL1RN) play significant roles in inflammatory processes, so functional polymorphisms in these genes may constitute genetic risk factors for implant failure. Genetic variations include IL1A −889 C/T and IL1B +3953 C/T that are associated with increased levels of IL-1. IL1RN +2018 T/C correlates with decreased levels of IL1RN and TNFA −308 G/A has been associated with a sevenfold increase in TNF-α expression. Several studies have linked IL-1 and IL-1RN polymorphisms to peri-implantitis , implant failure and peri-implant bone loss. TNF-α has been implicated in inflammation and bone resorption in an experimental model of periodontitis. Studies have reported an overrepresentation of TNFA gene variants in patients with peri-implantitis.
In order to establish diagnostic tools for individual risk assessment in dental medicine, the authors set out to define a set of markers suitable for predicting the risk of titanium implant failure. They carried out a retrospective study that investigated the influence of genetic variation in four cytokine genes (IL1A −889 C/T, IL1B +3954 C/T, IL1RN +2018 T/C and TNFA −308 G/A) as well as the influence of the individual titanium-induced cytokine release in a functional in vitro assay.
Materials and methods
This retrospective study included 109 unrelated Northern Caucasian individuals aged 14–79 years (average 51.6 years) who had received a two-component titanium implant system (CAMLOG Biotechnologies AG, FRIATEC ® AG, Straumann GmBH, FRIADENT GmBH, Nobel Biocare™, Astra Tech GmBH, BTI Deutschland GmBH, SIC Invent AG, Zimmer Dental GmBH), IL1A/IL1B/IL1RN/TNFA genotyping and IL-1β/TNF-α release assays for routine medical treatment at the same private dental clinic in the course of dental restoration between 1981 and 2008. The same oral surgeon placed all the implants. Any periodontal disease was treated adequately before insertion of the implants. Information on general medical conditions, bruxism, number of implants, oral hygiene and smoking was recorded. According to the Helsinki Declaration all patients gave written informed consent to share their data for scientific evaluation. Ethical approval was not required.
Titanium stimulation assays and lymphocyte transformation tests were performed using heparinized venous blood. For genetic analysis, genomic DNA was extracted from whole blood or buccal epithelial cells using the QIAamp DNA mini kit (Qiagen, Hilden, Germany).
IL-1β/TNF-α release assay
TNF-α- and IL-1β-production were measured after incubation of 1:2 v/v RPMI (Roswell Park Memorial Institute) diluted heparinized whole blood with titanium dioxide (1 × 10 5 particles/ml, Sigma Aldrich, Taufkirchen, Germany) in pyrogen-free 2 ml tubes (Eppendorf, Germany). Maximum titanium particle diameter was 2 μm. TNF-α and IL-1β levels were determined using the automated Immulite ® CLIA system (Siemens, Germany). The lower detection limit was 5 pg/ml for both cytokines. The upper detection limits were 1000 pg/ml and 3000 pg/ml for IL-1β and TNF-α, respectively.
Lymphocyte transformation test
Peripheral blood mononuclear cells were isolated by Ficoll-Plaque (Pharmacia, Uppsala, Sweden) density gradient centrifugation from heparinized venous blood and lymphocyte transformation test was performed as previously described.
IL1A (rs1800587), IL1B (rs1143634) and IL1RN (rs419598) were genotyped by hybridization to specific probes (GenoType IL-1; VER 1.0, HAIN Lifescience, Nehren, Germany). TNFA (rs1800629) was genotyped by polymerase chain reaction and melting curve analysis using a Light Cycler 1.5 (Roche Diagnostics, Mannheim, Germany) as published previously.
Genotype frequencies and their combinations were analyzed by standard χ 2 test. Differences of IL-1β and TNF-α-production were assessed by Mann–Whitney U -test. Multivariate logistic regression analysis was used to determine which factors influence the outcome of implant survival. The number of risk genotypes, the results of the TNF-α/IL-1β release assay, age, gender and smoking status were entered into the model, the significant determinants were determined and odds ratios (ORs) were calculated. For multivariant analysis, reference values for the titanium stimulation assay were defined as >30 pg/ml for TNF-α and >25 pg/ml for IL-1β. TNF-α/IL-1β release assays were scored as positive if either TNF-α, IL-1β or both exceeded their reference values. A p -value <0.05 was considered significant. Statistical analysis was performed using IBM SSP statistics version 19.
Demographic data for the study groups are summarized in Table 1 . The 109 participants in this retrospective study, comprised 41 patients with implant loss and 68 patients with functional implants as controls. In the implant loss group, 14 patients (34.1%) showed early implant loss before loading (average implant survival 4.2 months). The remaining 29 patients showed clinical signs of peri-implantitis after implant loading, resulting in implant loss (average implant survival 75.6 months). The 68 patients who served as controls maintained functional implants for at least 5.2 years (range 5.2–29.6 years) ( Table 2 ). Since smoking is considered a risk factor for peri-implantitis, smokers were distributed equally between patients and controls (14.7% in control group; 14.6% in cases group). When medical diseases (hypertension, diabetes, sensitization to nickel, allergic rhinitis, hypothyroidism, hyperthyroidism, neurodermatitis) or clinical findings (daily oral hygiene, bruxism, number of implants) were stratified, there were no significant differences between the two groups ( Table 2 ).
|Cases ( n = 41)||Controls ( n = 68)||p -Value *|
|Average age (years)||51.1||51.8|
|Implant failure ( n = 41)||Control ( n = 68)||p -Value *|
|General medical conditions|
|Sensitization to nickel||6||14.6||5||7.3||0.325|
|Daily oral hygiene|
|Number of implants|
|More than one||12||29.3||32||47.1|
|Early implant loss (unload)||14||34.1|
|Late implant loss (load)||27||65.9|
|Controls (no loss)||68|