Failure of Dental Implants and its Association with IL-10 A/G rs1800896 and TNF-α G308A rs1800629308 Polymorphisms, Systematic Review, and Meta-Analysis of Clinical Trials

The general objective of this work was to evaluate the IL-10 A/G rs1800896 and TNF-α G308A rs1800629308 polymorphisms with the loss of one or more dental implants through a systematic review and meta-analysis of clinical trials. The PRISMA guidelines established for systematic reviews and meta-analyses were followed. It was carried out on randomized controlled clinical trials, epidemiological studies obtained from the MEDLINE, EMBASE, LILACS, IBECS and Cochrane databases. The initial search returned 593 results. 28 articles were selected, and we excluded 12 studies that did not describe the genetic polymorphism we were looking for related to dental implant survival. Therefore, we included 4 articles in this review. We analyzed all these different categories of genes with their respective polymorphisms (IL-1A [-889], IL-1B [+3954], IL-1B [+3953], IL-1B [-511], IL-4 [+33 ], IL-4 [-590], IL-6 [-174], IL-10 [-1082], IL-10 [-819], IL-10 [-592], IL-1 receptor antagonist [IL-1ra]), tumor necrosis factor (TNF-α [-308]). There are very few comprehensive studies that have the same methodologies and similar data analyzes to be able to compare them that include the types of SNPs altered in the gene related to reduced implant survival to perform the meta-analysis. This systematic review confirms results that are still controversial due to the limited sample size, the definition in the terminology dental implant failure not yet defined equally in all studies, and the non-homogeneous inclusion and exclusion criteria, which include race, age, smoking, occlusion trauma, and periodontal disease.

Introduction

The success of more than 90% in oral rehabilitation with oral implants makes this technique an alternative therapeutic alternative to other prosthetic therapies, such as removable or fixed ones supported on dental elements [1].

Although these therapies are predictable, they are not free of complications; we frequently find peri-implant mucositis (19%–65%), peri-implantitis (1%–47%) [2], aesthetic imperfection and absolute loss of osseointegration before functional loading of the implants. The main characteristic reported in studies with follow-up over time is the implant survival rate, underestimating data on complications of a biological or technical nature in the procedures [3]. The studies are predominantly based on evaluations carried out in small groups of arbitrarily selected patients, that is, samples of the researcher’s convenience [2]. A significant correlation has been found between implant failures and habits such as smoking, patients with diabetes, cardiac pathologies, poor oral hygiene, pre-existing infections, poor bone quality, and bruxism or occlusal trauma. A higher failure rate was also observed for narrow, short, tapered, internally connected, grade IV titanium implants [4]–[6]. The use of temporary removable prostheses was significantly associated with an increased risk of implant failure. On the other hand, antibiotic prevention (before and after surgery) and some specific brands of implants were related to a lower risk of failure [7]. Also, osteoporosis, Crohn’s disease, smoking habits, certain characteristics of the implants (length, diameter, and location), and proximity to other natural dental elements were significantly associated with early implant loss (p < 0.05) [8]. Other authors report that tobacco consumption, male sex, not having any dental element in the oral cavity, implant diameter, and surgeries that increase bone volumes are associated with losses. early stages of implants [9]. The evidence supports that at least twelve polymorphisms could be related in some way to implantological complications, and the results would vary depending on the community studied. Among the research that has been published, we frequently find initial studies with small samples that use different methods to homogenize the data of the study groups.

The proteins, cytokines IL1A, IL1B, and IL1RN, play a role in the immuno-inflammatory response, are located on chromosome 2, and make up the IL-1 gene group. Now, numerous investigations related to peri-implantitis and early loss or failure of implants have been reported. However, studies have been somewhat inconsistent [10]. Tumor necrosis factor α (TNFα), a cytokine that is also pro-inflammatory, is strategically associated with the first stages of the inflammatory response as well as with insulin resistance [11].

The increase in TNFα levels in the crevicular fluid and saliva of patients diagnosed with peri-implantitis with active lesions is correlated with the progression of peri-implant disease [12]. TNFα, among its direct or indirect repercussions, has the ability to encourage bone resorption by promoting cell differentiation of the monocytic line and macrophages to osteoclasts, promoting osteoclastic maturation and enhancing its activity resorption [12], [13].

MMP Metalloproteinases show diverse activities and functions, and we understand that they participate in tumor growth, metastasis phenomena, and even cell death. The other cells that generate MMPs like osteoclasts are fibroblasts, gingival sulcus epithelial cells, endothelial cells, plasma cells, monocytes/macrophages, and neutrophils. There would be an increased risk of early implant failure in some polymorphisms of this genotype [14]. Polymorphisms in the CD-14 (159-) polymorphic region of the gene encode the CD-14 membrane receptor (Group of Differentiation-14). CD-14 has the property that monocytes, macrophages, and polymorphonuclear cells distinguish lipopolysaccharides from the walls of Gram-negative bacteria. The results of this study demonstrated that the presence of the CC genotype was associated with a fivefold risk of peri-implantitis, while the other genotype, the CT, would have a certain protective effect [12].

Campos et al., they studied the early failure of dental implants, TNF-alpha gene polymorphism (G-308A), and their results indicated that TNF-alpha (the G-308A gene polymorphism) would not be associated with early implant failure, suggesting that by itself it does not represent a genetic risk factor for implant loss in the Brazilian population studied [15]. Mo et al. also learned the link between the G-308A tumor necrosis factor-alpha polymorphism and the threat of dental peri-implant disease. It was consistent with previous results where TNF-α (G-308A) polymorphism was not significantly associated with the risk of dental implant disease. However, it suggests conducting studies on different ethnicities with increased sample sizes and gene-gene and gene-environment interactions to compare the results [16]. In the 2019 review by Chen and Zhao [17], they confirmed that many implant losses in patients cannot be explained by clinical factors alone, and several phenomena would imply the existence of genetic risk factors for implant failure. The finding of Mo et al. [16] regarding the genetic polymorphisms of IL-1A–889C/T or IL-1B + 3954C/T shows the association between peri-implantitis and periodontal status. He et al. [18] assessed three genetic variants: tumor necrosis factor-alpha (TNF-α)-308G/A, interleukin-1 alpha (IL-1A)-889C/T, and IL-1 beta (IL-1B) + 3954C/T-as risk factors for peri-implantitis in 144 patients with peri-implantitis and 174 healthy controls in non-smoking Chinese. Their results suggested that IL-1A – 889C/T or IL-1B + 3954C/T genetic polymorphisms showed associations for peri-implantitis risk and periodontium inflammation. Confirming that genetic polymorphisms would be constant and could be measured before the onset of the disease, so they would be beneficial for planning and prognosis in the early stages of the disease, being of benefit to the patient [18]. Therefore, with these data, we can affirm that inflammation is crucial for osseointegration and, therefore, implant success. The same, if maintained for a considerable time, leads to the loss of supporting tissues, mainly bone, which results in implant failure.

When a single nucleotide polymorphism occurs in proinflammatory genes, this would impair the expression and/or sequence of amino acids, and this would directly affect the host’s inflammatory response. Therefore, identifying the polymorphic variants that are related to implant losses would allow the development of protocols for prevention, which would reduce costs and complications for patients. Moreover, applying these methodologies to patients with other risk factors could also influence implant survival.

The clinical studies we carried out [19] were able to analyze the relationship between the IL-10 A/G rs1800896 and TNF-α G308A rs1800629308 polymorphisms with the loss of dental implants and periodontal disease. But the polymorphisms evaluated were not predictive of the failure of dental implants for the number of cases studied. If we verify that there is a significant relationship between periodontal disease and the TNF-α genotype. This could lead us to assume that these patients have osteolytic capacity given by TNF and a lower capacity to maintain the integrity and homeostasis of the epithelial layers given by the mutated allele of IL 10. This should be corroborated in new studies expanding the samples. Systematic reviews and meta-analyses allow for more robust and generalized conclusions by considering several studies and, therefore, larger groups of cases and controls to identify some models of risk markers. These models can aid in screening, early diagnosis, and/or therapy in the clinic.

The hypothesis we proposed for our work was that polymorphic population variations influence the survival rate of dental implants.

Algorithm-based diagnoses are applied worldwide in periodontics and other areas, which provide contributions to the design of health policies to raise people’s quality of life.

General and Specific Objectives

The general objective of this work was to evaluate the IL-10 A/G rs1800896 and TNF-α G308A rs1800629308 polymorphisms with the loss of one or more dental implants through a systematic review and meta-analysis of clinical trials.

Materials and Methods

Data Sources

PRISMA guidelines established for systematic reviews and meta-analyses were followed. It was carried out on randomized controlled clinical trials, epidemiological studies obtained from the MEDLINE, EMBASE, LILACS, IBECS and Cochrane databases.

Study Selection

It was carried out in a double-blind manner, coding each article and giving it to two independent reviewers with the checklist added at the end. The checklist that will be applied will be the one published by Bono et al. [19].

Inclusion Criteria

Articles (cases and controls) in English, with human subjects, that evaluated the specific types of genetic polymorphisms that could be associated with dental implant failure. The search was expanded as necessary, including the references of the articles that were included.

Exclusion Criteria

Original published studies that do not include quantifiable data in relation to the clinic. Studies with pregnant patients, with some syndrome, or some systemic pathology, or who were taking medication.

Data Analysis

The Mantel-Haenszel method and the Peto method were used to estimate the fixed and random effects because the type of result analyzed was binary, related to statistical tests of Odds ratio, Relative risk or Risk difference. The R program version 2.10.1 ( www.r-project.org) will be used, package meta function metabin. The Q test was used to evaluate heterogeneity. The following models were used to construct the prediction rule for the risk of implant loss:

• Logistic Regression

• Regression Trees–Classification (ARC)

Information Search

The following mesh terms in English were used to search for articles in the databases: MEDLINE, EMBASE, LILACS, IBECS and Cochrane:

• ((“Interleukin-10”[Mesh]) AND “Polymorphism, Genetic”[Mesh]) AND “Tumor Necrosis Factor-alpha”[Mesh]

• (((“Interleukin-10”[Mesh]) AND “Polymorphism, Genetic”[Mesh]) AND “Tumor Necrosis Factor-alpha”[Mesh]) AND “Dental Implants”[Mesh]

• ((“Interleukin-10”[Mesh]) AND “Polymorphism, Genetic”[Mesh]) AND “Tumor Necrosis Factor-alpha”[Mesh] and implant

The analysis, classification, and screening of articles were developed by two blinded researchers. The excluded studies were those that did not respond to the PICOS strategy, that were not RCT (Randomized Controlled Trial), and those studies in which the patients’ diagnosis did not correspond to lost implants.

The initial search returned 593 results. A manual search was performed, reviewing the title and abstract of each article. We excluded duplicates and studies that were systematic reviews, and 28 articles were selected. We read these 28 articles completely, excluding 24 studies that did not characterize the genetic polymorphism we were looking for related to dental implant survival. Therefore, we included 4 articles in this review (Fig. 1).

Fig. 1. Study selection–PRISMA flow chart.

Genetic polymorphisms of interleukin (IL)-1A, IL-1B, and tumor necrosis factor-α were considered in the included studies (Table I). In our analysis, we were unable to significantly correlate the polymorphisms with dental implant failure. Of the polymorphisms studied, LTA (TNF-β), IL-1A (−889), and IL-1B (+3,954) showed a higher association with dental implant loss, and IL1B (C-511T) presented a slight association with implant loss.

Table I. Characteristics of Studies

Gene/polymorphism	Authors	Origin	Design: CC: case-control study	Diagnosis	Cases(place)	Controls(place)	GenderAF (RF)	Age (years) Average ± SD	Smoking	No smoking
LTA, TNFA, and LTB	Broker et al. [20]	Brazil	CC	Loss of dental implants	n163 Latin-American Dental Research Institute (ILAPEO)	n81 Latin-American Dental Research Institute (ILAPEO)	Female: case 51 (63) control: 111 (68.1) Male: case 30 (37.0) control 52 (31.9)	Cases: 52.9 ± 11,0 Controls: 51 ± 13,0	Cases: 15 (18.5) Controls: 32 (19.6)	Cases: 2 Controls: 131 (80.4)
IL-10 A/G rs1800896, TNF-α G308A rs1800629	Bono et al. [19]	Argentina	CC	Loss of dental implants	n8 Specialization Career in Oral Implantology, Catholic University of Córdoba	n95 Specialization Career in Oral Implantology, Catholic University of Córdoba	Female: case 4 control: 51 Male: case 4 control 44	53.06 ± 16.22 (20–80)	Cases: 6 Controls: 35	Cases: 2 Controls: 60
L-1A (−889), IL-1B (−511), IL-1B	Cosyn et al. [21]	Belgium	CC	Early implant failure	n14 University Hospital in Ghent, Belgium	n14 University Hospital in Ghent, Belgium	Female: case 6 control: 8 Male: case 8 control 6	Cases 64 (11). Controls 67 (11)	Cases: 4 Controls: 4	Cases: 10 Controls: 10
IL1B (C-511T)	Dirschnabel et al. [22]	Brazil		Implant failure	n92 Latin-American Dental Research Institute (ILAPEO)	n185 Latin-American Dental Research Institute (ILAPEO)	Female: case 56 control: 122. Male: case 36 control 63	Cases 54.63 (10.44). Controls 53.13 (11.46)	Cases: 18 Controls: 43	Cases: 74 Controls: 142

The polymorphisms studied in the articles are not comparable since the rs (Reference SNP cluster ID or rsID) is different, which is the label assigned to each genetic variant or genetic polymorphism (SNP) found in a significant proportion of the population.

Characteristics of the Included Studies

We considered all these classes of genes with the polymorphisms (IL-1A [−889], tumor necrosis factor (TNF-α [−308]), IL-10 [−592], IL-1B [−511], IL-1B [+3954], IL-1B [+3953], IL-10 [−1082], IL-1 receptor antagonist [IL-1ra]), IL-10 [−819] (Table I).

The genetic polymorphisms of IL-1A (−889) were the subject of a study by one author, Cosyn et al. [21], who demonstrated an association with biological complications for the loss of dental implants and were statistically significant in one of them (Table II).

Table II. Summary of the Relationship of the Polymorphisms Studied with the Possible Association with Implant Loss/Periodontitis

Authors	Polymorphisms studied	Considerations
Broker et al. [20]	TNFA rs1800629	No association between implant loss
	LTA (TNF-β)	Associated with implant loss LTA rs2009658 (allele C)
	LTB (TNF-C)	No association between implant loss
Bono et al. [19]	TNFα rs 1800629	No association for implant loss but with periodontitis (GG)
	IL-10 rs1800896	No association for implant loss but with periodontitis (AG)
Cosyn et al. [21]	IL-1A (−889)	Association for implant loss (T allele)
	IL-1B (−511)	No association for implant loss
	IL-1B (+3,954)	Association for implant loss (T allele affects osseointegration).
Dirschnabel et al. [22]	IL1B (C-511T)	Association at the limit with implant loss

The genetic variations of IL-1B (−511) were explained in two articles Cosyn et al. [21] and Dirschnabel et al. [22], and found no association with implant failure, while IL-1B (+3954) was related to peri-implantitis and the poor life or survival of implants in only one of the studies we cite because it directly alters osseointegration (Table II).

Results

The failure of dental implants is a multifactorial process (systemic conditions of the patient, patient hygiene, habits, prosthetic overload), and when one observes failures in patients’ implants in the clinic, the question arises about the existence of susceptibility to their failure [23].

We rely on the inflammatory immune response for implant survival, but genetics can favor or be a problem, acting as protective or destructive factors, and modifying the patient’s immune response. [21], [22] Genetic alterations that increase the activity of IL-1A (−889), LTA (TNF-β), IL-1B (+3,954), and IL1B (C-511T) on the different cellular components of the immune system [19], [20]–[22] they manage to destroy the site because they generate greater activity of bone metabolism and this results in bone loss all around those implants and therefore a decrease in the survival of the implants (Fig. 2).

Fig. 2. Estimating the effects of each study in odds ratios.

ILs play a significant role in bone remodeling as a key mediator of the inflammatory process, initiating the cascade of decomposition of the components of the extracellular matrix through enzymes essential for the structural maintenance of the extracellular matrix and bone degradation due to the reciprocal action between RANK/osteoprogerin [22].

The most studied polymorphisms in the literature, which are normally found in the hyperinflammatory response, are the IL-1A (−889) and IL-1B (+3954) genes, with a high concentration in the periodontal sulcus of high-risk patients. Cosyn et al. [21] published this study of 14 cases and controls who had failed one or more implants within 6 months of their placement. They took blood samples and performed PCR. Their most important results were those of the T allele of IL-1A (−889) (p = 0.039) and the T allele of IL-1B (+3.954) (p = 0.003) in implant failure. Furthermore, the genotype distribution was very different between cases and controls for IL-1B (+3.954) (p = 0.015), which would indicate its significance in osseointegration (Table III) [21]–[23].

Table III. Results Polymorphisms Studied

Author	Country	Diagnosis	Polymorphism	Genotype	CI 95%	Cases	Controls	OR
Broker et al. [20]	Brazil	Loss of dental implants	TNFα rs1800629	A/G	0.49 −1.86	n = 81	n = 163	0,95
				Recessive Hz 1		63	128
				Recessive Hz 2		16	63
Bono et al. [19]	Argentina	Loss of dental implants	TNFα rs 1800629			n = 103	n = 103	2,09
				A/A		2	24	ref. category
				G/A		2	38	1,58333333
			IL-10 A/G	G/G		4	33	0,6875
Cosyn et al. [21]	l Belgium	Early implant failure	IL1B-511T			n = 14	n = 14
				A/A		3	2	ref. category
				A/G		6	6	1,5
				G/G		5	6	1,8
			IL-1B +3,954			n = 14	n = 14
				C/C		6	13	ref. category
				C/T		6	1	0,07692308
				T/T		2	0	no estimado
Dirschnabel et al. [22]	Brazil	Implant failure	IL1B C-511T			n = 92	n = 185
				C/C		30	69	ref. category
				C/T		41	88	0,93319194
				T/T		21	28	0,57971014

Cosyn et al. [21], and Sampaio Fernandes et al. [24], found that people carrying the T allele of IL-1A (−889) (p = 0.006) and IL-1B (+3954) (p = 0.03) multiplied their risk of suffering from peri-implantitis, for a Chinese population [21]–[24].

Our research team in a previous study observed that polymorphisms in IL-10 A/G rs1800896 and TNF-α G308A rs1800629 may affect the results of peri-implantitis treatment in genotypically positive individuals (Fig. 3).

Fig. 3. Effect size for TNFα rs 1800629 polymorphism associated with periodontitis (GG).

We evaluate the genetic relationship with the progression of the disease, for which we compare patients with peri-implantitis with patients with periodontal health, comparing their clinical parameters. We did PCR for IL genotyping. As a result, 8 patients in the peri-implantitis group were genotypically positive (80%), and 33 patients in the healthy group were also positive (40%). Both groups presented statistical significance (p < 0.0164). [19]

Melo et al. [25] also found, like Cosyn et al. [21], that there was nothing of significance between this polymorphism (IL-1B [+3,954]) and the failure of rehabilitation. He examined 31 implants without pathology, 16 implants with peri-implantitis, 31 healthy teeth with healthy implants, and 16 healthy teeth from patients with peri-implantitis) took crevicular fluid and performed genomic DNA analysis. Concluding that the concentration of IL-1B (+3.954) (p = 0.0814) found in the patients was not relevant, nor in any of the variables, be it the genetic polymorphism studied nor peri-implantitis, nor in implant failure [26], [27].

Conclusion

As shown, there are very few exhaustive studies that have the same methodologies and similar data analysis to be able to compare them that include the variability of modified SNPs in the gene related to the reduction in implant survival to carry out the meta-analysis. This systematic review confirms results that are still controversial due to the limited sample size, the definition in the terminology dental implant failure not yet defined equally in all studies, and the non-homogeneous inclusion and exclusion criteria, including race, age, smoking, occlusion trauma, and periodontal disease. In numerous genetic studies, the different types of possible allelic combinations are not differentiated; It is only detected whether or not there is a genetic influence. Allelic discrimination makes it possible to detect genetic polymorphisms and specific mutations, facilitating the understanding of the relationship between genotype and phenotype. Without this discrimination, there is a risk of generalizing genetic effects, overlooking how particular allelic combinations can influence gene expression and disease manifestation.