A Comprehensive Review of Peri-implantitis Risk Factors

The diagnosis, or history, of periodontal disease is the most researched factor associated with peri-implantitis. This is partially attributed to similarities in the subgingival microbiota between the diseased teeth and implants [10•]. Submucosal presence of certain periodontal pathogens has been significantly associated with peri-implantitis, with an odds ratio of 15.1 [11], but the results are still controversial [12]. Current data suggests that peri-implantitis is associated with a specific microbiota resembling that of periodontal lesions, in addition to other microorganisms not commonly related to periodontitis [13]. Nevertheless, it is well-accepted that peri-implantitis consistently presents with marked microbial diversity [13, 14], and that deeper peri-implant pockets exhibit significant microbial alterations and higher levels of dysbiosis [15].

Periodontal disease has been strongly associated with peri-implantitis [5•, 16]. Active periodontitis at the adjacent teeth is further considered a predictor of future peri-implantitis [9]. Periodontally compromised patients have twice the risk of developing peri-implantitis compared with healthy individuals [10•]. Moreover, those with a history of generalised aggressive periodontitis are 5 times more prone to implant failure, and 14 times more susceptible to peri-implantitis, compared with healthy controls [17]. Fortunately, successful treatment of periodontal disease prior to implant placement has been shown to lower the risk of peri-implantitis [18], and is therefore considered an essential initial part of the overall treatment plan.

Supportive therapy has been shown to significantly lower the risk of peri-implant biological complications, and a minimum recall interval of 5–6 months has thus been recommended [18, 19]. Maintenance programs should be tailored to the individual’s specific needs and susceptibility to both periodontal and peri-implant diseases. Factors used for risk assessment include the percentage of BOP, the prevalence of active residual pockets, oral hygiene level, smoking habits and the presence of systemic or genetic conditions [20]. Individuals with high-risk profiles require three to four annual visits [20, 21], and their attendance is detrimental for prevention and early detection of peri-implantitis [22]. One out of five non-compliant patients is diagnosed with peri-implantitis within 5 years [23]. On the other hand, compliance is associated with 86% fewer peri-implantitis cases. Unfortunately, those with greater needs have been known to be the least compliant. The extent and severity of periodontal disease, as well as the patient’s smoking habits, affect adherence to maintenance programs [22]. Therefore, it is the clinicians’ duty to adequately inform their patients of the importance of regular supportive therapy for the prevention of peri-implantitis.

The negative effects of smoking on periodontal health have long been well established. It impacts innate and adaptive immune responses, impairing the host’s defence mechanisms and its response to microbial challenges [24, 25]. Cigarette smoking also affects wound healing, as it is therefore detrimental to periodontal treatment [21, 26, 27]. Smoking further increases the oxidative stress and inflammatory burden with marked alterations in microbial flora [28]. It significantly affects implants’ colonisation with periodontal pathogens such as Porphyromonas gingivalis (Pg) and Fusobacterium nucleatum [28]. Besides, cigarettes are not only harmful to smokers, but mere exposure to environmental smoke increases the risk of developing periodontal disease by 28% [29].

Studies have repeatedly proven smoking as a risk factor for peri-implantitis [9, 30, 31]. Smokers are almost twice more at risk of developing peri-implantitis compared with non-smokers [5•]. Moreover, smoking is associated with increased severity of peri-implantitis lesions [16], with a dose-dependant relationship between smoking and tissue destruction [26, 32]. Nevertheless, smoking cessation has been shown to positively impact periodontal health, with favourable effects on both incidence and progression of the disease [26, 33].

In addition to cigarettes commerciality, the popularity of non-cigarette tobacco products has been alarmingly rising. Water pipes, also known as shisha, hookah or narjilah, have become a popular way of smoking tobacco among adolescents and adults alike [34, 35]. Their recreational use has become widely acceptable despite containing high levels of nicotine, and a multitude of carcinogens and heavy metals. In addition, water pipes emit a variety of pollutants generated by the charcoal used to heat the tobacco. They are smoked for hours in social settings, thus extending the amount of smoke inhalation and its side effects. Second-hand smoke inhalation should also be taken into account. The link between water pipe smoking and periodontal disease has already been established by several studies [35, 36]. Furthermore, water pipe smokers have a significantly higher risk of periodontitis compared with cigarette smokers, but adverse effects were strongly associated with the duration and the quantity of daily use [37]. However, studies have so far focused on periodontal conditions in general, and not on peri-implantitis in particular. But in analogy to cigarette smoking, water pipe smoking presents a possible risk factor for peri-implant disease.

Smokeless tobacco is yet another factor associated with periodontal disease, specially in the absence of adequate oral hygiene measures [38]. It is most commonly used in India and Southeast Asia [39]. The adverse effects of smokeless tobacco on both periodontal and peri-implant tissues are comparable with those of cigarette smoking [40, 41]. Deeper probing depths and higher degrees of peri-implant bone loss were found in cigarette smokers and smokeless tobacco users compared with non-tobacco users [41].

Electronic cigarettes (e-cigarettes), or vaping, have lately become an extremely widespread trend among individuals of all ages. They are widely misconceived as harmless recreational products. Yet their increasing popularity, combined with the lack of evidence on long-term health effects, has become rather disquieting [42]. Regardless of their nicotine content, e-cigarettes have been shown to increase oxidative/carbonyl stress and pro-inflammatory responses, with adverse effects on endothelial cells and fibroblasts, and concomitant dysregulation in periodontal repair [43‐45]. Clinical studies have associated vaping with periodontal attachment loss and marginal bone resorption [44, 46]. A recent cross-sectional study had further demonstrated significantly deeper peri-implant probing depths and increased marginal bone loss in vaping patients compared with never smokers. However, this study did not account for past cigarette smoking as a confounding factor that could have influenced the results [45]. Further research is required to shed the light on the extent and severity of peri-implant complications, as well as the impact of vaping on general health.

Among smokable illicit substances, cannabis is one of the most commonly used drugs worldwide [47, 48•]. Following its recent legalisation in several countries, the plausibility of an association between cannabis and peri-implantitis should be evaluated. Different studies have already established higher prevalence and severity of periodontitis in cannabis users, irrespective of concomitant tobacco smoking [48•, 49]. However, an animal study demonstrated bone loss on the periodontitis-affected teeth exposed to cannabis without significantly affecting the periodontally healthy teeth [50]. Meaning that cannabis only seems to aggravate periodontitis-associated bone loss. Since periodontitis is more prevalent in adults [51, 52], this could explain why clinical studies showed a higher impact of cannabis on older individuals compared with adolescents. However, the drug’s mechanism of action and its exact pathway to periodontal destruction is still unclear [48•, 53]. Finally, despite studies supporting an association between cannabis use and periodontitis, the evidence is still lacking regarding peri-implantitis.

The influence of certain systemic diseases on periodontal health has long been established [54‐56]. Due to its increasing prevalence, diabetes mellitus is one of the most thoroughly researched conditions in the literature. It affects 415 million adults worldwide, with 642 million projected in 2040 (The International Diabetes Federation; 2015). The disease affects insulin’s secretion, its function or both, causing disruption of glycaemic levels. This consequently results in a variety of neuropathological, retinal, microvascular and renal complications [57].

Poor glycaemic control plays a pivotal role in the progression and severity of periodontitis [58]. This association has been explained by several vascular and cellular responses, leading to enhanced tissue destruction and impaired healing response [59]. Similar mechanisms are triggered in peri-implant tissues; resulting in a higher susceptibility to peri-implantitis in individuals suffering from hyperglycaemia [56]. Poorly controlled diabetics are at 46% higher risk of developing peri-implantitis, with deeper peri-implant pockets and higher marginal bone loss, compared with their normoglycaemic controls [55]. Interestingly, smokers and poorly controlled diabetics are considered at a similar risk for peri-implantitis. On the other hand, non-smokers with poor glycaemic control are 3.39 times at higher risk of developing peri-implantitis compared with normoglycaemic individuals [56]. Therefore, hyperglycaemia, not diabetes per se [60••], presents a significant risk factor for peri-implantitis.

Obesity is another highly prevalent condition with detrimental effects on periodontal health [61, 62]. It is defined as abnormal or excessive body fat accumulation with debilitating effects on general health [63]. This is a major medical problem associated with marked physiological changes, including diabetes mellitus and coronary heart disease [64]. Obesity is also associated with a generalised and constant hyper-inflammatory state, causing an altered immune response and increased production of pro-inflammatory cytokines, which adversely affect periodontal tissues and alveolar bone levels [31, 65]. Clinical studies have established obesity as a risk factor for peri-implantitis [62, 66, 67]. When compared with individuals with normal body weight, obese patients present with significantly higher percentages of BOP, deeper peri-implant probing depths and increased marginal bone loss [66, 67]. The severity of peri-implant inflammation is significantly associated with the level of obesity [62].

Despite their prevalence, few studies have examined the association between cardiovascular diseases and peri-implantitis. Most showed a significantly higher risk of peri-implantitis and additional bone loss for patients suffering from heart disease [54, 68, 69]. Yet the results are still controversial [54, 70]. The influence of such conditions on implants’ success should be further explored in larger studies with adequate methodology.

Discussions have also been raised regarding the effect of different autoimmune diseases on peri-implantitis, but conclusions cannot be drawn due to the scarcity of evidence [54]. Rheumatoid arthritis with concomitant connective tissue disease had been associated with higher percentages of BOP and peri-implant bone loss in one study [71]. Another study [72] evaluated patients with Sjögren’s syndrome but could not show an increased prevalence of peri-implantitis. Yet the higher prevalence of mucositis in the diseased group might indicate an increase in their susceptibility to peri-implantitis over time. Further research is still required to shed the light on the association between autoimmune diseases and peri-implantitis.

While osteoporosis could not be linked to peri-implantitis, anti-resorptive medications, including bisphosphonates (BP) and hormone replacement therapy (HRT), have been gaining increasing attention. One study [73] showed significant increase in marginal bone loss and implant thread exposure with BP intake. However, a recent systematic review [74•] showed that a low-dose BP did not negatively affect peri-implant bone levels. On the other hand, HRT significantly compromised marginal bone levels. Still, the considerable risk of medication-related osteonecrosis, and its negative influence on peri-implant hard tissues, should not be underestimated.

Despite the general belief in a certain genetic predisposition to peri-implantitis, a clear association with specific risk factors is still to be determined [75]. Interleukin-1 (IL-1) polymorphism is the single most researched genetic factor in the literature. This is mainly due to the involvement of this gene cluster in the encoding of two main pro-inflammatory cytokines, IL-1α and IL-1β, as well as the anti-inflammatory IL-1 receptor antagonist. Increasing levels of both IL-1α and IL-1β have been associated with peri-implantitis, and their levels were correlated with the severity of the disease [75, 76]. Yet, studies evaluating the correlation between IL-1 polymorphism and peri-implantitis have shown conflicting results [75, 77‐79]. While several investigations [76, 78] could not find a significant association between the two conditions, a recent study [79] showed that subjects with IL-1 polymorphisms were 1.9–2.47 times more at risk of developing peri-implantitis. Discussions have also been raised regarding a synergistic effect of smoking on individuals with IL-1 polymorphisms without reaching a consensus [76, 80].

Tumour necrosis factor-alpha (TNF-α) is another pro-inflammatory cytokine associated with peri-implant inflammation and bone destruction [75]. Studies have shown that TNF-α polymorphism increased the risk of peri-implantitis five to eightfold [75, 81]. Yet, the meta-analysis of relevant studies could not establish a significant correlation [82].

Very few studies have examined other genetic polymorphisms, and conclusions cannot be extrapolated due to the scarcity of the evidence [75]. Currently, only preliminary investigations, including a wide variety of ethnic groups, have evaluated other genetic markers in association with biological implant complications. Additional research is still required, with larger sample sizes and reduced levels of bias.

Occlusal overload of implant-supported prostheses is a controversial subject, and the exact mechanism in which it causes marginal bone loss is still debatable [83, 84]. Yet several studies have demonstrated that overloading an implant beyond a certain threshold leads to marginal bone loss [83, 85, 86]. Moreover, under similar overload, peri-implantitis-affected sites show significantly higher marginal bone loss compared with those with mucositis [87]. Also, the patterns of bone resorption varied significantly around overloaded implants and those with ligature-induced peri-implantitis [84]. The effect of overloading on peri-implant bone levels can be accentuated by sub-optimal implant positioning, poorly designed prosthetic reconstructions, inadequate bone quantity or its poor quality. Para-functional habits leading to elevated non-axial occlusal forces may also increase marginal bone loss [83]. On the contrary, several animal studies demonstrate an insignificant effect of overload on bone levels in the absence of inflammation [83, 88, 89]. Unfortunately, ethical reasons impede resolution of such controversy using controlled clinical trials with experimental periodontitis models or intentionally overloaded implants.

Attrition and wear of natural dentition or prosthetic reconstructions may be used for diagnosis of occlusal overload and para-functional habits. The presence of wear facets on implant-supported prostheses is associated with a 2.4 increase in the prevalence of peri-implantitis [90]. However, case reports [91, 92] have demonstrated that occlusal adjustment may result in marked peri-implant bone repair. Therefore, occlusal overload may be considered a potential risk factor for peri-implant bone loss, with an aggravating influence on peri-implantitis-associated bone loss.

The influence of an implant’s surface topography on its susceptibility to peri-implantitis is still debatable [4, 5•, 90, 93, 94]. The implant’s roughness and surface energy have an impact on initial biofilm formation, but its long-term effect on the inflammatory process and bone stability is still controversial [93, 95, 96]. Few studies demonstrated that rough implants were more susceptible to peri-implantitis [5•, 97], while others could not show a significant difference between rough and moderately rough surfaces [5•,98]. A recent systematic review [4] calculated that moderately rough implants are three times less affected by peri-implantitis compared with rough or machined ones. Yet another review published only a year earlier [99•] completely contradicted these results. The later found significantly lower bone loss around minimally rough surfaces compared with moderately rough and rough ones. However, the longer follow-up periods of both minimally rough and rough implants in comparison with the new generation of moderately rough fixtures should be taken into consideration. Moreover, the presence of other risk factors could not always be ruled out. Therefore, definite conclusions cannot be drawn regarding the effect of an implant’s topography on peri-implantitis.

So far, titanium has been the material of choice in implant dentistry. Nonetheless, zirconia ceramic implants have been progressively emerging [100]. Zirconia’s greatest assets lie in its biocompatibility, superior soft tissue integration [101, 102••], low affinity to plaque [103] and reduced inflammatory processes when compared with titanium [104, 105]. It was hence hypothesised that zirconia implants would finally offer the solution for peri-implant disease. Unfortunately, a recent animal study had clearly demonstrated that zirconia implants can be affected by peri-implantitis [106•]. Still, zirconia demonstrated significantly lower marginal bone loss compared with titanium implants with similar surface topographies. Clinical studies have also demonstrated different degrees of bone loss around zirconia implants with variable designs [107], but additional long-term data is still required to establish both prevalence and treatment protocols.

The design of the prosthetic reconstruction largely contributes to the implants’ long-term success. Poorly designed superstructures significantly impede plaque control, which increases the risk of developing peri-implant disease [21, 108]. Patients with higher plaque indexes were predicted to be 7.9 times more at risk of developing peri-implantitis [9]. Moreover, the relationship between the implant and its prosthetic superstructure, whether centerer or offset, significantly affects its prognosis. An asymmetric prosthesis with a sub-optimal emergence profile favours plaque accumulation, consequently increasing the risk of peri-implantitis 4.3 times [9, 109]. A poor marginal fit is also a detrimental risk factor for the development of peri-implantitis [16].

Cemented implant restorations are 3.6 times more prone to peri-implantitis compared with screw-retained ones [90]. This is mostly attributed to the risk of leaving excess cement in the sub-mucosal region, especially when resin luting agents are utilised [110]. Therefore, deep sub-mucosal margins should be avoided in order to provide sufficient visibility and access for cement removal [111].

Full-mouth implant-supported fixed reconstructions have been associated with a 16-fold increase in peri-implantitis compared with single crowns [90]. This was mainly due to poor accessibility for plaque control. Furthermore, a history of advanced periodontal disease, leading to extensive loss of natural dentition, could not be ruled out in cases requiring complete rehabilitation.

Bone-level implant designs, combined with convex restorations at an angle exceeding 30°, significantly augment the risk of peri-implantitis [109]. Platform switching has been recommended to reduce peri-implant bone loss, but its benefits are still debatable [112]. Nevertheless, a systematic review evaluating earlier systematic analyses favoured platform switching for peri-implant bone preservation [113]. This could be due to the relocation of the microgap between the implant and the abutment. This microgap is wide enough to allow for bacterial colonisation, and its horizontal offset away from the bone is believed to reduce the risk of peri-implant inflammation [114•]. Platform switching has also been supported by a recent clinical study which showed a significantly reduced probability of developing peri-implantitis [115]. Another reason for advocating platform switching is the reduced amounts of tribocorrosion products released into peri-implant tissues. Alrabeah et al. [116•] had demonstrated that platform-matched implants released higher amounts of metal ions, and exhibited more surface damage, compared with platform-switched implants. But keep in mind that the association between titanium particles’ release and peri-implantitis is still debatable [117••].

Therefore, tissue-level implant designs are recommended for non-aesthetic implant restorations due to the supra-mucosal location of the microgap and their accessibility for plaque control. When bone-level implants are indicated, platform switching is advisable with screw-retained superstructures and anatomically shaped emergence profiles. Screw-retained prosthesis is also easily retrievable when better visibility and access are required for treatment.

The soft tissue condition around an implant may influence its susceptibility to peri-implant disease. Patients with thin periodontal phenotypes are more prone to peri-implant mucosal recessions [118]. The exposure of an implant’s rough surface to the oral cavity complicates plaque control and enhances bacterial adhesion, thus leading to a potential increase in its susceptibility to peri-implantitis. A recent clinical study had demonstrated a significant association between thin biotypes and the severity of peri-implantitis [119]. Yet the lack of additional research precludes definitive conclusions.

A minimal 2 mm zone of keratinised soft tissue (KT) has been advocated for peri-implant health and long-term stability. Yet its absolute necessity is still controversial [60••, 120]. It has been associated with better plaque control, lower soft-tissue inflammation, mucosal recession and attachment loss [121]. Brushing discomfort and higher plaque scores have been reported at sites with insufficient KT, even in patients with generally good oral hygiene [122, 123]. Moreover, peri-implantitis and marginal bone loss have been associated with KT width < 2 mm [23, 123, 124], particularly in patients not compliant with regular supportive therapy [125]. On the other hand, periodic maintenance resulted in a low incidence of peri-implant disease regardless of the width of KT [126]. Finally, despite the lack of evidence regarding a direct association with peri-implantitis, a 2-mm band of KT is highly recommended.

While the number of implants does not seem to influence the risk for peri-implantitis [127], their position is critical for long-term success [128]. Implant malpositioning represents a significant risk factor for peri-implantitis [128]. Crestal bone resorption could occur when an implant is placed too close to the natural teeth or even other implants [129]. This could compromise access for plaque control, and thus increase the risk of peri-implant disease. Also, fixtures located outside the bony envelope or those with thin facial bone (< 1 mm) are more prone to mucosal recession, especially in patients with thin biotypes. This exposure of the fixture’s rough surface increases plaque retention [118], and thus the risk of peri-implantitis. Bone and/or soft tissue grafting is recommended in such cases [129], keeping in mind that augmentation procedures do not increase the risk of biological complications [130].

Moreover, placing an implant 6 mm or more apical to the cemento-enamel junction of the neighbouring teeth increases its risk of peri-implantitis 8.5 times [9]. A deep sub-mucosal position also complicates plaque control and increases the susceptibility to peri-implant inflammation [9, 21].

Despite the availability of zirconia implants, titanium remains the material of choice in implantology. However, the release of titanium particles, and their impact on peri-implant tissues, has recently become subjects of heated debates [117••, 131]. Mechanical wear, chemical corrosion and implant surface treatment have been suggested as sources of titanium in the oral environment [132]. The term « tribocorrosion » has been used to describe the combination of wear and corrosion processes [133]. More specifically, corrosion can be observed once mechanical wear has disrupted the protective titanium oxide layer. This is not only confined to the surface of the implant, but can also affect the implant-abutment interface [116•]. From this perspective, the location of the microgap and the quality of the abutment connection are of major importance. From there on, tribocorrosion becomes auto-sustained, since a corroded surface becomes less resistant to mechanical wear.

Microbial contamination could also lead to the release of titanium particles. The acidic inflammatory environment contributes to the oxidation and destruction of the superficial implant layer, and hence the release of metal ions. This consequently amplifies peri-implant inflammation and disease progression [134]. The local release of titanium has also been associated with lipopolysaccharides producing gram-negative bacteria, such as Pg, and an over-expression of pro-inflammatory cytokines [135]. Greater levels of titanium particles have been detected in peri-implant soft tissue biopsies taken from fixtures with peri-implantitis compared with healthy sites [136, 137].

Certain peri-implantitis therapeutic measures may further contribute to titanium’s release into the peri-implant region. This includes chemicals used for implant surface decontamination, such as chlorhexidine and hydrogen peroxide [138], and mechanical devices like ultra-sonic tips. Implantoplasty, frequently used to remove exposed implant spears and polish rough surfaces, has also been associated with significant release of metallic particles [131, 139]. Moreover, titanium’s dissemination into distant organs, such as the lungs and lymph nodes, should be considered [134].

Regardless of the source, it remains to be determined if titanium particles could be the sole cause of peri-implantitis, or merely a consequence of microbiological, chemical and mechanical factors. Moreover, the effect of the particle’s size should be taken into account. Nanoparticles have a higher biological activity and hence are significantly more harmful than microparticles. Also, keep in mind that harmless levels of titanium oxide are commonly found in cosmetics, toothpastes, suncreams and even various food products [117••].

A patient’s self-performed plaque control is one of the most important factors influencing the implant’s prognosis [23, 60••, 108]. A high plaque index was associated with an eightfold increase in susceptibility to peri-implantitis [9]. The accumulation of bacterial biofilm on implant and abutment surfaces leads to peri-implant inflammation, also known as mucositis [140••]. This cause-and-effect relationship has been further validated by the complete resolution of experimental mucositis once oral hygiene measures have been reinstated [140••, 141].

Peri-implantitis is always preceded by a period of mucositis. The two share several risk factors including poor oral hygiene, smoking and sub-mucosal presence of excess cement [140••]. However, not all mucositis lesions progress to peri-implantitis, even when present for extensive periods of time [142]. Conversion to peri-implantitis is more likely to occur in patients non-compliant with regular supportive implant therapy, with an odds ratio of 5.92 [143]. A concomitant effect of different previously described risk factors could lead to the onset of peri-implantitis. Nevertheless, implants diagnosed with mucositis are at risk of developing peri-implantitis [140••].