Network of potential mechanisms involved in the pathogenesis of periodontitis in diabetes. The hyperglycaemic state that characterizes diabetes has several deleterious effects. It drives the formation of irreversible advanced glycation end-products(AGEs) and the expression of their cheif signalling receptor RAGE. This interaction, in turn, leads to immune cell dysfunction, alters phenotype and function of other key cells in the periodontium, and contributes to cytokine imbalance with increased generation of certain pro-inflammatory cytokines. Hyperglycaemia also contributes to enhanced levels of reactive oxygen species (ROS) and a state of oxidative stress, both directly and indirectly through the AGE/RAGE axis, promoting quantitative and qualitative shifts in cyto-kine profiles. Finally, hyperglycaemia modulates the RANKL/OPG ratio, again directly and indirectly via the AGE/RAGE axis, tipping the balance towards enhanced inflammation and destruction. All the above, complemented by the effects of ecological shifts in the subgingival biofilm and the circulating adipokines generated due to diabetes-associated adiposity and dyslipidaemia, drive this vicious cycle of cellular dysfunction and inflammation. The end result is a loss of equilibrium where enhanced periodontal tissue destruction and impaired repair ensue, leading to accelerated and severe periodontitis. Importantly, as shown, several of the associations between the different elements in the figure are bidirectional, for example, the pro-inflammatory state further feeds the generation of AGEs, ROS, and adipokines, increases the RANKL/OPG ratio and helps pathogenic subgingival bacteria thrive. It is also important to note that (a) the amount and quality of evidence supporting the various pathways in this figure varies, and (b) although the goal is to depict the major mechanisms and networks described in the literature, other pathways and links among the various elements shown do exist, but cannot easily be demonstrated in a single schematic. Finally, the processes outlined are potentially modified by several other factors, such as genetics, age, smoking, stress, all of which may contribute significantly to inter-individual variations in disease experience (From Taylor et al (2013a,b). Reproduced with permission from the American Academy of Periodontology, European Federation of Periodontology and John Wiley and Sons)
The significance of the AGE/RAGE binding has been demonstrated in experimental studies using administration of soluble RAGE, which is the extracellular ligand-binding domain of RAGE, and administration of a RAGE antagonist prevented periodontitis progression in hyperglycemic diabetic mice (Lalla et al. 2000). Also, decreased levels of TNF-α, IL-6, and matrix metalloproteinases (MMPs) in the gingival tissue were found. Other studies have demonstrated that RAGE has fundamental influence on the increased periodontal tissue destruction, which is why antagonists of RAGE have been proposed as a therapeutic tool for the management of DM-associated periodontitis (Lalla et al. 2001). Interaction of AGEs with toll-like receptors (TLRs) has been described, as well as increased expression of TLR2, TLR4, and TLR9 in periodontitis-affected tissues of DM patients, as compared with periodontitis-affected tissue from controls without DM (Rojo-Botello et al. 2012). Further investigations of the TLR-mediated pathways in DM and periodontitis are obviously needed.
B cells, which dominate the inflammatory reaction in the established periodontitis lesion, have been described as a major source of receptor activator of nuclear factor-κB ligand (RANKL) with a pro-osteoclastogenic effect (Onal et al. 2012), and since RANKL expression is increased in mice with type 2 DM (Cao et al. 2010), an exaggerated RANKL expression may potentiate periodontal bone destruction in type 2 DM patients (Zhu and Nikolajczyk 2014). Another source of RANKL is the T cell, but the role of T cells, including whether T-cell produced RANKL plays a role in DM-associated periodontitis, remains to be clarified.
Several studies have investigated the influence of DM on the cytokine profile of patients with periodontitis, and the results reported so far are inconsistent; they are cross-sectional, or they are lacking confirmative support from other studies. Elevated levels of IL-1β in serum and crevicular fluid from DM patients with chronic periodontitis seem to be the most consistent finding (reviewed by Taylor et al. (2013b) and by Atieh et al. (2014)). Studies in animal models have also emphasized the role of TNF-α in prolonging the bacteria-induced immune response in DM-related periodontitis, but evidence from clinical studies is so far inconclusive (Taylor et al. 2013b).
The role of neutrophils in the development of periodontitis, in general, is considered protective, and changes in neutrophil function may account for an increased susceptibility to periodontitis. Indeed, neutrophil function in DM patients with periodontitis has been studied intensively. The outcome of studies based on peripheral neutrophils may conceivably differ from that of neutrophils located in periodontal tissues. However, signs of compromised neutrophil function have been presented in humans, since neutrophil-derived β-glucuronidase and IL-8, which has a chemotactic effect on neutrophils, were depressed in type 2 DM patients with periodontitis (Engebretson et al. 2006). Experimental animal studies in rodent models of DM and/or periodontitis have also revealed reduced neutrophil function (Golub et al. 1982; Sima et al. 2010).
It is well established that hyperglycemia in DM patients may predispose to periodontal tissue destruction, and a large amount of studies have scrutinized the possible pathologic pathways, by which DM may have impact on the course of periodontitis. Fewer studies have dealt with the pathways by which periodontitis may affect the course of DM. High levels of CRP in patients with both diseases have been associated with increased HbA1c levels, and since periodontitis itself may account for higher levels of CRP, the additional systemic inflammation associated with periodontitis may be responsible for the increased HbA1c levels in DM patients with periodontitis (Demmer et al. 2010). Insulin resistance in periodontitis patients with DM may be promoted by hyperreactive neutrophils producing reactive oxygen species, which, in turn, may stimulate pro-inflammatory pathways (Allen and Matthews 2011). An interesting association of periodontal microbiota with prediabetes prevalence in young adults has been described in a recent cross-sectional study (Demmer et al. 2015). Although it is up to future longitudinal studies to determine whether such interrelationships are causal, the finding that levels of potentially periodontopathic subgingival bacteria are abundant in and predictive of prevalent prediabetes is new knowledge (Demmer et al. 2015).
A recent review has focused on the significance of resistin, a biomarker for the levels of which are increased in chronic inflammation including periodontitis. Since resistin has been shown to induce insulin resistance in mice, it has been proposed as a possible link between periodontitis and DM (Devanoorkar et al. 2014).
Several experimental studies in rodents have provided insight in the possible interactions between periodontitis and DM (for review, see Andersen et al. (2007b)). Interestingly, ligature-induced periodontitis has been shown to deteriorate metabolic control in type 2 DM rats with an increase in oral glucose tolerance test of as much as 30 %, and an increase in IL-1β in adipose tissue compared to diabetic rats without periodontitis (Andersen et al. 2006). In prediabetic rats with ligature-induced periodontitis, the glucose tolerance was also significantly impaired, which suggests that periodontitis may facilitate the development of manifest type 2 DM (Andersen et al. 2007a). Moreover, the prediabetic rats with periodontitis developed renal alterations including kidney hypertrophy and a tendency for increased glomerular volume (Andersen et al. 2008).
Although a number of interactions of periodontitis with DM may appear obvious, there is still little evidence to understand the mechanistic pathways of periodontitis’ influence on DM, and most is presently speculative.
4.2.4 Outcome of Periodontal Treatment
A large number of studies have examined the role of periodontal treatment for the course of DM, but long-term randomized clinical trials are scarce. The studies are characterized by different inclusion criteria of patients, including various types of DM and various diagnostic criteria for a case of periodontitis. Moreover, stratification for confounders such as smoking, overweight, and medication is difficult. The current evidence has been critically reviewed and analyzed in several papers. A meta-analysis of the outcome of nonsurgical periodontal treatment was performed based on 15 papers selected on the following criteria: randomized controlled study in humans, intervention study on diabetic patients with periodontal disease, minimum 3 months follow-up observation, including data on HbA1c and/or fasting plasma glucose change after treatment, and clear presentation of population demographic data (Corbella et al. 2013). The majority of the patients included in the studies were affected by uncontrolled type 2 DM, and only one study involved patients with type 1 DM. The meta-analyses showed that nonsurgical periodontal treatment significantly reduces the level of HbA1c and fasting plasma glucose in patients with DM. The mean decrease of HbA1c was 0.4 % after 3 months and 0.3 % after 6 months, and the decrease in fasting plasma glucose was 9.0 mg/dL after 3 months and 13.6 mg/dL after 6 months, and there was no positive effect of adjunctive antimicrobials. The authors stated that it was difficult to quantify the clinical relevance of the findings in terms of improved glycemic control. Another meta-analysis of randomized clinical trials included five studies of patients with type 2 DM (Sgolastra et al. 2013). The inclusion criteria were almost similar to the abovementioned, and the primary outcome variables were changes in HbA1c and fasting plasma glucose, while secondary outcomes were changes in total serum cholesterol, serum triglycerides, and high- and low-density lipoprotein cholesterol. The result of the meta-analysis was that the periodontal treatment after 3–6 months resulted in a significant reduction in HbA1c, and in fasting plasma glucose, the mean differences being 0.7 % and 9.0 mg/dL, respectively. Periodontal treatment resulted in no significant differences in the secondary outcomes. This meta-analysis has been criticized for the use of too restrictive exclusion criteria, which may limit the generalizability of the meta-analysis to a fraction of the relevant population (Janket 2014). Finally, a meta-analysis has been presented of the effect of nonsurgical periodontal treatment on systemic inflammation in patients with type 2 DM (Artese et al. 2015). Exclusion of studies due to study design and missing data resulted in four included studies involving associations with CRP and two involving associations with TNF-α, the primary outcome measures being high sensitivity CRP (hsCRP) or CRP, IL-6, and TNF-α. Adjunctive antimicrobial therapy was combined with scaling and root planing in four of the included studies. A significant reduction as the result of treatment was found for both TNF-α (−1.33 ng/L) and hsCRP (−1.28 mg/L). Taken together, the studies indicate a positive effect on metabolic control and systemic inflammation of nonsurgical periodontal treatment. This is particularly evident in type 2 DM patients. The clinical significance of the improvements obtained, however, is uncertain. Even small reductions in HbA1c may result in significant clinical improvements in diabetic complications and mortality. Thus, for every percentage point decrease in HbA1c, 35 % reduction in microvascular complications has been reported, and an average at 0.2 % point reduction in HbA1c level associates with 10 % lower mortality in type 2 DM patients (UK Prospective Diabetes Study UKPDS Group 1998). The abovementioned reductions in HbA1c levels of 0.31–0.65 % after periodontal treatment, thereby constitutes an important public health benefit. Also, it should be remembered that patients with poor glycemic control may have more insufficient oral hygiene, and they may visit the dentist more infrequently than those with a better blood sugar control, as pointed out by Aggarwal and Panat (2012). This is why special periodontal treatment efforts are recommendable for this group of patients. As mentioned above, a large proportion of type 2 DM and prediabetes patients remain undiagnosed (Glumer et al. 2003; Guariguata et al. 2011), which is a general problem for the prognosis of the patients’ health condition in general. However, for the prognosis of the periodontal condition and the result of periodontal treatment, it is very important that these patients are diagnosed as early as possible. An easy and cost-effective way to diagnose type 2 DM is to measure HbA1c level in peripheral blood sampled from the finger (Heianza et al. 2011). Since the majority of adults attend the dental clinic independently of medical treatment needs, and since it is important for the dental treatment to know about diabetic state, it has been proposed to involve dentists in screening of some of their patients for diabetes. In favor of such an arrangement is the fact that the attitude of dentists and their patients is positive to these medical examinations performed in the dental setting (Greenberg and Glick 2012; Greenberg et al. 2012).
The association of periodontitis with diabetes has been described as bidirectional, and there is substantial evidence that poor glycemic control in type 1 and type 2 DM patients is a risk of periodontitis, resulting in increased extension and severity of periodontitis. Due to the global increase in the prevalence of diabetes, the influence of diabetes on the development of periodontitis may be a growing problem. Current evidence also suggests that periodontitis may aggravate the course of DM, but further longitudinal studies are warranted for a firm conclusion to be drawn. The mechanisms by which the two diseases interact are uncertain, but presumably chronic low-grade inflammation enhanced by both diseases plays an important part in the interaction, which obviously involves inflammatory cells and their products, including cytokines and MMPs. The formation of AGE results in modified cellular functions. The existing clinical trials indicate a positive effect on metabolic control and systemic inflammation of nonsurgical periodontal treatment, which may result in a clinically relevant decrease of HbA1c. However, further studies are needed to robustly confirm this.
Al-Khabbaz AK. Type 2 diabetes mellitus and periodontal disease severity. Oral Health Prev Dent. 2014;12:77–82.PubMed
Andersen CCP, Buschard K, Flyvbjerg A, Stoltze K, Holmstrup P. Periodontitis deteriorates metabolic control in type 2 diabetic goto-kakizaki rats. J Periodontol. 2006;77:350–6.CrossRef
Andersen CCP, Flyvbjerg A, Buschard K, Holmstrup P. Periodontitis is associated with aggravation of prediabetes in zucker fatty rats. J Periodontol. 2007a;78:559–65.CrossRef
Andersen CCP, Flyvbjerg A, Buschard K, Holmstrup P. Relationship between periodontitis and diabetes: Lessons from rodent studies. J Periodontol. 2007b;78:1264–75.CrossRef
Andersen CCP, Holmstrup P, Buschard K, Flyvbjerg A. Renal alterations in prediabetic rats with periodontitis. J Periodontol. 2008;79:684–90.CrossRef
Anonymous. From the centers for disease control and prevention: blindness caused by diabetes – Massachusetts, 1987–1994. JAMA. 1996;276:1865–1866.
Artese HP, Foz AM, Rabelo Mde S, Gomes GH, Orlandi M, Suvan J, D’Aiuto F, Romito GA. Periodontal therapy and systemic inflammation in type 2 diabetes mellitus: a meta-analysis. PLoS One. 2015;10, e0128344. doi:10.1371/journal.pone.0128344. eCollection 2015.PubMedCentralCrossRef