Periodontal inflammation correlates with systemic inflammation and insulin resistance in patients with recent diagnosis of type 2 diabetes

Background: diabetes and periodontitis are common comorbidities; however, the clinical implications of this association remain only partially known. This study was aimed to characterize the periodontal status of type 2 diabetic (T2D) patients and its correlation with metabolic and inflammatory parameters. Methods: patients (n = 30) with 5 or less years since the diagnosis of T2D (18 – 65 years old) were recruited. Anthropometric (Body Mass Index, BMI), metabolic (fasting glucose, glycated hemoglobin, insulin, HOMA-IR, HDL, LDL and total cholesterol, triglycerides) and inflammatory parameters (ultrasensitive C reactive protein, usCRP) were quantified. Periodontal evaluation included clinical attachment level (CAL), probing depth (PD), gingival level (GL) and bleeding on probing (BOP) average. Statistical significance was assessed by Mann-Whitney and Spearman correlation tests. Results: mean values of BOP, CAL, PD and GL were 39.3, 2.8, 2.8, and 0.1, respectively. BOP significantly correlated with BMI and HOMA-IR and was higher in patients with elevated usCRP >3 mg/L (p<0.05). Age and duration of T2D directly and inversely correlated with CAL and GL, respectively. BOP correlated with HOMA-IR and usCRP but not with patients ́age, duration of T2D or BMI. Conclusions: in patients with recent diagnosis of T2D, BOP is associated with usCRP and HOMA-IR levels, suggesting that periodontal inflammation promotes insulin resistance possibly by increasing systemic inflammation.

Periodontitis is also a prevalent condition (Schatzle et al., 2003;Hajishengallis, 2015) that is associated with decreased quality of life (Gamonal et al., 2010).TD2 is a recognized risk factor for periodontitis across different populations ( Nelson et al., 1990;Genco & Borgnakke, 2013).Furthermore, the prevalence of periodontitis in diabetic patients is extremely high, such as it has been considered a diabetic complication by itself (Löe, 1993;Preshaw et al., 2012).
Pathophysiologically, periodontitis and diabetes may influence each other in bidirectional ways (Mealey, 2006;Preshaw et al., 2012).Diabetic complications are the result of micro and macrovascular abnormalities.In addition, chronic hyperglycemia impairs immunological function and tissue reparation, leading to chronic tissue inflammation and damage.These might be the underlying mechanisms for accelerated periodontal destruction in diabetic patients and animal models ( Pickup, 2004;Lakschevitz et al., 2011).By other hand, there is still no agreement in scientific community that treatment of periodontitis improves the glycemic control in this type of patients.A randomized study of 512 patients with DM2 from five medical centers in New York City identified that there is no relationship between the decrease in blood glucose levels among diabetic patients receiving periodontal treatment for a period of six months versus diabetic controls who did not receive it (Engebretson et al., 2013).While other authors recognize that the treatment has a positive impact on the improvement of blood sugar levels ( Borgnakke et al., 2013;Engebretson & Kocher, 2013;Casanova et al., 2014).A last meta-analysis shows that there was no evidence to support that one periodontal therapy was more effective than another in improving glycaemic control in people with diabetes mellitus (Simpson et al., 2015).
Systemic low grade inflammation is a common finding in T2D and it has been associated with increased insulin resistance (Colombo et al., 2012).Circulating levels of ultrasensitive C reactive protein (usCRP), a marker of subclinical inflammation, are increased in patients with periodontal disease, possibly as a consequence
Obesity is a major risk factor for T2D and appears to be an independent determinant of periodontitis (Saito et al., 2001;Chaffee & Weston, 2010).Recently, was reported that body mass index (BMI) and waist circumference (WC) to height ratio are significantly associated with periodontal progression (Gorman et al., 2012).
Finally, although the duration of T2D determines the risk for micro and macrovascular complications its influence on the prevalence and severity of destructive periodontal disease remains controversial (Cerda et al., 1994;Sandberg et al., 2000;Pranckeviciene et al., 2014).
The aims of this study were to quantify clinical parameters of periodontal inflammation and destruction in patients with recent diagnosis of T2D and to determine the relationship between periodontitis and the duration of diabetes as well as indicators of insulin resistance, systemic inflammation and nutritional status.

Periodontal evaluation
Periodontal determinations were performed by a single and trained periodontist that was blind for the metabolic and inflammatory status of the patients.Periodontal disease diagnosis and severity classification were based on American Academy of Periodontology guidelines (American Academy of Periodontology, 2000).
The periodontal parameters assessed were probing depth (PD), gingival level (GL) measured from cementoenamel junction, clinical attachment level (CAL) and bleeding on probing (BOP).

Anthropometric determinations
Body weight (BW), height, waist circumference (WC) and body mass index (BMI) were determined by a single investigator.

Laboratory determinations
Fasting glucose, insulin, glycated hemoglobin fraction C (HbA1c), insulin resistance (HOMA_IR), total cholesterol (TC), low density lipoprotein cholesterol (C-LDL), high density lipoprotein cholesterol (C-HDL), triglycerides (TG) and usCRP were determined in the central laboratory of Pontifical Catholic University of Chile Health Network.

Statistical analysis
Statistical significance of associations was determined with Mann-Whitney and Spearman correlation test using SPSS version 22.0 (IBM Corp. Released 2013) and GraphPad Prism version 7.0 (GraphPad Software, s.f.).

Results
Thirty patients were included in this study (21 women, 9 men).
The age of the subjects was 48 ± 10.4 years (mean ± S.D.) and the known duration of T2D was 2.1 ± 1.3 (mean ± S.D.).Anthropometric, metabolic, periodontal and dental characteristics of the subjects are shown in table 1.All the patients included in this study had periodontitis and ~15% had the most severe form of the disease (mean CAL 5-6 mm) (table 2).
The age of patients directly correlated with clinical attachment level (CAL) (figure 1A) and inversely with gingival level (GL) (figure 1B).Equivalent associations were found with the duration of T2D (figure 1C-D).By contrast, patients age or duration of T2D did not correlate with BOP, PD or metabolic parameters (table 3).
Obesity has been previously associated with periodontal disease (Saito et al., 2001) however, the impact of diabetes on periodontitis in obese patients remains uncertain.In our group of T2D patients, we found no association between BMI or waist circumference (WC) with periodontal indicators GL, CAL and BOP (table 3).
Combined, these data suggest that, in recently diagnosed patients with T2D, both age and duration of diabetes are determinants of gingival recession, indicated by lower GL, and interproximal loss of attachment indicated by elevated CAL.
The association between gingival inflammation and circulating levels of usCRP is shown in figure 2A.We found a direct correlation between usCRP and BOP levels in diabetic patients (r= 0.515; p<0.004).Concordantly, the median BOP was significantly higher in patients with usCRP ≥ 3 mg/L (figure 2B).Since periodontitis is a site-specific disease, with an heterogeneous intrapersonal distribution of its severity, we clustered the number of periodontitis sites at the individual level.For this purpose, only the sites with probing depth ≥4 mm was analyzed.We found no correlation between the proportion of periodontal pockets larger than 4 mm and usCRP or HOMA-IR (figure 3).

Discussion
Herein we report that periodontitis was present in all the middle-aged T2D patients studied and that 15% of them had severe periodontal inflammation and destruction.These findings agree with previous reports showing that periodontitis is significantly more frequent in diabetic patients than non-diabetic individuals of same age and sex ( Tsai et al., 2002;Lakschevitz et al., 2011;Colombo et al., 2012) Nonetheless, our main finding was that in patients with recent diagnosis of T2D (5 or less years), AL and GL positively correlate with the duration of diabetes.Cerda et al. (1994), reported a positive correlation between the time since diagnosis and the size of periodontal pockets and gingival recession, but in patients with duration of diabetes longer than 5 years.Therefore, our results suggest that T2D has a rapid deleterious influence on periodontal tissue which is, fastest that in other classically targeted tissues.
Oyarzo et al.Obesity has been recently proposed as an independent risk factor for periodontitis (Gorman et al., 2012;Papageorgiou et al., 2015).
Herein, we found that BOP, a clinical indicator of local periodontal inflammation, was directly correlated with BMI in patients with T2D.
Previously, Saito et al. (2001) reported that probing depth and BMI are associated in a population of young Japanese patients, whereas Gorman et al. (2012) found that white obese male individuals have increased risk for moderate and severe periodontal disease.We did not observe correlations between CAL or GL with BMI or WC in our group of T2D patients.
Another important finding of our study was the direct correlation between BOP and usCRP.It is currently accepted that periodontitis determines low grade systemic inflammation (Gurav, 2012;Hajishengallis, 2015) and this has been implicated in the progression of chronic metabolic conditions like atherosclerotic coronary artery disease (D'Aiuto et al., 2004).The mechanisms of periodontitis-associated systemic inflammation are partially understood but is possible that gingival ulceration at periodontal pockets facilitates the translocation of bacteria into the systemic circulation.In addition, locally generated pro-inflammatory cytokines, such as tumor necrosis factor, interleukin-1β and Iinterleukin-6, enter the circulation and promote systemic inflammatory responses (Hajishengallis, 2015).
Currently, it is proposed that both metabolic (lipotoxicity, mitochondrial dysfunction) and inflammatory determinants play pathogenic roles in insulin resistance (Samuel & Shulman, 2016).In non-diabetic subjects, HOMA-IR has been correlated with periodontitis severity Pejcic, suggesting that local periodontal inflammation influence systemic insulin sensitivity (Demmer et al., 2012).A pathophysiological link for this association might be systemic inflammation, as indicated by elevated inflammatory cytokine in patients with periodontitis (Loos, 2005).Conventional periodontal therapy has beneficial impacts on insulin resistance and systemic inflammation in T2D patients ( Sun et al., 2011;Sgolastra et al., 2013).In our study, we found that BOP directly correlated with HOMA-IR, independent of the number or percentage of sites ≥ 4mm, suggesting that periodontitis-associated insulin resistance is no related with the extension or severity of periodontal disease.Further studies are required to disclose the biological relationships between periodontitis and insulin resistance.
It has been proposed that the metabolic control of diabetes positively influence periodontitis progression (Chapple & Genco, 2013).In our study, we found no correlations between various periodontitis parameters and HbA1C levels, an indicator of glycemic control, possibly because the limited value of isolated HbA1c determinations to estimate the long-term glycemic control in diabetic patients.

Conclusion
The main limitation of this study was the small number of patients studied and lack of a control group.In spite of that, we were able to find that in recently diagnosed patients with T2D, age and time since T2D diagnosis correlate with GL and CAL, and that HOMA-IR and usCRP are directly correlated with gingival inflammation (BOP).
T2D patients were recruited from Pontifical Catholic University of Chile outpatient diabetes and odontology clinics.Inclusion criteria were: T2D with five years or less of formal diagnosis, age between 18 and 65 years and preservation of at least 20 teeth.Exclusion criteria were: use of antibiotics or periodontal treatment in the last six-month, concurrent acute infectious diseases, primary or secondary immunodeficiency unrelated to diabetes, active smoking and severe oral infection.Thirty-nine (39) patients were initially screened but only 9 men and 21 women completed the full series of clinical, anthropometric and laboratory determinations.Study design and informed consent document were approved by Pontifical Catholic University of Chile institutional ethical committee for clinical research.All participants signed informed consent.

Figure 1 :
Figure 1: Correlation between periodontal parameters (CAL, GL) with age and time of T2D. A. Correlation between CAL (mm) and age (years) B. Correlation between GL (mm) and age (years) C. Correlation between CAL (mm) and time of T2D (years) D: Correlation between GL (mm) and time of T2D (years) * p<0.05, **p < 0.005 (Spearman)

Figure 3 :
Figure 3: A. Correlation between percentage of sites equal or larger than 4 mm with us CRP and HOMA-IR B. Correlation between percentage (%) of sites ≥ 4mm of each patient and us CRP (p = 0.109, Spearman) C. Correlation between percentage (%) of sites ≥ 4mm of each patient and HOMA-OR.Not significant (p = 0.901, Spearman)

Table 1 :
Anthropometric, metabolic and clinical periodontal characteristics of subjects included in the study.