A pilot study on glycemia and insulin resistance in patients with severe periodontitis

Abstract

Context:

The potential impact of severe periodontitis on glycemia in systemically healthy individuals is not clearly established. It was hypothesized that among individuals who were previously undiagnosed for diabetes mellitus, patients with severe periodontitis have impaired glycemia and insulin resistance.

Aims:

The aim of our study was to assess and compare glycemia in severe periodontitis patients and in individuals with clinically healthy periodontium.

Materials and Methods

A cross-sectional analytical design was employed. From among individuals who were undiagnosed for diabetes mellitus, 37 patients with severe periodontitis and 37 individuals with healthy periodontium in the age group of 25–55 years were recruited for the study. The fasting blood sugar (FBS), glycosylated hemoglobin (HbA1c), and insulin resistance by the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) were assessed and compared between the two groups.

Results:

The mean FBS, HOMA-IR, and HbA1c were significantly higher for patients with severe periodontitis than those individuals with healthy periodontium. After adjustments for age, gender, and body mass index, patients with severe periodontitis had a statistically significant association with impaired glucose metabolism (HbA1c ≥5.7) (adjusted odds ratio [OR] of 9.56; 95% confidence interval [CI]: 1.819–46.08; P < 0.01). Furthermore, patients with severe periodontitis had significantly greater odds to develop impaired fasting glucose (adjusted OR of 7.489, 95% CI: 1.408–39.839; P < 0.01).

Conclusions:

The mean FBS, HbA1c, and HOMA-IR were significantly higher in severe periodontitis patients than in the control group. A higher proportion of patients presented with prediabetes, incident diabetes, and insulin resistance in the severe periodontitis group.

INTRODUCTION

More than 50% of adult population worldwide have periodontitis, and alarmingly, approximately 11.2% of them have severe periodontitis.[1,2] Periodontitis and diabetes mellitus have a “two-way street” relationship.[3] Diabetes mellitus is underlined by elevated blood glucose levels consequential to impaired insulin secretion or insulin resistance or both.[4] The global prevalence of diabetes varies from 4.6% in France to 8.3% in the United States and 8%–12% in the urban areas of India.[5,6] India is rightfully called the diabetic capital of the world as the nation houses more than 40 million diabetic patients. Another area of grave concern in this regard is that a large percentage of individuals remain undiagnosed for diabetes mellitus.[4,5] Diabetes and its complications such as retinopathy, nephropathy, neuropathy, macrovascular diseases, and altered wound healing are of great public health concern as they lead to considerable morbidity and mortality. Diabetes screening for prediabetes and incident diabetes among high-risk groups will help reduce this global health burden.

Periodontitis and diabetes mellitus have complex bidirectional associations.[3,7,8] The two-way relationship between the two chronic diseases is mediated by a complex and intricately orchestrated interplay between environmental, microbiologic, genetic, and immunologic factors.[8] Although periodontitis was understood as the sixth complication of diabetes mellitus, currently the two diseases are considered as “comorbid old friends.”[8,9,10] Progressive deterioration of glycemic control and onset of more severe diabetic complications have been reported in individuals in whom periodontitis coexists with diabetes mellitus.[3] Recent evidence has shed light into the potential impact of periodontitis on glycemia in the nondiabetic/undiagnosed individuals.[11,12] Prospective cohort studies have reported significant and positive associations between baseline periodontitis and incident diabetes mellitus.[13,14,15,16] The Joint International Workshop of the European Federation of Periodontology and the American Academy of Periodontology (AAP) stated that patients with moderate-to-severe periodontitis are at an increased risk for diabetogenesis.[17] A recommendation of the recent joint statement by the Indian Society of Periodontology (ISP) and Research Society for the Study of Diabetes in India (RSSDI) stated that periodontitis can cause insulin resistance and alterations in glycemic status.[18]

An association between impaired glucose tolerance and periodontitis has been previously reported.[19] Studies have also revealed an association between impaired fasting glucose (IFG) and periodontitis.[20,21,22,23] Few studies have reported higher glycosylated hemoglobin (HbA1c) levels in periodontitis patients than in individuals with healthy periodontium.[24,25,26,27] Further, studies have also showed an association between severe periodontitis, insulin resistance, and a dysmetabolic state.[28,29,30] We hypothesized that severe periodontitis patients who were previously undiagnosed for diabetes mellitus had alterations in glycemia when compared to individuals with healthy periodontium. The aim of our study was to assess and compare glycemia in severe periodontitis patients and in individuals with healthy periodontium. Our objective was to evaluate and compare fasting blood sugar (FBS), HbA1c, and insulin resistance by the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) in severe periodontitis patients and in individuals with clinically healthy periodontium.

MATERIALS AND METHODS

A cross-sectional analytical study was conducted. Individuals in the age group of 25–55 years who sought treatment at a dental care center and were previously undiagnosed for diabetes mellitus constituted the sampling frame. We excluded smokers, obese individuals[31] (body mass index [BMI] of ≥30 kg/m²), and pregnant/lactating women from our study. Individuals who reported having had received antibiotic therapy for the past 3 months or had a history of large recent blood loss were also excluded. A consecutive series of 37 patients who presented with severe periodontitis and with more than 20 natural teeth constituted the cases and 37 individuals with healthy periodontium formed the comparison group. Study participants were recruited during a period of 1 year, from June 2018 to June 2019. All participants were informed about the study, and they signed written consent. The study was approved by the Institutional Review Board (IRB Study Ref. No. 10/2017).

Periodontal examination

Full-mouth periodontal screening and recording was done for all study participants. For each tooth, the mesiobuccal, midbuccal, distobuccal, distolingual, mesiolingual, and midpalatal/lingual sites were carefully probed using a Williams graduated periodontal probe. Probing pocket depth (PPD) and clinical attachment levels were charted for each participant. Periodontal examinations were carried out by a single trained periodontist. As per the criteria proposed by the Centers for Disease Control and Prevention and AAP, case definition for severe periodontitis was ≥2 interproximal sites with attachment loss ≥6 mm (not on the same tooth) and ≥1 interproximal site with probing depth (PD) ≥5 mm.[32]

Glycemic status assessment

Seven milliliters of blood was collected from all consenting participants after an overnight fast of at least 8 h. FBS, fasting insulin (FI), and HbA1c levels were estimated. The Homeostasis Model Assessment Index-HOMA-IR was used to measure insulin resistance. HOMA-IR was calculated using values of FBS and FI. As per the American Diabetes Association-Standards of Medical Care in Diabetes 2020,[33] IFG was defined as a FBS ≥100 mg/dL but <126 mg/dL. Diabetes was defined as FBS ≥126 mg/dL. A HbA1c value of 5.7%–6.4% was considered as prediabetic and HbA1c ≥6.5% as diabetic.[18] The cutoff identified as diagnostic for prediabetes in HbA1c test was 5.7%.[33] HOMA-IR was calculated as fasting insulin in ulU/mL × fasting blood glucose in mg/dL ÷ 405.[34] HOMA-IR above 1.9 indicates early insulin resistance, and a cutoff value of ≥2.5 was chosen to identify insulin resistance.[35]

Statistical methods

The mean values of age, BMI, FBS, FI, HOMA-IR, and HbA1c were compared between patients with severe periodontitis and controls using unpaired t-test. The proportions of gender were compared with Chi-square test. Multivariate models were constructed using binary logistic regression using impaired glucose metabolism (HbA1c ≥5.7%), IFG (FBS >100 mg%), and insulin resistance (IR >2.5) as outcome variables. Age, gender, BMI (normal/high), and periodontal status were the independent variables included in the model. All variables were simultaneously entered in the model using enter method. The goodness of fit of the model was estimated using the Hosmer–Lemeshow statistic. The Wald statistic, P values, and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were determined for all independent variables. All analyses were performed using a statistical software program (SPSS 25, IBM SPSS Inc., Chicago, IL, USA). P < 0.05 was considered statistically significant for all tests.

RESULTS

The mean age, BMI, and proportions of male participants in the periodontitis group and the control group were comparable with no statistically significant differences [Table 1]. The mean FBS, HOMA-IR, and HbA1c were higher for patients with periodontitis than controls, and these differences were statistically significant, though these variables were within the normative clinical range. The participants were categorized based on impaired glucose metabolism (yes/no) using HbA1C values ≥5.7 [Table 2]. Participants were also categorized as having IFG and insulin resistance, as shown in Tables 3 and 4, respectively. Separate multivariate logistic regression models were constructed as HbA1c category, FBS category, and HOMA-IR category as the outcome variable, as shown in Tables 2-4, respectively. After adjustments for age, gender, and BMI, patients with severe periodontitis had a statistically significant association with impaired glucose metabolism [adjusted OR of 9.56; 95% CI: 1.819–46.08; P < 0.01, Table 2]. Furthermore, patients with severe periodontitis had significantly greater odds to develop IFG [adjusted OR of 7.489, 95% CI: 1.408–39.839; P < 0.01, Table 3]. However, the association between severe periodontitis and HOMA-IR was not statistically significant in the multivariate model [Table 4]. In this model, only male gender showed a significant association with insulin resistance. The goodness of fit for all the three multivariate models was satisfactory as Hosmer–Lemeshow statistic had P > 0.05.

Table 1.

Comparison of select characteristics between study groups

Characteristics	Periodontitis (n=37)	Controls (n=37)	P
Age (mean±SD)	43.35±7.70	40.70±6.19	0.107
BMI (mean±SD)	25.05±4.53	25.60±3.61	0.564
FBS (mean±SD)	96.81±23.59	83.27±8.28	0.002
HOMA-IR (mean±SD)	2.06±1.16	1.54±0.91	0.035
HbA1c (mean±SD)	5.87±1.46	5.29±0.20	0.021
Males, n (%)	20 (54.1)	13 (35.1)	0.102

P<0.05 statistically significant. BMI – Body mass index; FBS – Fasting blood sugar; HOMA-IR – Homeostatic Model Assessment of Insulin Resistance; HbA1c – Glycosylated hemoglobin; SD – Standard deviation, n – Number; P – Probability

Table 2.

Multivariate logistic regression model with impaired glucose metabolism (glycosylated hemoglobin ≥5.7%) as outcome variable

Independent variable	Wald	P	Adjusted OR	95% CI of adjusted OR
				Lower	Upper
Severe periodontitis	7.212	0.007	9.156	1.819	46.086
Age	0.027	0.870	0.993	0.914	1.079
Gender	0.559	0.455	0.621	0.178	2.165
BMI	0.098	0.754	0.820	0.237	2.835
Constant	1.397	0.237	0.109

P<0.05 statistically significant, BMI – Body mass index; CI – Confidence interval; OR – Odds ratio; P – Probability

Table 3.

Multivariate logistic regression model with impaired fasting glucose (fasting blood sugar >100 mg%) as outcome variable

Independent variable	Wald	P	Adjusted OR	95% CI of adjusted OR
				Lower	Upper
Severe periodontitis	5.574	0.018	7.489	1.408	39.839
Age	2.741	0.098	0.925	0.844	1.014
Gender	0.267	0.605	0.700	0.182	2.701
BMI	0.084	0.773	0.995	0.960	1.030
Constant	0.073	0.787	1.783

P<0.05 statistically significant. BMI – Body mass index; CI – Confidence interval; OR – Odds ratio; P – Probability

Table 4.

Multivariate logistic regression model with insulin resistance (Homeostatic Model Assessment of Insulin Resistance >2.5) as outcome variable

Independent variable	Wald	P	Adjusted OR	95% CI of adjusted OR
				Lower	Upper
Severe periodontitis	2.362	0.124	2.342	0.791	6.935
Age	1.857	0.173	1.058	0.844	1.014
Males	8.519	0.004	5.084	1.706	15.151
BMI	0.037	0.847	0.998	0.974	1.022
Constant	4.349	0.037	0.017

P<0.05 statistically significant. BMI – Body mass index; CI – Confidence interval; OR – Odds ratio; P – Probability

DISCUSSION

In this study, glycemia and insulin resistance were assessed and compared between a group of severe periodontitis patients and individuals with healthy periodontium. FBS and HbA1c measures helped to determine glycemia. Values of FBS and FI were used to calculate HOMA-IR. The mean FBS, HOMA-IR, and HbA1c were significantly higher for patients with severe periodontitis than controls. A higher proportion of patients presented with IFG, prediabetes, and incident diabetes in the severe periodontitis group than in the control group. We have taken only severe periodontitis cases in our test group as the “spill” of inflammatory mediators and the potential systemic impact of periodontitis on glycemia would be maximum in these cases.

The mean FBS was significantly higher in the test group with severe periodontitis than in the control group with healthy periodontium. In the severe periodontitis group, six patients presented with prediabetes (16%) and four patients (11%) presented with incident diabetes as per the results of FBS. Results of fasting blood glucose in our study indicate that the proportion of patients with impaired glycemia is more in the severe periodontitis group (27%) than in the control group (5%). Multivariate analysis showed statistically significant associations between impaired glycemia (as per FBS) and severe periodontitis after adjusting for age, gender, and BMI. Our results are in concordance with previous reports where there were proportionately more participants with IFG among patients with periodontitis. For example, Islam et al.[22] from their study in a representative sample drawn from Korean population showed a significantly higher prevalence of IFG (28.5% vs. 17.7%, P < 0.001) in the periodontitis group. They identified periodontitis as a risk factor for IFG. In another cross-sectional study, IFG was significantly associated with severe periodontitis in nonsmoking patients.[23] Our results for the IFG outcome also agree with findings of Choi et al.[20] who reported that among 12,000 participants in NHANES III, clinical attachment loss (CAL) and PPD were related to increase in odds of IFG. They reported that participants in the higher quintile categories of CAL and PPD had significantly higher odds of IFG (OR = 1.55, 95% CI: 1.16–2.07; OR = 1.39, 95% CI: 1.00–1.92; respectively).

Contrary to our results, Arora et al.[19] found that severe periodontitis was not significantly associated with IFG (OR = 1.93; 95% CI: 1.18–3.17) but was associated with impaired glucose tolerance.

Glucose binds to hemoglobin through an irreversible nonenzymatic reaction to form HbA1c. HbA1c is the major subconstituent of HbA. Being based on the average life span of a red blood cell, HbA1c levels display glycemic status over the previous 1–3 months.[24] In this study, the mean HbA1c in the severe periodontitis group is significantly higher in the test group than in the control group. Our results are in concordance with previous studies that reported higher mean HbA1c levels in periodontitis patients than in individuals without periodontitis.[24,25,26,27] A cutoff value of HbA1c ≥5.7% for prediabetes was used in our study. Interestingly, the mean values for HbA1c fell in the prediabetic range for the severe periodontitis group. Nineteen percent (seven) of the patients in the severe periodontitis group were prediabetic and 16% (six) presented with incident diabetes. In the group with healthy periodontium, two patients had prediabetes and none of them had incident diabetes. Results of HbA1c assay in our study point out that a proportion of patients with prediabetes and incident diabetes are more in the severe periodontitis group than in the control group. Our results agree with reports by Teeuw et al.[27] and Holm et al.[36] A FBS result would provide only a “snapshot” of glycemia in an individual, and recent evidence has suggested that a combined measurement of FBS and HbA1c may be recommendable for diabetes screening.[18,37]

A number of studies have investigated the periodontitis and insulin resistance association but have generated inconclusive results.[22,28,29,35,38,39,40] The mean HOMA-IR value in our study is significantly higher in the severe periodontitis group than in the control group. HOMA-IR value was >1.9 (suggestive of early insulin resistance) in 22 patients in the severe periodontitis group (59%), and seven of them had values >2.5 (22%). In the group with healthy periodontium, eight individuals had values above 1.9 and four had values above 2.5. The cutoff point for insulin resistance was proposed to be 2.5 for HOMA-IR in adults, and previous studies have employed this cutoff value.[35] However, the multivariate analysis with HOMA-IR as the outcome did not show any association with severe periodontitis.

The high prevalence of prediabetes, incident diabetes, and insulin resistance among severe periodontitis patients who were previously undiagnosed for diabetes can be attributed to a state of heightened systemic inflammation in severe periodontitis. Chronic inflammation could be the plausible biological mechanism linking infections and insulin resistance.[8] Evidence from animal models has shown that pro-inflammatory cytokines, such as tumor necrosis factor-α (TNF-α), can induce a state of insulin resistance. TNF-α may disrupt serine phosphorylation of insulin receptor substrate-1,[41] and experimental evidence has repeatedly shown that systemic inflammation is a common denominator for both insulin resistance[42] and T2 diabetes mellitus.[43] Retrospective evidence points out that the association between high blood glucose and periodontitis is often confounded by factors such as smoking, age, and gender.[44] In order to eliminate the confounding effect of smoking and age, the study population was confined to nonsmoking persons, 25–55 years old. Although our study is a pilot investigation, it has a few strengths. Most of the available evidences indicating an association between periodontitis and altered glycemia have been from representative samples drawn from nationwide surveys wherein there could be considerable heterogeneity. Our study population included a consecutive series of consenting patients who satisfied strict inclusion and exclusion criteria and was more homogeneous with respect to age, smoking status, body weight, and diabetic status. In our study, full-mouth periodontal assessment of the study population was done by a single trained periodontist minimizing measurement bias. To the best of our knowledge, no previous study has independently assessed IFG HbA1c and HOMA-IR among severe periodontitis patients and healthy controls.

The study was limited by its small sample size; another limitation was the cross-sectional study design. A cause–effect relationship of altered glycemia and/or insulin resistance to severe periodontitis could not be established by this design. However, our study highlights the importance of prediabetes and diabetes screening among severe periodontitis patients in agreement with the recommendations of Tonetti et al.[45] who proposed that screening for diabetes mellitus should be part of a standard periodontal examination. Clinical signs of exaggerated periodontal destruction should be considered a “red flag” for underlying diabetes mellitus.[18] The ISP-RSSDI good clinical practice guidelines on management of periodontal disease in patients with diabetes mellitus recommend screening for prediabetes among high-risk patients reporting to dental treatment facilities, a patient-centered collaborative treatment approach, and cross-referrals between diabetologists and periodontists.[18]

CONCLUSIONS

This study investigated glycemia and insulin resistance in severe periodontitis patients and healthy controls. The mean values for fasting blood glucose, HbA1c, and HOMA-IR were significantly higher in the severe periodontitis group than in the control group with a healthy periodontium. A higher proportion of individuals presented with prediabetes and incident diabetes among the severe periodontitis patients. This pilot investigation underlines the importance of prediabetes/diabetes screening among periodontitis patients.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.