Abstract
Japan Diabetes Complication and Prevention prospective (JDCP) study was conducted to examine the association between glycemic control and oral conditions in a large database of Japanese patients with diabetes. It included a total of 6099 patients with diabetes (range, 40–75 years) who had been treated as outpatients between 2007 and 2009. The mean number of present teeth at baseline was 19.8 and women with type 2 diabetes had fewer teeth than men with type 2 diabetes. Within the previous year, 17% of all patients had lost teeth. At baseline, 32% had experienced gingival swelling, 69% had brushed more than twice a day, 37% had used interdental cleaning aids, and 43% had undergone regular dental checkups. Multiple logistic regression analysis indicated that type 1 patients with HbA1c ≥ 7.0% were at higher risk of having fewer than 20 teeth (odds ratio [OR] 2.38; 95% confidence interval [CI] 1.25–4.78), and type 2 patients with HbA1c ≥ 8.0% also were at high risk of having fewer than 20 teeth (OR 1.16; 95% CI 1.00–1.34), after adjustment for nine possible confounding factors. In conclusion, patients with diabetes were found to be at high risk of tooth loss, and the poorer the glycemic control, the higher the risk of tooth loss in these patients.
Keywords: Japan diabetes complication and their prevention study, Periodontal disease, Complication
Introduction
In a super-aging society, it is becoming increasingly recognized not only by dental circles, but by the general public how critically important it is to ensure that older individuals continue to have no difficulty in chewing or crushing foods with their own teeth, as well as in engaging in pleasant conversations with as many around them as possible. Again, while the maintenance of their overall health is shown to be closely linked to the health of their periodontal tissue, conversely, the presence of periodontal disease is shown to adversely affect their overall health [2, 3]. Therefore, it is deemed crucial to maintain and promote oral health in older individuals not merely to maintain oral health (thus the ability to masticate), but to maintain overall health (thus the ability to live a fuller and more comfortable life through enjoyment of meals and conversations) [3].
Periodontal disease is a disease in which the periodontium (tooth-supporting tissue; i.e., gum, periodontal membrane, alveolar bone, and cementum) becomes progressively disrupted in the presence of persistent chronic inflammation caused by dental biofilms and whose onset and development are accounted for in large part not only by host susceptibility to the disease, but by genetic and environmental factors [3]. Again, periodontal disease calls for attention as a diabetic complication, as does diabetes as a risk factor for periodontal disease [3–6].
Given that, currently, the relationship between periodontal disease and other diabetic complications remains only insufficiently elucidated in Japan, this report focuses on baseline dental and oral findings in patients with diabetes from a large-scale registry-based study, the Japan Diabetes Complication and Their Prevention prospective (JDCP) study, to investigate the current status of chronic diabetic complications (e.g., retinopathy, nephropathy, neuropathy, macroangiopathy, and periodontal disease) and their treatment in patients with diabetes, to provide clues as to their status and interrelationship.
Patients and methods
Patients
The study subjects comprised outpatients with type 1 and 2 diabetes 40 years old or older but younger than 75 years old being treated nationwide at university hospitals, core local hospitals, and clinics offering essentially specialist diabetes care. The study provisionally included a total of 7700 patients who met the inclusion criteria and gave informed consent during the 30 months from June 2007 to November 2009 and finally included a total of 6338 patients (patients with type 1 diabetes, 394 [6.2%]; patients with type 2 diabetes, 5944 [93.8]) after excluding those who had met either of the exclusion criteria: (1) patients incapable of making regular visits to their outpatient diabetes clinics; (2) those shown to have progressed to diabetic proliferative retinopathy; (3) those on dialysis therapy; (4) those diagnosed with malignancy in the last 5 years; and (5) those the study investigators judged to be ineligible for the study.
Methods
Diet, physical activity, and periodontal disease were evaluated in a subset of the subjects by using the brief-type self-administered diet history questionnaire (BDHQ), the International Physical Activity Questionnaire (IPAQ), and the oral examination reports/panorama x-ray images, respectively. Event-related outcome measures included: nephropathy (defined as presence of albuminuria or overt proteinuria confirmed on two consecutive measurements; a more than twofold increase in serum creatinine levels; or initiation of dialysis); retinopathy (defined as progression from simple/pre-proliferative to proliferative retinopathy; loss of vision; or photocoagulation/vitreous surgery); neuropathy (lower extremity amputations [LEA]; or ulcer/gangrene development); macroangiopathy (myocardial infarction; angina pectoris; heart failure; arteriosclerosis obliterans [ASO]; LEA; stroke; transient ischemic attack (TIA); or revascularization); malignancy; progression of periodontal disease or tooth loss; and death. The duration of diabetes was determined for each patient, based on his/her date of disease onset as estimated by his/her physician in charge in the available questionnaire.
Basic patient information was available for use as a basis for this study, which included information on periodontal disease under six headings (history of tooth loss in the last year; self-reported number of present teeth, history of gingival swelling, number of oral cleaning performed daily, frequency of use of oral cleansing aids, and frequency of regular dental check ups). X-ray examinations were performed by dentists in all patients who gave informed consent.
Statistical analysis
Continuous and categorical variables were expressed as means ± standard deviations or medians (interquartile ranges) and as proportions, respectively, with the former compared using the Student’s t test or the Mann–Whitney test, and the latter using the Pearson’s chi-square test.
Further, in an attempt to explore the relationship between glycemic control and the number of present teeth in the study participants, multiple logistic regression analyses were also performed, using sex (male/female), age (< 60 years/ ≥ 60 years), history of smoking (yes/no), duration of diabetes (< 10 years/ ≥ 10 years), body mass index (BMI) (< 25/ ≥ 25 kg/m2), prior medical history (presence or absence of hypertension, dyslipidemia, myocardial infarction, or cerebrovascular disorder), use of interdental cleaning aids (yes/no), regular dental checkups (yes/no), and HbA1c (< 7.0%/ ≥ 7.0%, < 8.0%/ ≥ 8.0%) as explanatory variables, as well as number of present teeth (< 20/ ≥ 20) as a dependent variable, to calculate the odds ratios (ORs) and 95% confidence intervals (CIs), and the ORs and 95% CIs adjusted for the nine explanatory variables, for the association between glycemic control and the number of present teeth. All statistical tests performed were two-tailed with a significance level of < 0.05. All data analyses were independently performed at least twice by several investigators within the study group and those without to confirm accuracy. All analyses were carried out using JMP version 9.0 software (SAS Institute Inc, Cary, NC, USA) and SPSS 22.0 (IBM Corp, Armonk, NY, USA).
Ethical considerations
This study was conducted in accordance with the Declaration of Helsinki and the Ministry of Education, Culture, Sports, Science and Technology/Ministry of Health, Labor and Welfare of Japan “Ethical Guidelines for the Conduct of Epidemiological Research” [7] to ensure that the privacy of all study participants would be protected; that the study participants would not be subjected to harm or hazard in any way whatsoever; and that full informed consent would be obtained from all study participants prior to study initiation.
Again, care was also taken to include not only in the study protocol, but in the informed consent form (ICF) a statement clarifying that all necessary ophthalmologic and dental examinations would be performed under the usual insurance scheme, so that study participation would not result in any unnecessary time and financial burden on the part of the study participants.
This study was conducted with approval from the Japan Diabetes Society Research Ethics Committee and the institutional review board (IRB) of each participating facility and registered with the UMIN Clinical Trial Registry (registry ID, UMIN 000016519).
Results
Of the 6338 patients included in the study, a total of 6099 (96.2%) patients who had responded to the questionnaire items related to oral health were included in this baseline analysis, while the oral examination reports were only made available for 723 of these patients.
Patient characteristics at baseline (Table 1)
Table 1.
Clinical characteristics and variables in JDCP subjects at baseline
| Characteristic/variable | (n = 6099) | P value | |
|---|---|---|---|
| Age, years | 61.1 ± 8.1 (40–75) | ||
| Sex, n (%) | |||
| Men | 3596 (59.0%) | ||
| Women | 2503 (41.0%) | ||
| Type of diabetes | Type 1 | Type 2 | |
| n (%) | 378 (6.2) | 5721 (93.8) | |
| Age, years | 56.2 ± 9.3 (40–74) | 61.4 ± 7.9 (40–75) | < 0.01 |
| Sex, n (%) | |||
| Men | 168 (44.4) | 3428 (59.9) | |
| Women | 210 (55.6) | 2293 (40.1) | |
| Alcohol consumption, n (%) | 377 | 5704 | |
| Yes | 99 (26.3) | 2223 (39.0) | |
| No | 278 (73.7) | 3481 (61.1) | |
| Current smoking status | 377 | 5701 | |
| Yes | 144 (38.2) | 2158 (37.9) | |
| No | 233 (61.8) | 3543 (62.1) | |
| Duration of diabetes, years (n) | 11.6 ± 9.3 (376) | 10.8 ± 8.1 (5651) | 0.124 |
| Height, cm (n) | 160.1 ± 9.1 (377) | 161.3 ± 8.9 (5718) | < 0.05 |
| Weight, kg (n) | 56.8 ± 10.0 (378) | 63.9 ± 12.2 (5659) | < 0.01 |
| Maximum weight, kg (n) | 62.4 ± 10.6 (374) | 70.9 ± 13.3 (5694) | < 0.01 |
| Age at maximum weight, years (n) | 41.9 ± 13.8 (371) | 47.2 ± 12.9 (5693) | < 0.01 |
| Waist circumference, cm (n) | 78.1 ± 9.2 (347) | 86.4 ± 10.4 (5411) | < 0.01 |
| Body mass index (BMI), kg/m2 | 22.1 ± 3.0 (377) | 24.5 ± 3.9 (5656) | < 0.01 |
| HbA1c, % (n) | 7.8 ± 1.4 (376) | 7.4 ± 1.3 (5705) | < 0.01 |
| Fasting plasma glucose, mg/dL (n) | 134.8 ± 63.7 (93) | 135.6 ± 38.1 (2282) | 0.901 |
| Fasting IRI, μU/mL (n) | 1.8 (1.1–7.6) (6) | 5.5 (3.5–8.7) (1110) | < 0.05 |
| Systolic blood pressure, mmHg (n) | 125.1 ± 15.3 (375) | 129.7 ± 15.1 (5661) | < 0.01 |
| Diastolic blood pressure, mmHg (n) | 72.2 ± 9.6 (375) | 74.7 ± 10.2 (5661) | < 0.01 |
| Lipid profile | |||
| Total cholesterol, mg/dL (n) | 198.5 ± 29.9 (350) | 195.0 ± 33.1 (5447) | < 0.05 |
| Low-density lipoprotein cholesterol, mg/dL (n) | 107.4 ± 24.3 (355) | 112.7 ± 28.1 (5491) | < 0.01 |
| High-density lipoprotein cholesterol, mg/dL (n) | 73.1 ± 18.2 (372) | 57.4 ± 16.0 (5627) | < 0.01 |
| Triglycerides, mg/dL (n) | 73.0 (56.0–92.0) (167) | 107.0 (77.0–152.0) (2755) | < 0.01 |
| Serum creatinine, mg/dL (n) | 0.7 ± 0.2 (367) | 0.8 ± 0.3 ( 5612) | < 0.01 |
| eGFR, mL/min/1.73 m2 (n) | 81.1 ± 17.5 (367) | 77.1 ± 18.9 (5612) | < 0.01 |
| Serum albumin, g/dL (n) | 4.2 ± 0.3 (339) | 4.3 ± 0.3 (5251) | < 0.01 |
Continuous variables were presented as mean ± standard deviation and tested for significance by using the Student’s t test. Non-normally distributed variables (triglycerides and fasting immunoreactive insulin) were presented as median (interquartile range) and tested for significance by using the Mann–Whitney U test. For each variable with missing data, the number of subjects with complete data is shown in parentheses
eGFR estimated glomerular filtration rate, IRI immunoreactive insulin
This study included a total of 378 (6.2%) patients with type 1 diabetes (mean age, 56.2 ± 9.3 years; duration of diabetes, 11.6 ± 9.3 years) and a total of 5721 (93.8%) patients with type 2 diabetes (mean age, 61.4 ± 7.9 years; duration of diabetes, 10.8 ± 8.1 years), and, of these, men and women accounted for 59.0% (3596) and 41.0% (2503), respectively.
Patient responses to the oral health-related questionnaire items (Table 2)
Table 2.
Dental and oral findings in JDCP subjects at baseline
| Overall (n = 6099) | Type 1 diabetes (n = 378) | Type 2 diabetes (n = 5721) | |
|---|---|---|---|
| Self-reported number of present teeth, n (patient number) | 19.8 ± 9.4 (5601) | 20.8 ± 9.3**a (343) | 19.7 ± 9.4**a (5258) |
| Poor glycemic control (HbA1c, ≥ 7.0%), n (patient number) | 19.6 ± 9.5 (3360) | 20.1 ± 9.5*b (253) | 19.5 ± 9.5 (3107) |
| Favorable glycemic control (HbA1c, < 7.0%) n (patient number) | 20.0 ± 9.3 (2224) | 22.9 ± 8.4*b (88) | 19.9 ± 9.3 (2136) |
| Edentulous jaw, n (%) | 341 (5.6) | 21 (5.6) | 320 (5.6) |
| More than 28 teeth, n (%) | 1361 (22.3) | 102 (27.0) | 1259 (22.0) |
| More than 20 teeth, n (%) | 3564 (58.4) | 241 (63.8) | 3323 (58.1) |
| Men | 20.1 ± 9.4**c (3295) | 21.5 ± 8.6 (148) | 20.1 ± 9.5**d (3147) |
| Women | 19.3 ± 9.4**c (2306) | 20.4 ± 9.8 (195) | 19.2 ± 9.4**d (2111) |
| History of tooth loss in the past one year, n (%) | 6078 | 377 | 5701 |
| No | 5056 (83.2) | 317 (84.1) | 4.739 (83.1) |
| Yes | 1022 (16.8) | 60 (15.9) | 962 (16.9) |
| History of gingival swelling, n (%) | 5933 | 369 | 5564 |
| No | 4033 (68.0) | 244 (66.1) | 3789 (68.1) |
| Yes | 1900 (32.0) | 125 (33.9) | 1775 (31.9) |
| Tooth brushing (times/day), n (%) | 6064 | 375 | 5689 |
| Not on a daily basis | 139 (2.3) | 3 (0.8) | 136 (2.4) |
| Once a day | 1754 (28.9) | 75 (20.0) | 1679 (29.5) |
| Twice a day | 2826 (46.6) | 175 (46.7) | 2651 (46.6) |
| 3 times a day | 1122 (18.5) | 102 (27.2) | 1020 (17.9) |
| 4 or more times a day | 223 (3.7) | 20 (5.3) | 203 (3.6) |
| Interdental cleaning, n (%) | 6050 | 375 | 5675 |
| None | 3807 (62.9) | 220 (58.7) | 3587 (63.2) |
| 1–3 times/month | 533 (8.8) | 41 (10.9) | 492 (8.7) |
| 1–2 times/week | 459 (7.6) | 29 (7.7) | 430 (7.6) |
| 3–4 times/week | 256 (4.2) | 19 (5.1) | 237 (4.2) |
| On a daily basis | 995 (16.5) | 66 (17.6) | 929 (16.3) |
| Regular dental checkups, n (%) | 6069 | 376 | 5693 |
| None | 3429 (56.5) | 196 (52.1) | 3233 (56.8) |
| 1–2 times/year | 1774 (29.2) | 122 (32.4) | 1652 (29.0) |
| 3–5 times/year | 365 (6.0) | 24 (6.4) | 341 (6.0) |
| More than 5 times/year | 501 (8.3) | 34 (9.1) | 467 (8.2) |
Continuous variable were presented as mean ± standard deviation unless otherwise indicated, and categorical variables were presented as percentages
The Student t-test and the χ2 test were used to test continuous and categorical variables, respectively, for significance
For each variable with missing data, the number of subjects with complete data is shown in parentheses
Significant differences are shown in bold. * P < 0.05; ** P < 0.01
aStatistically significant difference between patients with type 1 and type 2 diabetes (P < 0.01)
bStatistically significant difference between favorable and poor glycemic control in patients with type 1 diabetes (P < 0.05)
cStatistically significant difference between men and women patients (P < 0.01)
dStatistically significant difference between men and women patients with type 2 diabetes (P < 0.01)
The mean number of present teeth as reported by the study participants (n = 5601) was 19.8 ± 9.4 overall but fewer in patients with type 2 diabetes than in patients with type 1 diabetes (19.7 ± 9.4 vs. 20.8 ± 9.3; P < 0.01). Again, a comparison of patients with type 2 diabetes by sex showed that women had significantly fewer present teeth than men (19.2 ± 9.4 vs. 20.1 ± 9.5; P < 0.01). Furthermore, while a total of 341 (5.6%) patients were found to be edentulous, a total of 3564 (58.4%) patients were found to have 20 or more present teeth.
A comparison of the number of present teeth by glycemic control and type of diabetes showed that while there was no difference in the number of present teeth between type 2 diabetic patients with favorable glycemic control (HbA1c, < 7.0%) and those with poor glycemic control (HbA1c, > 7.0%), type 1 diabetic patients with favorable glycemic control had retained significantly more teeth than those with poor glycemic control (22.9 ± 8.4 vs. 20.1 ± 9.5; P < 0.05).
Of the 6078 study participants who responded to the questionnaire item on tooth loss, a total of 5056 (83.2%) reported having lost no teeth and 1022 (16.8%) reported having lost teeth in the past 1 year. Again, a total of 317 (84.1%) patients with type 1 diabetes reported having lost no teeth and 60 (15.9%) reported having lost teeth in the past 1 year, and a total of 4739 (83.1%) patients with type 2 diabetes reported having lost none and 962 (16.9%) reported having lost teeth in the past 1 year, demonstrating that more patients with type 2 diabetes had lost teeth than those with type 1 diabetes in 1 year.
Of the 5933 participants who responded to the questionnaire item on gingival swelling, a total of 4033 (68.0%) reported having had no gingival swelling and 1900 (32.0%) reported having had gingival swelling, showing that close to one-third of the participants had a history of gingival swelling. Again, a total of 125 (33.9%) patients with type 1 diabetes and 1775 (31.9%) patients with type 2 diabetes reported having had gingival swelling, with no difference between the two groups.
Of the 6064 participants who responded to the questionnaire item on the daily frequency of oral hygiene, a total of 1754 (28.9%) reported performing oral hygiene once daily, 2826 (46.6%) twice daily, 1122 (18.5%) three times daily, 223 (3.7%) four or more times daily, and 139 (2.3%) less than daily, showing that those performing oral hygiene twice daily accounted for the greatest proportion.
Of the 6050 participants responding to the questionnaire item on the frequency of use of interdental cleaning aids, a total of 3807 (62.9%) reported using them very rarely, 533 (8.8%) once to three times a month, 459 (7.6%) once or twice a week, 256 (4.2%) three to four times a week, and 995 (16.5%) five or more times a week.
Of the 6069 participants who responded to the questionnaire item on the frequency of regular dental checkups, a total of 3429 (56.5%) reported undergoing regular dental checkups very rarely, 1774 (29.2%) once or twice a year, 365 (6.0%) three to five times a year, and 501 (8.3%) five or more times a year, showing that more than 50% reported not undergoing any.
Relationship between glycemic control and the number of present teeth
As for the relationship between glycemic control and the number of present teeth in the participants, type 1 patients with HbA1 > 7.0% were found to have 2.32 (odds ratio [OR] 2.32; 95% CI 1.25–4.29) times the odds of having fewer than 20 present teeth or 2.38 times the odds of doing so (OR 2.38; 95% CI 1.25–4.78) after adjustment for the nine confounding factors, compared to those with HbA1c < 7.0% (Table 3). Again, of all explanatory variables, age (≥ 60 years) and use of interdental cleaning aids (no) were found to have 3.92 (OR 3.92; 95% CI 2.32–6.76) and 2.27 (OR 2.27; 95% CI 1.27–4.15) times the odds of resulting in fewer than 20 present teeth, respectively, compared to their opposites, with each demonstrating a significant association.
Furthermore, type 2 patients with HbA1 ≥ 8.0% were found to have 1.02 (OR 1.02; 95% CI 0.89–1.17) times the odds of having fewer than 20 present teeth or 1.16 (OR 1.16; 95% CI 1.00–1.34) times the odds of dosing so after adjustment for the nine confounding factors, compared to those with HbA1c < 8.0% (Table 4). Again, of all explanatory variables, sex (female), age (≥ 60 years), duration of diabetes (≥ 10 years), past medical history (yes), use of interdental cleaning aids (no), and regular dental checkups (no) were found to have 1.21 (OR 1.21; 95% CI 1.07–1.37), 2.69 (OR, 2.69; 95% CI 2.36–3.08), 1.24 (OR 1.24; 95% CI 1.09–1.40), 1.18 (OR 1.18; 95% CI 1.02–1.36), 2.02 (OR 2.02; 95% CI 1.76–2.32) and 1.24 (OR 1.24; 95% CI 1.09–1.41) times the odds of resulting in fewer than 20 present teeth, respectively, compared to their opposites, with each demonstrating a significant association.
Table 3.
Baseline odds ratios (OR) of the number of present teeth < 20 in type 1 diabetic subjects (n = 335) derived from multiple logistic regression analysis
| OR (95% CI) | P value | Adjusted OR** (95% CI) | P value | Adjusted OR*** (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Men | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Women | 1.14 (0.71–1.83) | 0.59 | 1.38 (0.81–2.38) | 0.23 | 1.44 (0.85–2.47) | 0.18 |
| Age (years) | ||||||
| < 60 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 60 | 4.01 (2.45–6.56) | < 0.01 | 3.92 (2.32–6.76) | < 0.01 | 3.87 (2.30–6.63) | < 0.01 |
| Smoking | ||||||
| No | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Yes | 0.95 (059–1.55) | 0.85 | 1.11 (0.65–1.89) | 0.71 | 1.11 (0.65–1.88) | 0.71 |
| Duration of diabetes (years) | ||||||
| < 10 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 10 | 1.32 (0.83–2.12) | 0.24 | 1.28 (0.76–2.17) | 0.35 | 1.33 (0.79–2.26) | 0.28 |
| Body mass index (kg/m2) | ||||||
| < 25 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 25 | 1.23 (0.63–2.40) | 0.55 | 0.92 (0.44–1.89) | 0.83 | 0.94 (0.45–1.91) | 0.87 |
| Past medical history* | ||||||
| No | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Yes | 1.87 (1.16–3.02) | 0.01 | 1.16 (0.67–2.00) | 0.60 | 1.17 (0.68–2.01) | 0.57 |
| Use of interdental cleaning aids | ||||||
| Yes | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| No | 2.29 (1.38–3.79) | < 0.01 | 2.27 (1.27–4.15) | < 0.01 | 2.31 (1.30–4.21) | < 0.01 |
| Regular dental checkups | ||||||
| Yes | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| No | 1.30 (0.81–2.09) | 0.27 | 1.02 (0.59–1.78) | 0.94 | 1.04 (0.60–1.80) | 0.90 |
| HbA1c (%) | ||||||
| < 7.0 | 1.0 (reference) | 1.0 (reference) | – | |||
| ≥ 7.0 | 2.32 (1.25–4.29) | < 0.01 | 2.38 (1.25–4.78) | 0.01 | – | |
| < 8.0 | 1.0 (reference) | – | 1.0 (reference) | |||
| ≥ 8.0 | 1.53 (0.94–2.47) | 0.08 | – | 1.55 (0.92–2.62) | 0.10 | |
*No, no history of hypertension, hyperlipidemia, myocardial infarction, and cerebrovascular disease
**Adjusted for age, sex, body mass index, time from onset of diabetes, smoking status, past medical history, interdental cleaning, regular dental checkups, and HbA1c (7%)
***Adjusted for age, sex, body mass index, time from onset of diabetes, smoking status, past medical history, interdental cleaning, regular dental checkups, and HbA1c (8%)
Significant differences are shown in bold
Table 4.
Baseline odds ratio (OR) of the number of present teeth < 20 in type 2 diabetic subjects (n = 5,065) derived from multiple logistic regression analysis
| OR (95% CI) | P value | Adjusted OR** (95% CI) | P value | Adjusted OR*** (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Men | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Women | 1.19 (1.06–1.33) | < 0.01 | 1.21 (1.07–1.37) | < 0.01 | 1.21 (1.07–1.37) | < 0.01 |
| Age (years) | ||||||
| < 60 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 60 | 2.68 (2.36–3.04) | < 0.01 | 2.66 (2.33–3.05) | < 0.01 | 2.69 (2.36–3.08) | < 0.01 |
| Smoking | ||||||
| No | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Yes | 1.04 (0.93–1.17) | 0.47 | 1.06 (0.94–1.20) | 0.35 | 1.06 (0.94–1.20) | 0.37 |
| Duration of diabetes (years) | ||||||
| < 10 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 10 | 1.41 (1.26–1.58) | < 0.01 | 1.24 (1.09–1.40) | < 0.01 | 1.24 (1.09–1.40) | < 0.01 |
| Body mass index (kg/m2) | ||||||
| < 25 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| ≥ 25 | 0.88 (0.79–0.99) | < 0.05 | 0.95 (0.84–1.08) | 0.47 | 0.95 (0.83–1.08) | 0.41 |
| Past medical history* | ||||||
| No | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| Yes | 1.27 (1.10–1.45) | < 0.01 | 1.17 (1.01–1.36) | < 0.05 | 1.18 (1.02–1.36) | < 0.05 |
| Use of interdental cleaning aids | ||||||
| Yes | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| No | 1.93 (1.71–2.18) | < 0.01 | 2.02 (1.76–2.31) | < 0.01 | 2.02 (1.76–2.32) | < 0.01 |
| Regular dental checkups | ||||||
| Yes | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |||
| No | 1.37 (1.22–1.53) | < 0.01 | 1.24 (1.09–1.41) | < 0.01 | 1.24 (1.09–1.41) | < 0.01 |
| HbA1c (%) | ||||||
| < 7.0 | 1.0 (reference) | 1.0 (reference) | – | |||
| ≥ 7.0 | 1.02 (0.91–1.14) | 0.76 | 1.05 (0.93–1.19) | 0.40 | – | |
| < 8.0 | 1.0 (reference) | – | 1.0 (reference) | |||
| ≥ 8.0 | 1.02 (0.89–1.17) | 0.77 | – | 1.16 (1.00–1.34) | < 0.05 | |
*No, no history of hypertension, hyperlipidemia, myocardial infarction, and cerebrovascular disease
**Adjusted for age, sex, body mass index, time from onset of diabetes, smoking status, past medical history, interdental cleaning, regular dental checkups, and HbA1c (7%)
***Adjusted for age, sex, body mass index, time from onset of diabetes, smoking status, past medical history, interdental cleaning, regular dental checkups, and HbA1c (8%)
Significant differences are shown in bold
Discussion
Despite the wide array of its clinical presentations, periodontal disease is broadly divided into gingivitis (in which the inflammation is found to be confined to the gum) and periodontitis (in which the inflammation is found to have extended beyond the gum line to the periodontium) [3]. While, as noted earlier, the causes of periodontal disease vary, traumatic factors (e.g., malocclusion), systemic disease (e.g., diabetes mellitus), and lifestyle (e.g., smoking) or oral habits are also assumed to come into play as disease modifiers to affect the clinical course of periodontal disease. Thus, removal of dental biofilms, i.e., appropriate routine dental cleansing, remains the key to prevention and treatment of periodontal disease.3
While a 2016 survey on dental and oral diseases showed that those who reported brushing their teeth every day and those who reported doing so twice or more frequently every day accounted for 95.3% and 77.0% of all respondents, respectively [8], the present survey in patients with diabetes showed that those who reported brushing their teeth twice or more frequently every day accounted for 68.8% of the respondents, a somewhat lower proportion than that in the 2016 survey. Furthermore, the 2016 survey also showed that those with suspected gingivitis accounted for 24.6% or one-fourth of all respondents with suspected gingivitis at the ages of 10–14 years and gradually increased over time, so that those with periodontal pockets deeper than 4 mm suggestive of periodontitis accounted for more than half (57.9%) of all respondents with suspected periodontitis at the ages of 60–64 years (close to the mean patient age in this survey) [8]. While the current study focused on periodontal disease cannot readily be compared with the 2016 survey on general dental and oral health, about one-third of both patients with type 1 (33.9%) and type 2 (31.9%) diabetes in this study reported having had gingival swelling, which was thought likely to be associated with progression of periodontitis. Likewise, about one-fifth of both patients with type 1 (15.9%) and type 2 (16.9%) diabetes reported having lost teeth in the past 1 year.
Furthermore, the number of teeth retained by the respondents at the ages of 60–64 years (close to the mean age of the participants in this survey) was 23.9 (males, 23.7; females, 24.0) in the 2016 survey on dental and oral diseases [8], but was 19.8 among the patients with diabetes in this study, i.e., 4 teeth fewer than in the 2016 survey. Again, those aged 60–64 years who retained 20 or more teeth accounted for 85.2% and 58.4%, respectively, in the 2016 survey [8] and the current study, showing that patients with diabetes who had retained 20 or more teeth accounted for a smaller proportion.
Non-use of interdental cleaning aids may be counted among the potential reasons for the high prevalence of periodontal disease in the Japanese population despite oral and dental cleaning being practiced on a daily basis by the majority. Of note here is that the interdental gingiva, also called the “col”, is a saddle (concave)-shaped tissue and thus is not only thought to be prone to biofilm deposit formation, covered by non-keratinized epithelium, but is also thought to be less resistant to bacterial infection. Furthermore, given that biofilm deposits in the interdental gingiva cannot be removed by brushing alone, interdental cleaning aids (e.g., interdental brush and/or dental floss) are deemed essential for dental hygiene and their improper use is likely to lead to onset of periodontal disease [2]. On the other hand, despite implementation of the health-promoting policy “Healthy Japan 21” [9], which aimed at increasing the proportion of users of interdental cleaning aids in their 40 s and 50 s to more than 50% of all individuals in these age groups in 2010 as a step toward nationwide prevention of periodontal disease, dental floss users and interdental brush users accounted for only about 13% and 20% of the Japanese population, respectively, in the 2010 National Health/Nutrition Survey [10]. Again, while those who reported using interdental cleaning aids accounted for 30.6% of all respondents and those aged 60–64 years (close to the mean patient age in this survey) who reported using these aids accounted for 49.9% of all respondents in this age group (thus tending to increase over time) in the 2016 survey [8], those who reported using interdental cleaning aids accounted for only 37.1% in this study, with those who reported using these aids five or more times a week or almost every day of the week accounting for a still smaller proportion (16.5%).
While the “Healthy Japan 21” [9] proposes to increase the proportion of individuals undergoing regular dental checkups to more than 30% of the entire population, 29.2%, 6.0%, and 8.3% of diabetic patients in this study reported undergoing these checkups once or twice a year, three to five times a year, and more than five times a year, respectively, showing that about 45% of patients had undergone regular dental checkups, far in excess of the target expected to be achieved under the “Healthy Japan 21” scheme. These results provide the impetus to focus attention in the years to come on promoting not only appropriate oral cleaning including day-to-day aggressive interdental cleaning but also regular dental checkups, among patients with diabetes.
Given that while less prevalent than type 2 diabetes, type 1 diabetes is more likely to affect younger individuals and thus allows its association with periodontal disease to be evaluated without involving systemic confounding factors (e.g., smoking and age), there are numerous reports available on the association between type 1 diabetes and periodontal disease in the literature. Of these, cross-sectional studies have shown no change in attachment level (AL) or alveolar bone resorption as an indicator of periodontal attachment loss, reflecting the fact that the majority of patients included were younger. Again, despite numerous reports suggesting an association between diabetes and periodontal disease, the available systematic reviews remain rather inconclusive [11, 12]. Thus, overall, early diagnosis and treatment of periodontal disease and rigorous glycemic control are deemed essential in type 1 diabetic patients, given that, in these patients, gingivitis is likely to become predominant even with the smallest amount of dental biofilm, and the longer the course of gingivitis, the more is the attachment between the teeth and the periodontal tissue is likely to become lost, leading to increased alveolar bone resorption [3, 6, 11–13]. Of note, as gingival disease associated with chronic disease, not only diabetes-associated gingivitis but more advanced periodontitis are classified by the Japanese Society of Periodontology as chronic periodontitis associated with diabetes [3].
In this light, a series of epidemiological studies in Pima Indian adults initiated in 1983 by the State University of New York at Buffalo are of note as the first large-scale studies to elucidate the relationship between type 2 diabetes and periodontal disease. In this series, Shlossman et al. [14] evaluated a total of 2878 Pima Indians and demonstrated that those with type 2 diabetes had more severe periodontal disease than those without. Shortly afterward, of these patients, Emrich et al. [15] closely examined a total of 1342 non-edentulous patients stratified by age and not only confirmed that those with type 2 diabetes had more severe periodontal disease than those without but also demonstrated that those with type 2 diabetes had 2.8 (OR 2.81; 95% CI 1.91–4.13) times the odds of having severe periodontitis when evaluated using AL as an indicator of periodontitis and 3.4 (OR 3.43; 95% CI 2.28–5.16) times the odds of dosing so when evaluated using alveolar bone resorption as a measure of periodontitis, compared to those without.
Likewise, in a study involving Japanese subjects, of all subjects without periodontitis at baseline (n = 5856), those with HbA1c ≥ 6.5% were found to have 1.2 (relative risk [RR] 1.17; 95% CI 1.01–1.36) times the risk of developing periodontitis 5 years later, while, conversely, of all subjects with HbA1c < 6.5% at baseline (n = 6125), those with periodontitis at baseline were found to have 3.5 (RR 3.45; 95% CI 1.08–11.02) times the risk of having HbA1c ≥ 6.5% 5 years later [16]. Indeed, several systematic reviews suggest that the poorer the glycemic control in patients with diabetes, the more likely they are to have periodontitis and the more likely they are to have severe periodontitis and lose teeth [4, 6, 11, 12].
As for the relationship between glycemic control and the number of present teeth, this study showed that type 1 diabetic patients with HbA1c ≥ 7.0% were 2.4 (9 variable-adjusted OR 2.38; 95% CI 1.25–4.78) times more likely to have fewer than 20 present teeth than those with HbA1c < 7.0%; and that type 2 diabetic patients with HbA1c ≥ 8.0% were 1.2 (9 variable-adjusted OR 1.16; 95% CI 1.00–1.34) times more likely to have fewer than 20 present teeth than those with HbA1c < 8.0%, while there was no difference between those with HbA1c ≥ 7.0% and those with HbA1c < 7.0%. In other words, patients with type 2 diabetes did not differ in their odds of having fewer than 20 present teeth when evaluated using HbA1c 7.0% as the threshold for prevention of diabetic complications [6, 17], but they significantly differed when evaluated using HbA1c 8.0 as the threshold for those less amenable to treatment intensification [6, 17]. Again, as regards the risk of periodontal disease in patients with diabetes, it becomes clear from reports in the literature that HbA1c 7.0% places patients at high risk of worsening periodontal disease, while HbA1c ≥ 9.0% places them at marked risk of doing so [4, 11, 12]. In this study, the number of present teeth significantly differed among patients with type 2 diabetes, depending on whether their HbA1c remains below or exceeds the threshold of 8.0% for those less amenable to treatment intensification, thus demonstrating a similar trend to that reported to date. In contrast, no clear relationship was found between HbA1c and the number of present teeth among patients with type 1 diabetes in this study, reflecting the fact that they not only were fewer in number than patients with type 2 diabetes but also consisted variously of those with acute-onset, slowly progressive, and fulminant-type diabetes.
Glycemic control and the number of present teeth in patients with type 1 and type 2 diabetes
This study examined the relationship between baseline glycemic control and oral findings in a total of 6099 patients with diabetes (type 1, 6%; type 2, 94%) included in the JDCP study and found that the mean number of present teeth was 19.8 overall, but tended to be fewer in patients with type 2 diabetes than in patients with type 1 diabetes and in women than in men. Using nine confounding factors as explanatory variables and number of present teeth (< 20/ ≥ 20) as a dependent variable, multiple logistic regression analysis demonstrated that type 1 diabetic patients with HbA1c ≥ 7% had 2.38 (OR 2.38; 95% CI 1.25–4.78) times the odds of having fewer than 20 present teeth compared to those with HbA1c < 7.0% and that type 2 diabetes with HbA1c ≥ 8.0% had 1.16 (OR 1.16; 95% CI 1.00–1.34) times the odds of having fewer than 20 present teeth, suggesting that patients with diabetes are associated with fewer present teeth and that the poorer their glycemic control becomes, the higher is the risk of losing teeth.
Acknowledgements
The authors would like to express their sincere thanks to all physicians and medical staff at the 464 participating facilities for their cooperation and to all diabetic patients for their participation in the study.
Compliance with ethical standards
Funding
Funding to support this study was received from Japan Diabetes Society.
Conflict of interest
Rimei Nishimura: lecture fees (Sanofi K.K., Medtronic Japan Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd., Takeda Pharmaceutical Co., Ltd., Kissei Pharmaceutical Co., Ltd., Novartis Pharma K.K., Eli Lily Japan Co., Ltd., Novo Nordisk Pharma Ltd., Astellas Pharma Inc., MSD K.K.); scholarship funds (Japan Diabetes Foundation, Nippon Boehringer Ingelheim Co., Ltd.). Naoko Tajima: lecture fees (Takeda Pharmaceutical Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd.)
Footnotes
The Japan Diabetes Society launched the Diabetes Registry Configuration Committee to conduct the Japan Diabetes Complication and its Prevention prospective (JDCP) study, which reported, in Japanese, the results of a large-scale observational study to investigate the current status of diabetes complications and their prevention in Japan [1]. This is an English version of that report.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Inagaki K, Kikuchi T, Noguchi T, et al. A Large-scale Observational Study to Investigate the Current Status of Diabetic Complication and Their Prevention in Japan (JDCP study 6): Baseline Dental and Oral Findings. J Jpn Diabet Soc. 2020;63(4):195–205. doi: 10.1007/s13340-020-00465-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Williams RC, Offenbacher S. Periodontal medicine: the emergence of a new branch of periodontology. Periodontol. 2000;23:9–12. doi: 10.1034/j.1600-0757.2000.2230101.x. [DOI] [PubMed] [Google Scholar]
- 3.Japanese Society of Periodontology (2016) JSP Clinical Practice Guideline for Periodontal Treatment 2015, Ishiyaku Pub Inc., Tokyo, pp 8–17 (in Japanese)
- 4.Japanese Society of Periodontology (2015) JSP Clinical Practice Guideline for Periodontal Treatment in Diabetic Patients 2014, 2nd Edition, Ishiyaku Pub Inc., Tokyo, pp 12–23 (in Japanese)
- 5.Japanese Society of Periodontology (2016) Periodontal Disease and Systemic Health, Ishiyaku Pub Inc., Tokyo, pp 88–91 (in Japanese)
- 6.Japan Diabetes Society (2019) Clinical Practice Guidelines for Treatment of Diabetes 2019, Nankodo Co., Ltd., Tokyo, pp 24–27, 219–228 (in Japanese)
- 7.Ministry of Education, Culture, Sports, Science and Technology/Ministry of Health, Labour and Welfare of Japan (2011) Ethical Guidelines for the Conduct of Epidemiological Research, https://www.mhlw.go.jp/general/seido/kousei/i-kenkyu/ekigaku/0504sisin.html. Accessed 17 Aug 2020 (in Japanese)
- 8.Ministry of Health, Labour and Welfare of Japan (2016) Results from the 2016 Survey on Dental and Oral Diseases. https://www.mhlw.go.jp/toukei/list/dl/62-28-01.pdf. Accessed 17 Aug 2020 (in Japanese)
- 9.Ministry of Health, Labour and Welfare (2000) Healthy Japan 21. Dental health. https://www.mhlw.go.jp/www1/topics/kenko21_11/b6.html. Accessed 17 Aug 2020 (in Japanese)
- 10.Ministry of Health, Labour and Welfare of Japan (2010) Summary of results from the 2010 National Health/ Nutrition Survey. https://www.mhlw.go.jp/stf/houdou/2r98520000020qbb.html. Accessed 17 Aug 2020 (in Japanese)
- 11.Borgnakke WS, Ylöstalo PV, Taylor GW, Genco RJ. Effect of periodontal disease on diabetes: systematic review of epidemiologic observational evidence. J Periodontol. 2013;84(4 Suppl):S135–152. doi: 10.1902/jop.2013.1340013. [DOI] [PubMed] [Google Scholar]
- 12.Graziani F, Gennai S, Solini A, Petrini M. A systematic review and meta-analysis of epidemiologic observational evidence on the effect of periodontitis on diabetes. An update of the EFP-AAP review. J Clin Periodontol. 2018;45(2):167–187. doi: 10.1111/jcpe.12837. [DOI] [PubMed] [Google Scholar]
- 13.Inagaki K, Kikuchi T, Mitani A, Naruse K, Matsubara T. Prevention of diabetic complications, periodontal disease. Naika (Intern Med) 2018;212(1):121–126. [Google Scholar]
- 14.Shlossman M, Knowler WC, Pettitt DJ, Genco RJ. Type 2 diabetes mellitus and periodontal disease. J Am Dent Assoc. 1990;121(4):532–536. doi: 10.14219/jada.archive.1990.0211. [DOI] [PubMed] [Google Scholar]
- 15.Emrich LJ, Shlossman M, Genco RJ. Periodontal disease in non-insulin-dependent diabetes mellitus. J Periodontal. 1991;62(2):123–131. doi: 10.1902/jop.1991.62.2.123. [DOI] [PubMed] [Google Scholar]
- 16.Morita I, Inagaki K, Nkamura F, Noguchi T, Matsubara T, Yoshii S, Nakagaki H, Mizuno K, Sheiham A, Sabbah W. Relationship between periodontal status and levels of glycated hemoglobin. J Dent Res. 2012;91(2):161–166. doi: 10.1177/0022034511431583. [DOI] [PubMed] [Google Scholar]
- 17.Council for Promotion of Countermeasures against Diabetes (Japan Medical Association, Japan Diabetes Society, Japan Association for Diabetes Education and Care) (2017) Essentials of Diabetes Treatment 2017, Bunkodo Co., Ltd., Tokyo, p 7 (in Japanese)