People with type 1 diabetes mellitus manifest a greater burden of both periodontal disease and coronary artery disease (CAD); however, little is known about their interrelation. Coronary artery calcium (CAC) measures subclinical atherosclerosis and predicts major adverse coronary events. The relation between periodontal disease and CAC progression in individuals with type 1 diabetes has not been previously described. We determined the prevalence and progression of CAC in relation to self-reported periodontal disease. Multivariate logistic and tobit regression models were used to examine the relation between periodontal disease duration and CAC progression and whether this relation differs by diabetes status after controlling for age, gender, total and high-density lipoprotein cholesterol, hypertension, smoking, body mass index (BMI), duration of diabetes, and baseline CAC. A total of 473 patients with type 1 diabetes and 548 without diabetes were followed for a mean of 6.1 years. At baseline, the prevalence and duration of periodontal disease did not differ between subjects with and without diabetes (14.5% vs 13.4%, p = 0.60; 6 vs 9 years, p = 0.18). Duration of periodontal disease was not significantly associated with baseline CAC prevalence. In patients with type 1 diabetes, periodontal disease duration was significantly related to CAC progression (p = 0.004) but not in subjects without diabetes (p = 0.63). In conclusion, this study suggests that periodontal disease is an independent predictor of long-term progression of CAC in patients with type 1 diabetes.
Coronary artery calcium (CAC) is an established marker of atherosclerotic plaque burden that is increased in patients with type 1 diabetes. To date, the influence of periodontal disease on CAC progression in patients with type 1 diabetes has not been established. The Coronary Artery Calcification in Type 1 Diabetes (CACTI) Study was used to examine the relation between periodontal disease and CAC progression over prolonged follow-up. We hypothesized that periodontal disease would be increased in type 1 diabetic individuals compared with non-diabetic individuals, and that the presence and duration of periodontal disease would predict CAC progression.
Methods
Subjects were identified from the CACTI Study. CACTI is a prospective cohort study examining the prevalence of CAC in patients with type 1 diabetes and a comparable group of controls with no history of atherosclerotic cardiovascular disease. Detailed descriptions of the study design have previously been published. Participants in this study included those who had completed the 6-year (range 4.1 to 8.5 years) follow-up visit and had complete data for all covariates (n = 1,021). Informed consent was provided by all study participants, and the protocol was reviewed and approved by the Colorado Multiple Institutional Review Board.
Physical examination measurements included height, weight, waist and hip circumference, and systolic and diastolic blood pressure. Hypertension was defined as systolic blood pressure ≥140 mm Hg, diastolic blood pressure ≥90 mm Hg, or treatment with antihypertensive medication. A fasting blood sample was collected and stored at −80°C until assayed for measurement of total cholesterol, high-density lipoprotein, and triglyceride levels. Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald equation. All subjects were given standardized questionnaires to obtain demographics, medical history, medication use, and current and past smoking status. Subjects were asked if they had ever been told by a dentist or dental hygienist that they have gum disease. If yes, then they were asked the age of diagnosis. Self-reported periodontal disease has been evaluated in other studies and demonstrated to be reasonably accurate.
Two sets of images for CAC scoring were obtained between March 2000 and May 2002 using an ultrafast Imatron C-150XLP electron beam computed tomography scanner (Imatron, San Francisco, California). The estimated radiation dose for each examination was 2.25 mSv. The CAC score was assessed using the Agatston method, and the total volume score using the volumetric method was calculated from the images, as described previously. Follow-up scans were performed approximately 6 years after the baseline visit between February 2006 and March 2009. Prevalence of CAC was defined as a CAC score >0. Progression of CAC was defined as the change in the square-root–transformed volume score between the follow-up visit and the baseline visit. We used the difference in the square-root–transformed volume for all analysis.
Differences in characteristics at baseline and follow-up were compared by diabetes status. Parametric continuous data are presented as means and standard deviation. Nonparametric data are presented as the median and interquartile range with the exception of the CAC scores. CAC scores were presented as the prevalence of positive CAC (>0) and as the median and interquartile range of only the positive values. Categorical data were tabulated as the number of subjects and the percent. Statistical testing to detect differences between groups included the t test for parametric continuous data, the Wilcoxon rank-sum test for nonparametric data, and the chi-square test for categorical data.
Logistic and tobit regression were used to determine the independent effect of periodontal disease presence at baseline and duration on the prevalence and progression of CAC. Previous data suggests that values of change in the square-root–transformed volume of CAC ≥2.5 are robust to interscan variability. Therefore, we used tobit regression for the progression of CAC models to left censor values of change (<2.5) that may represent interscan variability while estimating linear parameters for values representing significant progression (≥2.5). Tobit models assume the presence of a latent outcome variable that equals the observed outcome variable above the censored value, in this case, ≥2.5, and is unobserved at or below the censored value. The models use maximum likelihood estimation to linearly regress the latent outcome variable on the covariates, and therefore, the beta coefficients from these models provide an estimate of the linear relation between the independent variables in the model and the latent outcome variable. Potential confounding variables were considered for inclusion in the models on the basis of a priori criteria: significance in previous studies or confounding the association between the main variable of interest and the outcome by >10%. Age at baseline, gender, total cholesterol, HDL cholesterol, hypertension, smoking, body mass index (BMI), duration of type 1 diabetes, prevalence of periodontal disease, and baseline CAC were included in the final models. Additional variables that were considered for inclusion but not found to meet the a priori criteria were race, Hispanic ethnicity, years of follow-up, waist circumference, waist-to-hip ratio, triglycerides, on lipid-lowering drugs, high sensitivity C-reactive protein (hsCRP), and adiponectin. All analyses were performed using SAS/STAT software, Version [9.3] of the SAS System for Windows (SAS Institute Inc., Cary, North Carolina).
Results
A total of 473 patients with established type 1 diabetes and 548 nondiabetic individuals were assessed. Participant characteristics at baseline stratified by diabetes status are presented in Table 1 . Those patients with diabetes were younger and more likely to be receiving medications for hypertension and dyslipidemia. Corresponding diastolic blood pressure, total cholesterol, LDL cholesterol, and triglycerides were lower in those with diabetes compared to those without diabetes. Gender, smoking status, BMI, and hsCRP did not differ between the groups. At baseline, the prevalence and duration of periodontal disease were similar between those with and without diabetes. However, at the 6-year follow-up, the prevalence of periodontal disease was significantly higher in those with diabetes compared to those without diabetes. CAC prevalence, extent, and progression were higher in those with diabetes compared to those without at both the baseline and follow-up examinations.
| Variable | Overall (n=1021) | Diabetes (n=473) | No Diabetes (n=548) | p-value |
|---|---|---|---|---|
| Age (years) ∗ | 39 (±9) | 37 (±9) | 41 (±9) | <0.001 |
| Sex (male) † | 496 (48.6%) | 217 (45.9%) | 279 (50.9%) | 0.11 |
| Race (white) † | 911 (89.6%) | 448 (95.1%) | 463 (84.8%) | <0.001 |
| Hispanic † | 67 (6.6%) | 16 (3.4%) | 51 (9.4%) | <0.001 |
| Past smoking † ‡ | 214 (21.3%) | 94 (20.3%) | 120 (22.2%) | 0.46 |
| Current smoking † ‡ | 94 (9.4%) | 50 (10.8%) | 44 (8.2%) | 0.15 |
| Hypertension † | 273 (26.8%) | 190 (40.3%) | 83 (15.2%) | <0.001 |
| Antihypertensive drugs † | 203 (19.9%) | 163 (34.5%) | 40 (7.3%) | <0.001 |
| Lipid lowering drugs † | 112 (11.0%) | 77 (16.3%) | 35 (6.4%) | <0.001 |
| Hemoglobin A 1c (%) ∗ | 6.6 (±1.5) | 7.9 (±1.2) | 5.5 (±0.4) | <0.001 |
| Systolic blood pressure (mm Hg) ∗ | 115.4 (±12.8) | 116.7 (±13.3) | 114.3 (±12.3) | 0.004 |
| Diastolic blood pressure (mm Hg) ∗ | 78.1 (±8.5) | 77.1 (±8.6) | 79.0 (±8.4) | <0.001 |
| Total cholesterol (mg/dL) ∗ | 184 (±37) | 173 (±33) | 193 (±37) | <0.001 |
| High-density lipoprotein cholesterol (mg/dL) ∗ | 53 (±16) | 56 (±17) | 51 (±15) | <0.001 |
| Low-density lipoprotein cholesterol (mg/dL) ∗ | 108 (±32) | 99 (±28) | 116 (±33) | <0.001 |
| Triglycerides (mg/dL) § | 90 (67-129) | 76 (60-105) | 104 (77-155) | <0.001 |
| Body mass index (kg/m 2 ) ∗ | 26.2 (±4.5) | 26.1 (±4.2) | 26.2 (±4.8) | 0.74 |
| C-reactive protein (mg/l) § | 1.2 (0.89-2.0) | 1.2 (0.89-2.1) | 1.2 (0.88-1.9) | 0.49 |
| Self-reported periodontal disease at baseline † | 141 (13.9%) | 68 (14.5%) | 73 (13.4%) | 0.60 |
| Self-reported periodontal disease at follow-up † | 185 (18.2%) | 103 (21.9%) | 82 (15.0%) | 0.005 |
| Duration of periodontal disease (years) § | 9 (3-13) | 6 (2-11) | 9 (4-15) | 0.18 |
| Duration of diabetes (years) ∗ | N/A | 23 (±9) | N/A | N/A |
| Coronary artery calcification prevalence (>0) at baseline † | 317 (31.1%) | 169 (35.7%) | 148 (27.0%) | 0.003 |
| Coronary artery calcification prevalence (>0) at follow-up † | 534 (52.9%) | 278 (59.5%) | 256 (47.2%) | <0.001 |
| Coronary artery calcification score of positives at baseline § | 8.2 (1.4-72.4) | 14.6 (1.4-98.9) | 4.7 (1.1-31.8) | 0.011 |
| Coronary artery calcification score of positives at follow-up § | 13.7 (2.2-100.4) | 30.9 (3.3-201.7) | 8.0 (1.5-51.2) | <0.001 |
| Coronary artery calcification progression prevalence ¶ | 359 (35.2%) | 201 (42.5%) | 158 (28.8%) | <0.001 |
| Coronary artery calcification progression ≥2.5 § ‖ | 5.7 (3.9-9.5) | 6.5 (4.2-10.9) | 5.0 (3.6-7.6) | <0.001 |
∗ Data presented as mean ± SD, p-value from t-test.
† Data presented as number (%), p-value from chi-square or Fisher’s exact.
‡ Referent category never smoking.
§ Data presented as median (interquartile range), p-value from Wilcoxon rank sum test.
‖ Difference in square-root transformed volume scores between baseline and follow-up for those with positive progression (≥2.5) only.
¶ Data presented as number (%) of those with progression of CAC (≥2.5 square-root transformed units difference in volume scores between baseline and follow-up).
The median CAC score for positive values stratified by diabetes and periodontal disease status is shown in Figure 1 . For those with diabetes, subjects with baseline periodontal disease had greater median CAC scores at both the baseline (p ≤0.001) and follow-up (p = 0.002) examinations. For subjects without diabetes, there were no significant differences between the CAC scores at baseline or follow-up by periodontal disease at baseline.
Table 2 presents the results of age- and gender-adjusted logistic and tobit regression models for CAC prevalence at baseline and progression by the baseline characteristics. Self-reported periodontal disease was significantly associated with progression of CAC but not prevalence of CAC.
| Variable | Coronary Artery Calcium Prevalence | Coronary Artery Calcium Progression ∗ | |
|---|---|---|---|
| OR (95% CI) † | β ±SE | p-value | |
| Age (years) ‡ | 1.09 (1.07, 1.11) | 0.34 ±0.03 | <0.001 |
| Type 1 diabetes | 2.57 (1.88, 3.50) | 5.3 ±0.53 | <0.001 |
| Male § | 3.02 (2.25, 4.04) | 3.7 ±0.54 | <0.001 |
| Race (Non-white vs white) | 0.79 (0.49, 1.29) | -1.4 ±0.90 | 0.12 |
| Hispanic | 1.05 (0.60, 1.86) | -0.33 ±0.37 | 0.38 |
| Past smoking | 1.24 (0.87, 1.76) | 0.001 ±0.66 | 0.99 |
| Current smoking | 1.83 (1.13, 2.96) | 2.5 ±0.87 | 0.004 |
| Hypertension | 2.69 (1.97, 3.68) | 5.2 ±0.54 | <0.001 |
| Antihypertensive drugs | 2.94 (2.08, 4.16) | 5.3 ±0.58 | <0.001 |
| Lipid lowering drugs | 2.97 (1.92, 4.59) | 4.9 ±0.73 | <0.001 |
| Hemoglobin A 1c (%) | 1.40 (1.26, 1.55) | 1.65 ±0.17 | <0.001 |
| Systolic blood pressure (mmHg) | 1.03 (1.02, 1.04) | 0.16 ±0.02 | <0.001 |
| Diastolic blood pressure (mmHg) | 1.02 (1.00, 1.04) | 0.09 ±0.03 | 0.004 |
| Total cholesterol (mg/dL) | 1.00 (1.00, 1.01) | -0.02 ±0.007 | 0.03 |
| High-density lipoprotein cholesterol (mg/dL) | 0.99 (0.98, 1.00) | -0.03 ±0.02 | 0. 10 |
| Low-density lipoprotein cholesterol (mg/dL) | 1.00 (1.00, 1.01) | -0.02 ±0.009 | 0.048 |
| Log triglycerides (mg/dL) | 1.54 (1.18, 2.03) | 0.44 ±0.50 | 0.381 |
| Body mass index (kg/m 2 ) | 1.15 (1.11, 1.19) | 0.30 ±0.06 | <0.001 |
| Log C-reactive protein(mg/l) | 1.53 (1.22, 1.92) | 1.3 ±0.42 | 0.002 |
| Self-reported periodontal disease | 1.16 (0.78, 1.74) | 1.2 ±0.73 | 0.090 |
| Duration of diabetes (years) | 1.05 (1.02, 1.07) | 0.24 ±0.05 | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
