OBJECTIVE

Information on the relationship among dysglycemia (prediabetes or diabetes), myocardial infarction (MI), and periodontitis (PD) is limited. This study tests the hypothesis that undetected dysglycemia is associated with both conditions.

RESEARCH DESIGN AND METHODS

The PAROKRANK (Periodontitis and Its Relation to Coronary Artery Disease) study included 805 patients with a first MI and 805 matched control subjects. All participants without diabetes (91%) were examined with an oral glucose tolerance test. Abnormal glucose tolerance (AGT) (impaired glucose tolerance or diabetes) was categorized according to the World Health Organization. Periodontal status was categorized from dental X-rays as healthy (≥80% remaining alveolar bone height), moderate (79–66%), or severe (<66%) PD. Odds ratios (ORs) and 95% CIs were calculated by logistic regression and were adjusted for age, sex, smoking, education, marital status, and explored associated risks of dysglycemia to PD and MI, respectively.

RESULTS

AGT was more common in patients than in control subjects (32% vs. 19%; P < 0.001) and was associated with MI (OR 2.03; 95% CI 1.58–2.60). Undetected diabetes was associated with severe PD (2.50; 1.36–4.63) and more strongly in patients (2.35; 1.15–4.80) than in control subjects (1.80; 0.48–6.78), but not when categorized as AGT (total cohort: 1.07; 0.67–1.72). Severe PD was most frequent in subjects with undetected diabetes, and reversely undetected diabetes was most frequent in patients with severe PD.

CONCLUSIONS

In this large case-control study previously undetected dysglycemia was independently associated to both MI and severe PD. In principal, it doubled the risk of a first MI and of severe PD. This supports the hypothesis that dysglycemia drives two common diseases, MI and PD.

The global prevalence of dysglycemia (diabetes or prediabetes) is rapidly increasing (1). Diabetes is an important risk factor for myocardial infarction (MI) and periodontitis (PD), both of which are significant public health problems. The progress of atherosclerosis is promoted by chronic inflammatory conditions, and active systemic inflammation increases the risk for atherosclerotic plaque rupture leading to unstable angina or MI (2,3). Counteracting inflammation reduces cardiovascular (CV) events in patients with established CV disease (CVD) as recently shown by the Canakinumab Antiinflammatory Thrombosis Outcome Study (CANTOS) trial, which established a causal relationship between inflammation and atherosclerosis (4). PD is a chronic inflammatory condition, predominantly induced by Gram-negative bacteria colonizing the gingival crevice successively degrading the tissues attaching the teeth to the alveolar bone (5). The Swedish case-control study PAROKRANK (Periodontitis and Its Relation to Coronary Artery Disease) recently revealed that the risk for MI among those with PD was increased even after adjustment for relevant confounders, thereby strengthening the possibility of an independent relationship between PD and MI (6).

A two-way relationship between diabetes and PD has been proposed and was recently reviewed (i.e., that hyperglycemia increases the risk for the development of PD and that the presence of PD increases the risk for diabetes) (7). Whether this association also extends to persons with previously undetected dysglycemia has, however, not been extensively elucidated, especially not in populations including subjects both with and without prevalent CVD (810). Since both established diabetes and undetected dysglycemia are associated with MI (11,12) and since PD seems to be related to MI (6), it can be hypothesized that PD should be more common among patients with previously undetected dysglycemia who have experienced an MI than in healthy control subjects.

The objective with this report was to test the hypothesis that undetected dysglycemia is associated with both MI and PD.

Data Sources and Study Population

For this analysis, data collected in the case-control study PAROKRANK were used. A detailed description of the study procedures and selection of control subjects has been published previously (6), whereas information of particular importance for this report is described below. Participants with established diabetes (patients 10%; control subjects 8%) were excluded from the present report. An oral glucose tolerance test (OGTT) was performed at all study sites since the study protocol contained the prespecified objective to analyze PD severity in relation to glycemic state.

Patients (n = 805) were <75 years of age and had received a diagnosis of a first MI according to international guidelines (13,14). Exclusion criteria were previous heart valve replacement, any conditions limiting the ability to participate in the study, or unwillingness to participate. Recruitment occurred from May 2010 to February 2014 at 17 Swedish coronary care units. OGTTs and dental examinations were performed at the local cardiology outpatient clinic and the hospital dental care unit within 6–10 weeks after hospital discharge. After the exclusion of patients with already established diabetes, 719 patients remained.

Control subjects (n = 805) were persons matched for age, sex, and geography (by postal code) who had been selected from the national population registry. To be considered, control subjects were free from prior MI or heart valve replacement and willing to participate in the study investigations. Study visits at the Departments of Cardiology and Dental Care were scheduled in close proximity to the matched patient’s study visit. After the exclusion of control subjects with already established diabetes, 739 control subjects remained.

Study Protocol

The matched control subjects were selected and investigated soon after the outpatient visit of their corresponding patients. Prior to the visit at the cardiology outpatient clinic, study participants fasted for 12 h, including abstaining from smoking. Blood samples were collected during the study visit 6–10 weeks after the MI in patients and at baseline in control subjects. Blood samples were analyzed at the local hospital laboratory for white blood cell count, glycated hemoglobin A1c (HbA1c), and fibrinogen. hs-CRP was analyzed at a central laboratory (Redhot Diagnostics, Södertälje, Sweden) by an ELISA method (MP Biomedicals).

The national quality registry SWEDEHEART (Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies; www.swedeheart.se), modified to comply with the study needs, was used to collect medical information from the patients at the time of their initial hospitalization and at the follow-up 6–10 weeks after the MI. Equivalent information was collected for the control population with data entered into a separate database.

OGTT and Definitions of Glucose Tolerance States

Participants without previously known diabetes underwent a standardized OGTT consisting of 75-g glucose diluted in 200 mL of water (15). Venous plasma glucose was measured before ingestion of the glucose solution and 120 min after using a bedside point-of-care system (HemoCue 201 System; HemoCue AB, Ängelholm, Sweden). Glucose levels obtained during the OGTT were used to classify study participants according to the World Health Organization definition (15) as normal glucose tolerance (NGT), impaired glucose tolerance (IGT), or diabetes (Supplementary Table 1). Abnormal glucose tolerance (AGT) was defined as the presence of either IGT or diabetes detected by the OGTT performed according to the study protocol.

Dental Examination

A dental examination, following a standardized protocol, was completed, and a panoramic dental X-ray was performed (6). The X-rays were analyzed according to standardized protocol centrally at the Department of Dental Medicine, Karolinska Institute Huddinge by dentists blinded to whether the study participants were patients or control subjects. Each tooth was measured. Measurements were made from the marginal bone crest to the tooth apex (total bone height) and from the cemento-enamel junction to the tooth apex (total root length) mesially and distally (for further details, see the original PAROKRANK publication) (6). The arithmetic mean, calculated from the total root length and bone height, was used as a measure of the proportion of remaining bone height supporting each tooth. Measurements were made of all teeth with visible cemento-enamel junctions and visible apices. Twenty patients had five or less reaming teeth, but no one had only one tooth. No lower limits were used as a criterion for analyzing marginal bone height.

Dental implants were not examined. Based on alveolar bone height, the study participants were allocated to one of three groups: Healthy (≥80% remaining bone), Moderate PD (79–66% remaining bone), and Severe PD (<66% remaining bone). Edentulous participants were not included in the analyses.

The dental X-ray analysis was performed by three dentists using ImageJ (Image Tool 3.0 software program; Department of Dental Diagnostics Science, University of Texas Health Science Center, San Antonio, TX) on a high-resolution computer in a darkened room. An interindividual calibration was performed in a random sample of 42 panoramic radiographs that were examined by the three dentists. These dental X-rays were graded by the three dentists in 126 separate observations. The three graders were in complete agreement in 120 of these observations (95%). The correlation between dentist 1 and 2 was 0.95; between 1 and 3, 0.90; and between 2 and 3, 0.90 (κ value 0.82) (6).

Definitions of Comorbidities

The presence of a family history of CVD (close relative with CVD below the age of 60 years) and rheumatic and pulmonary disease was based on self-reported information. The diagnoses of hypertension were based on a medical history obtained by the study personnel.

Statistical Analyses

Statistical comparisons of baseline demographics between the patients and control subjects were performed by means of the Student t test for continuous variables and the χ2 test for categorical variables. A two-sided P value <0.05 was considered significant. Odds ratios (ORs) and corresponding 95% CIs were calculated by use of logistic regression to estimate the association of glucose variables and MI within the total cohort. For the association of glucose variables and severe PD, the association was analyzed in the total cohort as well as in patients and control subjects separately. Glucose variables were analyzed as both categorized variables (e.g., as AGT) and as continuous variables (HbA1c/fasting/2-h glucose). OR and 95% CI were further adjusted for the following variables: age, sex, smoking, education, and marital status at admission since these variables were associated with MI and PD in the original PAROKRANK study (6).

All statistical analyses were performed using the SAS system (SAS system for Windows 9.4; SAS Institute Inc., Cary, NC).

Ethical Approval

The PAROKRANK study was approved by the Regional Ethics Committee at Stockholm (Dnr:2008/152–31/2) prior to the study, and all patients provided written informed consent. The PAROKRANK study was conducted according to principles outlined in the Helsinki Declaration.

In all, 719 patients and 739 control subjects underwent an OGTT and were classified into the three categories of glucose tolerance states (Fig. 1). Their mean age was 62 ± 8 years, and 19% were women. The proportion of AGT among patients and control subjects is depicted in Fig. 1 and was more common in patients than in control subjects (AGT: 32% vs. 19%, P < 0.001; diabetes: 10% vs. 6%, P = 0.001; and IGT: 22% vs. 13%, P < 0.001). Pertinent baseline demographics by glucose tolerance status in patients and control subjects separately are detailed in Table 1. In general, patients more often had a family history of CVD, were more often smokers, and had higher levels of fibrinogen and white blood cell counts than control subjects. Patients and control subjects with undetected diabetes tended to have higher mean age, BMI, waist circumference, hs-CRP, and more often a history of previous hypertension, pulmonary, and rheumatic disease compared with those categorized as NGT. Post-MI patients were subjected to intensified secondary prevention, which is reflected by higher prescription rates of CV preventive pharmaceutical agents compared with control subjects but with no major difference between glucose tolerance groups (Supplementary Table 2).

Figure 1

Proportion of AGT in patients with a first MI and matched control subjects without previously known diabetes investigated by OGTT revealing 32% of patients vs. 19% of control subjects (P < 0.001) with previously unrecognized AGT. DM, diabetes.

Figure 1

Proportion of AGT in patients with a first MI and matched control subjects without previously known diabetes investigated by OGTT revealing 32% of patients vs. 19% of control subjects (P < 0.001) with previously unrecognized AGT. DM, diabetes.

Close modal
Table 1

Pertinent baseline characteristics divided by glucose tolerance status on OGTT

Baseline characteristicsNGT
IGT
DM
Patient (n = 486)Control (n = 601)P valuePatient (n = 159)Control (n = 96)P valuePatient (n = 74)Control (n = 42)P value
Age (years) 61 ± 8 61 ± 8 0.284 63 ± 8 65 ± 6 0.191 64 ± 7 65 ± 6 0.608 
Male sex 404 (83) 479 (80) 0.150 125 (79) 80 (83) 0.358 61 (82) 35 (83) 0.901 
Waist circumference (cm) 98 ± 10 97 ± 11 0.165 100 ± 11 102 ± 10 0.113 103 ± 12 101 ± 12 0.412 
BMI (kg/m227 ± 4 26 ± 4 0.418 27 ± 4 28 ± 3 0.109 29 ± 5 28 ± 4 0.315 
Family history of CVD 193 (40) 142 (24) <0.001 57 (36) 16 (17) 0.004 25 (34) 9 (21) 0.201 
Medical history          
 Hypertension 131 (27) 152 (25) 0.529 67 (42) 42 (45) 0.724 35 (47) 22 (52) 0.599 
 Rheumatic disease 87 (18) 97 (16) 0.386 35 (22) 20 (21) 0.805 14 (19) 9 (21) 0.745 
 Pulmonary disease 53 (11) 61 (10) 0.640 20 (13) 12 (12) 0.956 12 (16) 5 (12) 0.528 
Smoking (at admission)          
 Current 123 (26) 76 (13) <0.001 45 (29) 10 (10) 0.002 20 (28) 3 (7) 0.027 
 Previous 160 (34) 255 (42)  60 (38) 50 (52)  28 (39) 22 (52)  
 Never 195 (41) 270 (45)  51 (33) 36 (38)  23 (32) 17 (40)  
Laboratory values          
 HbA1c (mmol/mol) 38 ± 4 37 ± 4 <0.001 40 ± 4 39 ± 4 0.103 45 ± 8 45 ± 11 0.976 
 HbA1c (%) 5.6 5.5 <0.001 5.8 5.7 0.103 6.3 6.3 0.976 
 F-glucose (mmol/L) 5.5 ± 0.6 5.3 ± 0.6 <0.001 5.9 ± 0.6 5.8 ± 0.6 0.618 7.0 ± 0.9 7.1 ± 1.6 0.751 
 2-h glucose (mmol/L) 5.7 ± 1.2 5.4 ± 1.3 <0.001 8.9 ± 0.8 9.0 ± 0.8 0.586 11.0 ± 3.0 12.1 ± 3.1 0.098 
 Fibrinogen (g/L) 3.3 ± 0.7 3.1 ± 0.7 <0.001 3.5 ± 0.9 3.3 ± 0.9 0.056 3.7 ± 0.8 3.1 ± 0.9 <0.001 
 hs-CRP (mg/L) 1.9 ± 2.2 2.0 ± 2.3 0.318 2.7 ± 3.0 2.6 ± 2.8 0.749 3.1 ± 3.1 3.1 ± 4.0 0.974 
 LPK (×109/L) 6.2 ± 3.4 5.6 ± 3.3 0. 001 6.8 ± 4.2 6.0 ± 1.6 0.065 8.7 ± 11.5 6.2 ± 1.6 0.156 
Education level          
 1–12 years 309 (64) 359 (60) 0. 229 115 (73) 58 (60) 0.033 48 (65) 32 (76) 0.205 
 University 176 (36) 238 (40)  42 (27) 38 (40)  26 (35) 10 (24)  
PD status (n = 480) (n = 595)  (n = 159) (n = 96)  (n = 74) (n = 40)  
 Tooth (number) 24 ± 6 25 ± 5 0.016 23 ± 6 24 ± 5 0.169 23 ± 6 23 ± 6 0.885 
 Periodontal grade (X-ray)          
  Healthy 285 (59) 408 (69)  90 (57) 68 (71)  41 (55) 22 (55)  
  Moderate 149 (31) 164 (28) <0.001 57 (36) 24 (25) 0.085 19 (26) 15 (38) 0.175 
  Severe 46 (10) 23 (4)  11 (7) 4 (4)  14 (19) 3 (8)  
Baseline characteristicsNGT
IGT
DM
Patient (n = 486)Control (n = 601)P valuePatient (n = 159)Control (n = 96)P valuePatient (n = 74)Control (n = 42)P value
Age (years) 61 ± 8 61 ± 8 0.284 63 ± 8 65 ± 6 0.191 64 ± 7 65 ± 6 0.608 
Male sex 404 (83) 479 (80) 0.150 125 (79) 80 (83) 0.358 61 (82) 35 (83) 0.901 
Waist circumference (cm) 98 ± 10 97 ± 11 0.165 100 ± 11 102 ± 10 0.113 103 ± 12 101 ± 12 0.412 
BMI (kg/m227 ± 4 26 ± 4 0.418 27 ± 4 28 ± 3 0.109 29 ± 5 28 ± 4 0.315 
Family history of CVD 193 (40) 142 (24) <0.001 57 (36) 16 (17) 0.004 25 (34) 9 (21) 0.201 
Medical history          
 Hypertension 131 (27) 152 (25) 0.529 67 (42) 42 (45) 0.724 35 (47) 22 (52) 0.599 
 Rheumatic disease 87 (18) 97 (16) 0.386 35 (22) 20 (21) 0.805 14 (19) 9 (21) 0.745 
 Pulmonary disease 53 (11) 61 (10) 0.640 20 (13) 12 (12) 0.956 12 (16) 5 (12) 0.528 
Smoking (at admission)          
 Current 123 (26) 76 (13) <0.001 45 (29) 10 (10) 0.002 20 (28) 3 (7) 0.027 
 Previous 160 (34) 255 (42)  60 (38) 50 (52)  28 (39) 22 (52)  
 Never 195 (41) 270 (45)  51 (33) 36 (38)  23 (32) 17 (40)  
Laboratory values          
 HbA1c (mmol/mol) 38 ± 4 37 ± 4 <0.001 40 ± 4 39 ± 4 0.103 45 ± 8 45 ± 11 0.976 
 HbA1c (%) 5.6 5.5 <0.001 5.8 5.7 0.103 6.3 6.3 0.976 
 F-glucose (mmol/L) 5.5 ± 0.6 5.3 ± 0.6 <0.001 5.9 ± 0.6 5.8 ± 0.6 0.618 7.0 ± 0.9 7.1 ± 1.6 0.751 
 2-h glucose (mmol/L) 5.7 ± 1.2 5.4 ± 1.3 <0.001 8.9 ± 0.8 9.0 ± 0.8 0.586 11.0 ± 3.0 12.1 ± 3.1 0.098 
 Fibrinogen (g/L) 3.3 ± 0.7 3.1 ± 0.7 <0.001 3.5 ± 0.9 3.3 ± 0.9 0.056 3.7 ± 0.8 3.1 ± 0.9 <0.001 
 hs-CRP (mg/L) 1.9 ± 2.2 2.0 ± 2.3 0.318 2.7 ± 3.0 2.6 ± 2.8 0.749 3.1 ± 3.1 3.1 ± 4.0 0.974 
 LPK (×109/L) 6.2 ± 3.4 5.6 ± 3.3 0. 001 6.8 ± 4.2 6.0 ± 1.6 0.065 8.7 ± 11.5 6.2 ± 1.6 0.156 
Education level          
 1–12 years 309 (64) 359 (60) 0. 229 115 (73) 58 (60) 0.033 48 (65) 32 (76) 0.205 
 University 176 (36) 238 (40)  42 (27) 38 (40)  26 (35) 10 (24)  
PD status (n = 480) (n = 595)  (n = 159) (n = 96)  (n = 74) (n = 40)  
 Tooth (number) 24 ± 6 25 ± 5 0.016 23 ± 6 24 ± 5 0.169 23 ± 6 23 ± 6 0.885 
 Periodontal grade (X-ray)          
  Healthy 285 (59) 408 (69)  90 (57) 68 (71)  41 (55) 22 (55)  
  Moderate 149 (31) 164 (28) <0.001 57 (36) 24 (25) 0.085 19 (26) 15 (38) 0.175 
  Severe 46 (10) 23 (4)  11 (7) 4 (4)  14 (19) 3 (8)  

Data are n (%) or mean ± SD. DM, diabetes; F-glucose, fasting glucose; LPK, leukocyte count.

Dental Health

Information on both the PD status and the glucose status was available in 712 patients and 731 control subjects. Table 1 at the bottom, describes details on dental status at dental X-ray examinations categorized by glucose status in patients and control subjects separately. Severe PD was present in 10% (n = 71 of 712) of patients and in 4% (n = 30 of 731) of control subjects (P < 0.001). Severe PD was most common in patients who have had an MI with undetected diabetes, although not significantly different from proportions in control subjects (patients [n = 14; 19%] vs. control subjects [n = 3; 8%], overall P value = 0.175).

Supplementary Table 3 describes the reverse situation (i.e., the frequency of NGT/IGT/diabetes by periodontal status in patients and control subjects separately). The highest proportion of previously undetected diabetes was observed in patients who had an MI with severe PD (20%, n = 14). In control subjects, the corresponding frequency was 10% (n = 3; overall P value = 0.233).

A detailed description of dental status including caries and edentulous information is provided in Supplementary Table 4. As can be seen, there are no major differences between the groups, apart from that patients had a more severe clinically evaluated PD and slightly more often severe PD if they had undetected diabetes.

Association of Glucose and Myocardial Infarction

Figure 2 shows a Forest plot demonstrating univariate and adjusted associations between glucose variables and first MI. AGT was associated to MI (OR 2.09; 95% CI 1.64–2.66), which remained after adjustment (2.03; 1.58–2.60). Continuous glucose variables were associated with MI and remained after adjustment (fasting glucose: 1.66; 1.43–1.92; 2-h glucose: 1.14; 1.09–1.20; HbA1c: 1.07; 1.05–1.10).

Figure 2

Associations (OR; 95% CI) of glucose variables and MI before and after adjustments for age, sex, smoking, education, and marital status. P-glucose, plasma glucose.

Figure 2

Associations (OR; 95% CI) of glucose variables and MI before and after adjustments for age, sex, smoking, education, and marital status. P-glucose, plasma glucose.

Close modal

Association of Glucose and Severe Periodontitis

Figure 3 shows details on associations of glucose variables to the presence of severe PD status on dental examinations in the total cohort. Undetected diabetes was associated with severe PD (OR 2.60; 95% CI 1.48–4.55 [adjusted OR 2.50; 95% CI 1.36–4.63]) (Fig. 3). When analyzing patients and control subjects, separately undetected diabetes remained associated with severe PD in patients (2.35; 1.15–4.80) but not in control subjects (1.80; 0.48–6.78) (Supplementary Fig. 1A and B).

Figure 3

Associations (OR; 95% CI) of glucose variables and severe periodontal disease in the total cohort before and after adjustments for age, sex, smoking, education, and marital status. P-glucose, plasma glucose.

Figure 3

Associations (OR; 95% CI) of glucose variables and severe periodontal disease in the total cohort before and after adjustments for age, sex, smoking, education, and marital status. P-glucose, plasma glucose.

Close modal

There was no association between AGT and severe PD in the total cohort (adjusted OR 1.07; 95% CI 0.67–1.72) or for IGT and severe PD (0.57; 0.31–1.05) (Fig. 3). After adjustments, none of the continuous glucose variables were significantly associated with severe PD (in the total cohort 2-h glucose: 1.09; 1.00–1.19; and in control subjects 2-h glucose 1.14; 1.00–1.29).

The main finding was the independent association between dysglycemia and a first MI and severe PD with an in principle doubled risk for both conditions even after adjustment for relevant confounders. Undetected diabetes was more frequent among subjects with signs of severe PD, especially in the presence of a first MI. To the best of our knowledge, this is the only report addressing such association in a contemporary cohort of patients investigating both diseases at the same time. The findings support the hypothesis that dysglycemia is a driver for two common diseases in the general population, MI and PD, thereby highlighting the impact that such disturbances might have on health care demands.

The PAROKRANK study provided support for an independent association between PD and MI (6). This observation is extended by the present investigation, indicating that this association is especially apparent in the presence of previously undetected diabetes. In contrast to undetected diabetes, neither AGT, which includes IGT, nor IGT, analyzed as separate entity, were significantly associated with severe PD. Thus, we found that glucose levels in the diabetes range were more clearly associated with severe PD than glucose levels in the IGT range. With the lack of previous reports on this issue, a potential explanation could be that severe PD may represent a microvascular complication to diabetes and, as such, may be more likely to develop when glucose reaches levels in that range (i.e., fasting glucose >7 mmol/L and postload values >11.0 mmol/L) (15). In contrast, continuous glucose variables were independently associated with MI for all three glucose variables (HbA1c, fasting plasma glucose, and 2-h glucose), but not with severe PD, where an association was seen only when glucose variables were dichotomized at a higher cutoff value. This finding fits with the fact that macrovascular complications related to dysglycemia start at glucose levels below those commonly considered to be indicative of diabetes, but that microvascular complications are related to the conventional cut points for diabetes (16,17). Thus, categorizing dysglycemia by OGTT, including both fasting and 2-h glucose information, seems to be the most rewarding way to identify persons at risk for both MI and PD.

In the last decade, there have been increasing numbers of publications and reviews on a two-way relationship between diabetes and PD (710,18). Several studies have shown an association of prediabetes and severe PD and reversely, when assessing glucose status by OGTT (810) or fasting glucose (18). However, these studies analyzed PD by clinical periodontal status examinations in contrast to the current study, which used objective signs mirroring a probable long-standing PD disease with alveolar bone loss. Furthermore, previous studies did not take prevalent CVD into consideration (810), whereas the current study examined subjects both with and without CVD (i.e., with and without MI). Thus, the current study gives further support to the previously suggested two-way relationship between diabetes and PD and extends the evidence to also include undetected diabetes.

Although possible mechanistic explanations for these associations are beyond the scope of the current study, a reasonable assumption is that a PD-triggered inflammation will result in increased insulin resistance with elevated glucose levels and that increased glucose levels enhance bacterial accumulation in the periodontal tissue (19). By 1993, it had been suggested that PD should be regarded as “the sixth diabetes complication” in addition to the five affecting eye, kidney, nerve, heart, and vessels (20). Since PD can impede satisfactory glucose control and has been associated with more severe diabetes complications (21,22), management guidelines issued by the American Diabetes Association and the International Diabetes Federation underline that patients with diabetes should undergo a regular periodontal examination (2325). The current study supports these recommendations. Moreover, it suggests that this risk is already present before an overt diagnosis of diabetes and that screening for PD in diabetes conditions and for diabetes in patients with severe PD should be considered in the future. However, the largest randomized intervention trial conducted so far, the Diabetes and Periodontal Therapy Trial (DPTT), did not find improved HbA1c levels with nonsurgical periodontal treatment in 514 patients with type 2 diabetes who had chronic PD during 6 months of treatment. A problem with this study is that it had to be stopped prematurely due to futility (26). It has also been criticized as misleading because of a lack of sufficient control of the parodontal infection and since glycemic control was rather satisfactory already at study start (27).

Up to 60% of MI populations have undetected dysglycemia (i.e., AGT) when screened with OGTT (12,2830). In the current study, a lower proportion (32%) of MI patients had undetected AGT. The varying frequencies are likely explained by different population characteristics. Previous reports included older patients and reinfarctions, but the current study recruited only patients with a first MI at a rather young age. Still, a twofold increased risk for MI was found, which was in line with the previous case-control study by Bartnik et al. (30).

Since undetected dysglycemia is a driver for both MI and PD, there is a need for increased collaboration between representatives for the two specialties in charge of such patients, cardiologists and diabetologists. They should be aware that undetected dysglycemia and PD are common among patients who have had an MI and that undetected diabetes is rather prevalent in the presence of severe PD, causing an increased risk for a future MI. Thus, screening for undetected dysglycemia if severe PD is found by dentists at dental clinics, and in cardiology clinics after an MI, should be considered. Since routine dental examinations are more common than routine medical checkups, such an approach would likely detect patients with prediabetes earlier. Indeed, when a NHANES (National Health and Nutrition Examination Survey) III study explored the screening of diabetes at the dental office, researchers found 25–57% probability to have diabetes diagnosed by fasting glucose levels in persons with the clustering of risk factors such as family history of diabetes, hypertension, hypercholesterolemia, and clinical signs of PD (31).

The strength of the current study is the careful study design with a standardized OGTT identifying previous undetected dysglycemia and objective measures of PD with centrally evaluated alveolar bone loss on X-ray. Further, the study population was recruited from 17 hospitals in Sweden, covering a wide geographical area and a variety of educational and socioeconomic conditions.

There are some limitations with the current study. First, the low number of patients with severe PD and undetected diabetes hampered the significance testing of the association between dysglycemia and severe PD in patients and control subjects separately and for separate interactions tests. Thus, the different findings of associations between undetected diabetes and severe PD in patients (significant) and control subjects (not significant) could be a matter of sample size rather than a true difference in associated risks. The relatively low proportion of severe PD could partly be explained by the upper age limit of 75 years. This age limit and the fact that only patients with a first MI were included further explains the lower proportion of AGT seen in this report compared with previous reports (12,2830) with higher age limits and mixed MI populations. Second, the present analyses excluded subjects with established diabetes. Third, the current study design cannot, as already underlined, give a definite answer to the temporal relationship between hyperglycemia and PD. Still, and from a clinical point of view, the present findings underline that undiscovered dysglycemia could be important to consider as a part of the development of PD, especially since it is preventable. Fourth, there is a lack information on other infections (e.g., apical PD).

Moreover, the secondary prescription of pharmaceutical agents was more intense in the post-MI case patients than in control subjects, however, without any difference between the glycemic groups. It is, however, unlikely that these pharmacological prescriptions would have influenced the bone height levels during the short time span of 6–10 weeks post-MI.

In conclusion, previously undetected dysglycemia was independently associated with both MI and severe PD in this large case-control study. In principle, it doubled the risk of a first MI and of severe PD. This supports the hypothesis that dysglycemia may be a driver for two common diseases, MI and PD.

Acknowledgments. The authors are grateful to Merja Heinonen, Department of Medicine K2, Karolinska Institutet, Stockholm, Sweden, for devoted work at the coordinating center and for primary contacts with all control subjects.

Funding. The PAROKRANK study was academic driven and was supported by grants from AFA Insurance Foundation, the Swedish Heart-Lung Foundation, the Swedish Research Council, and the Swedish Society of Medicine and the Stockholm County Council (ALF project and Steering committee KI/SLL for odontological research).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. A.N. was the responsible cardiologist, A.G. was the responsible odontologist, and B.Kl. and L.R. chaired the PAROKRANK study. A.N., B.Kj., N.H., and L.R. wrote the manuscript with input from all of the authors. A.N., B.Kj., P.N., and L.R. had access to relevant data, and all authors had final responsibility for the decision to submit for publication. A.N. and L.R. designed the glucose objectives in the PAROKRANK study. A.N. and L.R. handled the control subjects in the Stockholm area and some of the patients. B.Kj. coordinated the participating centers. P.N. performed the statistical analyses. A.N. and B.Kj. contributed to discussion about statistical analyses. A.N., A.G., B.Kl., Å.N., E.S., and L.R. participated in the original design of the PAROKRANK study. All authors had input on the manuscript. A.N. and L.R. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at ESC Congress 2015, London, U.K., 29 August to 2 September 2015.

1.
International Diabetes Federation. IDF Diabetes Atlas 8th Edition, 2017 [Internet]. Available from www.idf.org. Accessed 10 December 2017
2.
Libby
P
,
Aikawa
M
.
Stabilization of atherosclerotic plaques: new mechanisms and clinical targets
.
Nat Med
2002
;
8
:
1257
1262
[PubMed]
3.
Hansson
GK
.
Inflammation, atherosclerosis, and coronary artery disease
.
N Engl J Med
2005
;
352
:
1685
1695
[PubMed]
4.
Ridker
PM
,
Everett
B
,
Thuren
T
, et al.;
CANTOS Trial Group
.
Antiinflammatory therapy with canakinumab for atherosclerotic disease
.
N Engl J Med
2017
;
377
:
1119
1131
[PubMed]
5.
Pihlstrom
BL
,
Michalowicz
BS
,
Johnson
NW
.
Periodontal diseases
.
Lancet
2005
;
366
:
1809
1820
[PubMed]
6.
Rydén
L
,
Buhlin
K
,
Ekstrand
E
, et al
.
Periodontitis increases the risk of a first myocardial infarction: a report from the PAROKRANK Study
.
Circulation
2016
;
133
:
576
583
[PubMed]
7.
Casanova
L
,
Hughes
FJ
,
Preshaw
PM
.
Diabetes and periodontal disease: a two-way relationship
.
Br Dent J
2014
;
217
:
433
437
8.
Marugame
T
,
Hayasaki
H
,
Lee
K
,
Eguchi
H
,
Matsumoto
S
.
Alveolar bone loss associated with glucose tolerance in Japanese men
.
Diabet Med
2003
;
20
:
746
751
[PubMed]
9.
Saito
T
,
Shimazaki
Y
,
Kiyohara
Y
, et al
.
The severity of periodontal disease is associated with the development of glucose intolerance in non-diabetics: the Hisayama study
.
J Dent Res
2004
;
83
:
485
490
[PubMed]
10.
Arora
N
,
Papapanou
PN
,
Rosenbaum
M
,
Jacobs
DR
 Jr
.,
Desvarieux
M
,
Demmer
RT
.
Periodontal infection, impaired fasting glucose and impaired glucose tolerance: results from the Continuous National Health and Nutrition Examination Survey 2009-2010
.
J Clin Periodontol
2014
;
41
:
643
652
[PubMed]
11.
Yusuf
S
,
Hawken
S
,
Ôunpuu
S
, et al.;
INTERHEART Study Investigators
.
Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study
.
Lancet
2004
;
364
:
937
952
[PubMed]
12.
Norhammar
A
,
Tenerz
A
,
Nilsson
G
, et al
.
Glucose metabolism in patients with acute myocardial infarction and no previous diagnosis of diabetes mellitus: a prospective study
.
Lancet
2002
;
359
:
2140
2144
[PubMed]
13.
Thygesen
K
,
Alpert
JS
,
White
HD
, et al.;
Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction
.
Universal definition of myocardial infarction
.
Circulation
2007
;
116
:
2634
2653
[PubMed]
14.
Thygesen
K
,
Alpert
JS
,
Jaffe
AS
, et al.;
Joint ESC/ACCF/AHA/WHF Task Force for the Universal Definition of Myocardial Infarction
.
Third universal definition of myocardial infarction
.
Circulation
2012
;
126
:
2020
2035
[PubMed]
15.
World Health Organization
.
Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycaemia: Report of a WHO/IDF Consultation
.
Geneva
,
World Health Organization
,
2006
16.
Shahim
B
,
De Bacquer
D
,
De Backer
G
, et al
.
The prognostic value of fasting plasma glucose, two-hour postload glucose, and HbA1c in patients with coronary artery disease: a report from EUROASPIRE IV: a survey from the European Society of Cardiology
.
Diabetes Care
2017
;
40
:
1233
1240
[PubMed]
17.
Sarwar
N
,
Gao
P
,
Seshasai
SR
, et al.;
Emerging Risk Factors Collaboration
.
Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies
[published correction appears in Lancet 2010;376:958].
Lancet
2010
;
375
:
2215
2222
18.
Chiu
SY
,
Lai
H
,
Yen
AM
,
Fann
JC
,
Chen
LS
,
Chen
HH
.
Temporal sequence of the bidirectional relationship between hyperglycemia and periodontal disease: a community-based study of 5,885 Taiwanese aged 35-44 years (KCIS No. 32)
.
Acta Diabetol
2015
;
52
:
123
131
[PubMed]
19.
Polak
D
,
Shapira
L
.
An update on the evidence for pathogenic mechanisms that may link periodontitis and diabetes
.
J Clin Periodontol
2018
;
45
:
150
166
[PubMed]
20.
Löe
H
.
Periodontal disease. The sixth complication of diabetes mellitus
.
Diabetes Care
1993
;
16
:
329
334
[PubMed]
21.
Borgnakke
WS
,
Ylöstalo
PV
,
Taylor
GW
,
Genco
RJ
.
Effect of periodontal disease on diabetes: systematic review of epidemiologic observational evidence
.
J Periodontol
2013
;
84
(
Suppl.
):
S135
S152
[PubMed]
22.
Simpson
TC
,
Weldon
JC
,
Worthington
HV
, et al
.
Treatment of periodontal disease for glycaemic control in people with diabetes mellitus
.
Cochrane Database Syst Rev
2015
;
11
:CD004714
[PubMed]
23.
American Diabetes Association
.
3. Initial evaluation and diabetes management planning. In Standards of Medical Care in Diabetes
.
Diabetes Care
2015
;
38
:
S17
S19
24.
IDF Clinical Guidelines Task Force
.
IDF Guideline on Oral Health for People with Diabetes
.
Brussels
,
International Diabetes Federation
,
2009
25.
Sanz
M
,
Ceriello
A
,
Buysschaert
M
, et al
.
Scientific evidence on the links between periodontal diseases and diabetes: consensus report and guidelines of the joint workshop on periodontal diseases and diabetes by the International Diabetes Federation and the European Federation of Periodontology
.
Diabetes Res Clin Pract
2018
;
137
:
231
241
[PubMed]
26.
Engebretson
SP
,
Hyman
LG
,
Michalowicz
BS
, et al
.
The effect of nonsurgical periodontal therapy on hemoglobin A1c levels in persons with type 2 diabetes and chronic periodontitis: a randomized clinical trial
.
JAMA
2013
;
310
:
2523
2532
27.
Borgnakke
WS
,
Chapple
IL
,
Genco
RJ
, et al
.
The multi-center randomized controlled trial (RCT) published by the Journal of the American Medical Association (JAMA) on the effect of periodontal therapy on glycated hemoglobin (HbA1c) has fundamental problems
.
J Evid Based Dent Pract
2014
;
14
:
127
132
[PubMed]
28.
Gyberg
V
,
De Bacquer
D
,
De Backer
G
, et al.;
EUROASPIRE Investigators
.
Patients with coronary artery disease and diabetes need improved management: a report from the EUROASPIRE IV survey: a registry from the EuroObservational Research Programme of the European Society of Cardiology
.
Cardiovasc Diabetol
2015
;
14
:
133
[PubMed]
29.
Bartnik
M
,
Rydén
L
,
Ferrari
R
, et al.;
Euro Heart Survey Investigators
.
The prevalence of abnormal glucose regulation in patients with coronary artery disease across Europe. The Euro Heart Survey on diabetes and the heart
.
Eur Heart J
2004
;
25
:
1880
1890
[PubMed]
30.
Bartnik
M
,
Malmberg
K
,
Hamsten
A
, et al
.
Abnormal glucose tolerance--a common risk factor in patients with acute myocardial infarction in comparison with population-based controls
.
J Intern Med
2004
;
256
:
288
297
[PubMed]
31.
Borrell
LNC
,
Kunzel
C
,
Lamster
I
,
Lalla
E
.
Diabetes in the dental office: using NHANES III to estimate the probability of undiagnosed disease
.
J Periodontal Res
2007
;
42
:
559
565
[PubMed]
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.

Supplementary data