Periodontitis related to cardiovascular events and mortality: a long-time longitudinal study

Periodontitis is a chronic disease with an infectious etiology, causing an inflammatory response resulting in the breakdown of soft and hard tissues around teeth [1]. Severe periodontitis has been identified as the sixth most prevalent disease in the world [2]. In the USA, the prevalence of periodontitis is reported to be 47% [3]. In older individuals, it is even more prevalent [4].

Cardiovascular diseases (CVDs) include all diseases associated with the heart and blood vessels, such as stroke, coronary heart disease, and heart failure [5]. CVDs are the most common causes of death in the USA [6]. Atherosclerosis is considered the leading cause of CVDs [7, 8]. Periodontal infections may cause bacteremia triggering host systemic inflammatory responses and chronic inflammation and related to the pathogenesis of atherosclerosis [9]. Data suggest that periodontitis is associated with subclinical atherosclerosis [7, 8, 10].

Periodontitis has been associated with an increased risk for CVDs [8, 11‐13]. Having periodontitis has been reported to increase the risk for stroke [14‐16]. Periodontal disease has been pointed out as a risk factor for stroke, especially in men and in younger subjects [17]. In a review and meta-analysis of the literature, periodontitis was reported to be associated with the occurrence of stroke [18]. Periodontitis has also been associated with myocardial heart infarction [19, 20]. Individuals diagnosed with the first event of acute myocardial infarction (AMI) were matched with subjects with no evidence of AMI. A clear association was reported between periodontitis and AMI [19]. In another control-matched study by Rydén et al. 2016[20], an increased risk for a first myocardial infarction was reported among individuals with periodontitis. In a review by Dietrich et al. [21], six case-control and cohort epidemiological studies described an increased risk for a first coronary event in individuals with diagnosed periodontitis. In general, epidemiologic data are linking periodontitis to CVDs. Periodontitis is presently believed as a CVD risk factor. Several studies assessing associations between periodontitis and CVDs are however cross-sectional cohort or case-control studies [7, 8, 16]. Few studies with a prospective longitudinal study design have been reported.

Accordingly, the present study aimed to assess if individuals ≥ 60 years of age with periodontitis are more likely to develop stroke or ischemic heart diseases or are at a higher risk of death over 17 years.

The present study identified that over the 17-year follow-up period, periodontitis increased the risk of future incidences of ischemic heart diseases in all individuals, in women, and in the OO age group. In woman with periodontitis, the HR for the incidence of ischemic heart disease was 2.1. Another study has confirmed that after the menopause, women have a higher incidence of AMI compared to age-matched men [22] which is in agreement with the results of the present study in older people. There are few other longitudinal studies concerning periodontitis and incidences of CVDs. In a Danish national register–based cohort study with a follow-up period of 15 years, patients with periodontitis were reported to have an increased risk of CVDs [23]. In the study by Hansen et al. (2013), ICD codes were used to define periodontitis. This classification of periodontitis is different from the one used in the present study. Additionally, the age of the study population was > 18 years, and gender differences were not reported making comparisons between results observed in the present study impossible. A Korean nationwide cohort follow-up study of 7.6 years showed a dose-dependent association with tooth loss and incident myocardial infarction, heart failure, and ischemic stroke, especially in individuals with periodontitis [24]. The circumstances between the study above and our study are not the same. The classification of periodontitis was in the study by Lee et al. [24] not defined, and the study individuals included were from 20 years, and the mean age not mentioned. Also, the included individuals had a history of a CVD event, meaning that they were at an increased risk for a subsequent CVD [25].

Another main finding in this study was that individuals with periodontitis were at a higher risk to die during the 17-year follow-up compared to individuals without a diagnosis of periodontitis. Recently published data have shown that periodontitis increased the risk for all-cause mortality within 15 years [23]. In a 3-year follow-up study, individuals with severe periodontitis developed the combined endpoint (myocardial infarction, stroke/transient ischemic attack, cardiovascular death, and death caused by stroke) more often compared to individuals without periodontitis (18.9% versus 14.2%) [26]. As we were unaware of the exact reasons for death in the present study, such a combined endpoint was not possible to include. In the present study, it was demonstrated that in men and the YO (60–72 years), there was an association between periodontitis and mortality. Young individuals (30–40 years) with periodontitis and missing molars have been reported to have an increased risk for early death over 16 years [27]. Missing teeth, due to periodontal disease, could be a proxy for a previous inflammation and partly explain the identified association with an increased risk for mortality among young-old in the present study that is in line with the results from the study by Lee et al. [24]. In another recent study, ≥ nine missing teeth were also associated with mortality [28]. In the present study, an association between periodontitis and the number of lost teeth at baseline was found. When lost teeth at baseline were used as a proxy for periodontitis, an association between all-cause mortality in all individuals was found. Such an association is in line with the results obtained using periodontitis as the independent variable in the Cox regression analysis. However, lost teeth at baseline were also associated with mortality among women and to an increased risk for ischemic heart diseases in all individuals, in women, and in the YO group. Accordingly, using lost teeth as a proxy for periodontitis does not seem to be an adequate approximation for periodontal disease.

Deaths by CVDs have decreased during the latest 60 years, as a result of preventive care and advances in medicine [29]. Many older individuals are using preventive medications that may delay or even prevent a CVD event from occurring [30].

Recent data reported that the numbers of missing teeth were related to heart failure and myocardial infarction. In contrast, the number of teeth missing was not significantly related to stroke in a longitudinal study with a median follow-up time of 15.8 years [31]. In another study, the same tendency was reported; missing teeth (≥ five missing teeth) were statistically associated with an event of coronary heart disease and acute myocardial infarction. In contrast, tooth loss was not associated with stroke in a 13-year prospective longitudinal study [28].

In the literature, different classifications and parameters for periodontitis have been used. Missing teeth have also been proposed as a proxy for current or past periodontitis as it is considered to reflect an accumulation of oral inflammation [32]. Correctly, if untreated, periodontitis may result in tooth loss and is one of the primary reasons for tooth loss in adulthood [33]. In a recent study by Lee et al. [24], edentulous individuals demonstrated the highest cardiovascular risk. It is, however, difficult to be sure of the reason for missing teeth unless it is reported in the dental records. In the present study, it was not possible to within certainty decide the reason for tooth loss. Alveolar bone loss, as used in the present study as a proxy for periodontal disease, is an indicator of a patient who has had periodontal inflammation during a period.

The prevalence of periodontitis defined by loss of alveolar bone was relatively low (24.7%). The reported prevalence of periodontitis in individuals 65 years and older has been reported to be 70% in the USA and with increasing prevalence with increasing age [3]. Differences in prevalence may be related to the different classifications of periodontitis used. Eke et al. [3], when defining “total periodontitis,” included mild, moderate, and severe stages of periodontitis and used clinical attachment–level loss as an indicator for periodontitis. The definition of periodontitis used in the present study is based on bone loss of ≥ 5 mm from the CEJ to the alveolar bone level on ≥ 30% of sites. The use of the 5-mm level was chosen to account for possible measurement errors. Individuals defined as mild periodontitis cases in the study by Eke et al. [3] were not defined as periodontitis patients in the present study, which to some degree may explain the differences in prevalence figures reported in the two studies. Individuals affected with bone loss in older ages may reflect a long history of periodontitis and accordingly a long time of an inflammatory response.

The new accepted classification of periodontitis includes clinical variables (CAL and probing pocket depth) as well as bone loss in radiographs [34]. The bone loss reflects the accumulated progressive effect of periodontitis over a long time [35, 36]. Clinical parameters partly reflecting the inflammatory activity could be more transient, giving the information of the periodontal status at that specific moment. Studies have reported that the mean proportion of bone loss increased with age, but the proportion of teeth with periodontal pockets remained unmodified. In another study, the alveolar bone loss progressed with age but was limited after the age of 50 [37]. In older individuals, gingival recession is the main reason for attachment loss [38]. Crestal bone height and CAL have shown a good correlation [39]. It has been proven that attachment loss precedes radiographic evidence of crestal alveolar bone loss during periods of periodontal disease activity [40] whereas, over time, these differences seem to level off [39]. In our study, the limit of alveolar bone loss ≥ 5-mm distance from CEJ to marginal bone level reflects a definitive bone loss and such a bone loss is present on ≥ 30% of sites corresponding to a general disease distribution.

One limitation with the present study was that the most fragile and medically compromised individuals were not able to participate. The fact that the most fragile individuals did not participate may have affected the results, possibly lowering the associations between periodontitis and CVDs. The individuals in the present study were 60 years or older. While the follow-up period was long, obviously many died during the study. The causes of death were not known, but the main reasons for death are still CVDs [41]. The consequence of not knowing the reasons for death makes it impossible to include mortality by CVDs in a combined endpoint, among others, which should have to strengthen the associations between periodontitis and CVDs. The risk factors in the Cox regression analysis were adjusted only from the baseline data, which is another limitation in the study.

The long-time follow-up and that CVD events were easy to follow and control is a strength with the present study. It would be interesting to study if well-designed preventive dental programs can influence the incidence of CVDs in long-term studies. Intervention studies are needed to verify a valid link between periodontitis and CVDs.

In conclusion, this study demonstrated that in older adults, periodontitis was a statistical risk indicator for ischemic heart diseases. Over the time studied, periodontitis was significantly associated with mortality. This is the first long-time follow-up report on periodontitis and the incidence of cardiovascular diseases and death.