Articles
Palle HOLMSTRUP * Nils-Erik FIEHN **
*University of Copenhagen
School of Dentistry
Copenhagen, Denmark
The impact of several systemic diseases on the periodontium is well established. These diseases include diabetes, neutrophil deficiences, neutropenia, immunodeficiences and others (Kinane and Davies, 1990 ; Holmstrup and Westergaard, 1998). Much less is known about the influence of periodontal disease on systemic health. In the past decade, however, a number of studies have suggested that general...
The influence of a number of systemic diseases on the periodontal tissues is well-established whereas the impact of periodontal disease on general health is less approved. In the past decade, however, the pathogenesis of a number of systemic diseases have been associated with periodontal infection with varying scientific evidence. These diseases include cardiovascular disease, infective endocarditis, respiratory infections, rheumatoid arthritis, diabetes mellitus and brain abscess. The purpose of this article is to review the most recent evidence to demonstrate the influence of periodontal disease on systemic diseases with emphasis on cardiovascular disease, infective endocarditis and respiratory infection.
It is concluded that an association between periodontal infection and cardiovascular disease has been documented in several epidemiologic studies, but more research is needed to confirm and understand such an association. Infective endocarditis has been shown in some patients to be caused by dental plaque bacteria. Associations between respiratory infection and periodontal disease have been found but remains to be firmly established in future studies, and the possible involvement of periodontal infection in the pathogenesis and course of diabetes mellitus, rheumatoid arthritis and brain abscess is obvious but should be further documented to firmly conclude about its existence and pathogenic background.
The impact of several systemic diseases on the periodontium is well established. These diseases include diabetes, neutrophil deficiences, neutropenia, immunodeficiences and others (Kinane and Davies, 1990 ; Holmstrup and Westergaard, 1998). Much less is known about the influence of periodontal disease on systemic health. In the past decade, however, a number of studies have suggested that general health may be more affected by periodontal disease than previously recognized (Seymour and Steele, 1998 ; Research, Science and Therapy Committee of the American Academy of Periodontology, 1998). The present review presents data on the influence of periodontal disease on a number of systemic diseases with emphasis on cardiovascular disease (CVD), infective endocarditis (IE) and respiratory infections (RI). If important life-threatening diseases are induced by periodontal disease, obviously periodontal treatment is even more important than hitherto approved.
Common CVD is associated with atherosclerosis which results in coronary thrombosis, ischemic heart disease and stroke. These are the major causes of death in the Western World (Becker, 1995 ; Whelton, 1995 ; Braunwald, 1997 ; Breslow, 1997 ; Harvard School of Public Health, 1997). Atherosclerosis is caused by the formation of atheromatous plaques within affected large and middle sized arteries. These plaques contain cholesterol and other lipids as well as necrotic cells, fibrin and fibrinogen, which are the source of thrombi that can either occlude the vessel of their origin or after release occlude other vessels resulting in infarction at remote sites.
A number of studies have associated periodontitis with CVD on the basis of epidemiologic findings. Four of these studies are Finnish. An early study compared the general dental status (as evaluated on the basis of number of carious lesions, probing pocket depths, periapical lesions and pericoronitis) of 100 patients who had recently suffered from acute myocardial infarction with age- and gender-matched controls. Registration of other risk factors included serum cholesterol, triglyceride, high density lipoprotein cholesterol concentration, hypertension, other infections, smoking habits and socioeconomic status (Mattila et al., 1989). In the acute myocardial infarction group the general dental health status was significantly lower than in the control group. A regression analysis showed that this association was independent of total cholesterol, triglycerides, hypertension, presence of diabetes, smoking and age.
In another Finnish study of 12 women and 88 men with coronary heart disease (CHD) radiographic examination of the oral cavity and coronary angiography showed significant correlation between the presence of periodontal/periapical infection and the degree of coronary atherosclerosis in males (Mattila et al., 1993). In multivariate analysis, other factors of CHD were taken into account, including lipid profile, body mass, hypertension, smoking, age and socioeconomic status, and there was an independent association between dental infections and the severity of coronary atherosclerosis.
A cross-sectional study (Paunio et al., 1993), which also originates from Finland, including 1 384 men aged 45-64 years showed a significant association of missing teeth with coronary heart disease. In a multivariate model for CHD which included smoking and other variables, smoking was squeezed as a risk factor when missing teeth was entered.
A fourth study from Finland is a 7 year follow-up study of 214 patients (182 males and 32 females) with coronary artery disease. In this study dental health was a significant predictor of new coronary problems. Other common risk factors for CHD were controlled for and additional risk factors included number of previous infarctions and a history of diabetes (Mattila et al., 1995).
Further evidence to support a relationship between dental disease and CHD has been added by a population based follow-up study including 9 760 individuals aged 25-74 years (De Stefano et al., 1993). After baseline dental and medical examination these individuals were followed for 14 years. When adjusting for other known risk factors, individuals with periodontitis showed a 25 % increased risk as compared to those with no or minimal periodontitis. However, periodontitis increased the risk of CHD by 70 % in males aged 25-49 years. Poor oral hygiene was also associated with increased risk for CHD.
In another follow-up study (Beck et al., 1996), which included 1 147 males followed for a mean of 18 years, it was shown that 18 % of the individuals developed CHD-type events including angina pectoris, non-fatal myocardial infarct and CHD-related death. Patients with more than half of the teeth with probing depths over 3 mm showed a 2-fold increase in CHD-risk after adjusting for other CHD-risks. Patients with more than 3 mm probing depths of all teeth showed an adjusted CHD odds ratio of 3.1. Subsequent strokes in patients with more than 20 % bone loss at baseline showed an odds ratio of 2.8 after adjustment for other factors.
A follow-up study of 44 119 male health professionals followed for 6 years showed a positive association between CVD and loss of teeth due to periodontal disease, the relative risk being 1.67 after adjusting for the common CVD risk factors (Joshipura et al., 1996).
The reviewed studies support an association of increased CHD, particularly in men younger than 60 years of age, but the epidemiologic findings primarily refer to studies with a dominance of males. In addition, some of the studies were not designed to determine the role of periodontal disease, but tooth loss and other data were used to characterize the oral condition. When CHD was not related specifically to periodontitis, the role of this disease is obviously unknown. Although some of the studies mentioned appear to control for some commonly accepted risk factors of CVD, others are not examined as confounding factors (Genco et al., 1999). It is, consequently, necessary to be cautious about interpreting the role of oral health. As further stated by Seymour and Steele (1998), there is a risk that the presence of periodontal disease is just a good way of summarizing a range of dental health and lifestyle variables.
To further substantiate an association between periodontitis and CVD it is important to understand a possible pathogenic mechanism whereby periodontitis can contribute to CVD.
Although such mechanisms are not clearly delineated, several pathways appear to be possible.
Oral bacteria may interact directly in the pathogenesis of CVD. Periodontitis predisposes to an increased incidence of bacteriemia (Silver et al., 1977) which frequently occurs as the result of daily activities as mastication and oral hygiene procedures (Guntheroth, 1984). Thereby, interactions between oral bacteria as the result of bacteriemia and platelets may play a role in CVD. One of the important periodontal bacteria, Porphyromonas gingivalis (P.g.) may induce platelet aggregation and thereby be associated with thrombus formation (Emrich et al., 1991). This finding is important because P.g. has been revealed in coronary and carotid atheromas (Chin, 1999). Furthermore, platelets selectively appear to bind strains of Streptococcus sanguis (S.s.) which is a common component of supragingival plaque, resulting in thrombus formation (Herzberg et al., 1983). A possible role for S.s. in CHD has been further substantiated by experimental studies in rabbits, which had intravenous injections with platelet aggregating or non-aggregating strains of S.s. The platelet aggregating strains resulted in larger heart valve vegetations, more signs of myocardial ischemia and higher mortality (Herzberg et al., 1992). In addition, increased platelet aggregation in response to S.s. injection has been described in rabbits on a high-fat diet (Herzberg and Meyer, 1996).
An indirect pathogenic role of periodontal bacteria may be due to influx of inflammatory cells into the large blood vessels in association with bacteriemia and LPS produced by Gram-negative bacteria. The resulting production of cytokines such as IL-1β and TNF-α, prostaglandins and growth factors including platelet derived growth factor, fibroblast growth factor and granulocyte macrophage colony-stimulating factor may contribute to thickening of the arterial wall because several of these substances are atherogenic (Marcus and Hajjar, 1993).
It is now recognized that the inflammatory response to bacterial challenge shows considerable interindividual variation, and some patients respond to periodontal bacteria or bacterial products with an inflammatory response resulting in high levels of inflammatory mediators (Hernichel-Gorbach et al., 1994 ; Shapira et al., 1994 ; Kornman et al., 1997), which may account for the rapid tissue destruction characteristic of early or rapidly progressive periodontitis. As cytokines appear to be important for the pathogenesis of both atherosclerosis and periodontitis, a specific phenotype (MØ+) characterized by increased monocytic production of IL-1β and TNF-α may link a risk for both destructive periodontitis and CVD (Beck et al., 1996 ; Offenbacher, 1996).
Finally, the inflammatory response associated with periodontitis elevates the levels of acute phase proteins including fibrinogen and C-reactive protein in the general circulation (Shklair et al., 1968 ; Pederson et al., 1995 ; Ebersole et al., 1997). Fibrinogen and C-reactive protein are risk factors for CHD (Danesh et al., 1998 ; Ridker et al., 1998 ; Koenig et al., 1999) and if induced by periodontitis, this may also explain a link between CHD and periodontitis.
At the present stage there is no scientific background for designing a detailed therapeutic strategy to prevent possible CVD-complications of periodontitis other than common periodontal treatment modalities which reduce the bacterial load of the oral cavity.
Infective endocarditis (IE) is an infection of the endothelial lining of the heart (Durack, 1990) with an incidence in the Western World of about 50 cases per million citizens per year (Gutschik and Lippert, 1990). The incidence increases significantly with age in the developed countries, and with an increasing population of elderly people and drug abusers, IE is a disease of increasing medical importance (Hogevik et al., 1995). IE occurs mainly in patients with predisposed heart lesions. High and moderate risk patients are those with a history of IE and rheumatic valvular disease, prosthetic heart valves, and surgical-pulmonary shunts (Kaye and Abrutyn, 1991 ; Scheld and Sand, 1995). However, the disease can also occur in an uncompromized host (Fiehn et al., 1995). Despite the relatively low number of cases, the disease must be regarded as very serious because of a mortality as high as 30-50 % (Yogev et al., 1988 ; Janatuinen et al., 1991 ; Douglas and Cobbs, 1992).
On the basis of blood samples from IE patients the identity of the infecting microorganism can be established. Previous studies have shown that more than 50 % of the cases can be attributed to streptococci, and among those most frequently encountered are viridans streptococci, which are normal inhabitants of the oral cavity, located in dental plaque associated with gingivitis and periodontitis (Bayliss et al., 1983 ; Douglas et al., 1993 ; Kjerulf et al., 1993). Other oral bacterial species such as Actinobacillus actinomycetemcomitans, Eikenella corrodens, and Haemophilus species originating from dental plaque in periodontitis patients have also been observed (Nord and Heimdahl, 1990). Among the viridans streptococci, Streptococcus sanguis is most often found in the blood stream of IE patients. In 1995, Fiehn et al. revealed total identity after ribotyping between blood isolates and dental plaque isolates which was the final proof of a possible oral origin of infecting bacteria in IE.
The initial event in the pathogenesis of IE is the invasion of the blood stream by the microorganisms resulting in bacteremia. Oral bacteria may gain access to the circulation as a result of lost mucosal integrity due to dental therapy e.g. oral surgical operation, subgingival scaling, but also, as discussed above, after mastication and oral hygiene procedures (Silver et al., 1977 ; Guntheroth, 1984 ; Durack, 1995 ; Gutschik, 1995). The circulating bacteria may adhere to the endothelium of the heart if there is a previous damage, or if there, by accident, is a microscopic traumatization of the endothelium, which has resulted in a sterile lesion composed primarily of platelets and fibrin, a so-called non-bacterial thrombotic endocarditis (NBTE) (Durack and Beeson, 1972 ; Jen and Lin, 1991). The bacterial adhesion to the NBTE is favoured, and a multiplication of bacteria will subsequently follow and initiate a focus of intravascular infection (Durack, 1975).
IE is difficult to diagnose and it is often diagnosed late in the disease process. The therapy demands susceptibility testing of the infecting microorganism and appropriate antibiotic treatment for about six weeks (Baldassarre and Kaye, 1992). For risk patients, prevention is vital in connection with dental treatment. The prophylactic procedures include oral rinsing with antimicrobial agents to minimize the microbial burden in the oral cavity, systemic use of antibiotics, and, in particular, minimizing gingival inflammation in risk patients. Most often, the antibiotic prophylaxis consists of one high dosis of amoxicillin just before and, at the latest, one or two hours after the dental treatment is finished (Durack, 1998).
Pneumonia is an infection of the pulmonary parenchyma, caused by bacteria or other infectious agents including viruses, fungi and mycoplasma. Bacterial pneumonia in adults usually derives from aspiration of oropharyngeal bacteria, which cause infection due to insufficient host defence mechanisms. It is thereby obvious, that the oropharyngeal microflora including periodontal bacteria may serve as a potentially important cause of respiratory infection.
Pneumonia is basically classified as community- or hospital-aquired, and the spectrum of bacteria involved in these two types of disease usually differs. In community-aquired bacterial pneumonia the offending bacteria are most often Streptococcus pneumoniae or Haemophilus influenzae, but other species may be involved too (Ostergaard and Anderson, 1993). Community-aquired pneumonia appears to be associated with anaerobic organisms in as much as 21-33 % of the cases (Ries et al., 1974 ; Pollock et al., 1983). Most cases of community-aquired bacterial pneumonia, however, respond successfully to antibiotics.
In contrast, hospital-aquired (nosocomial) bacterial pneumonia, which is one of the most common nosocomial infections, has a high mortality rate, accounting for 19-50 % of the mortality rate of all nosocomial infections (Bartlett et al., 1986 ; Craven et al., 1991). Seriously ill patients are primary targets of nosocomial bacterial pneumonia. The typical infective organisms are Staphylococcus aureus, and facultative or aerobic Gram-negative bacteria, including Enterobacteriaceae, e.g. Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Enterobacter species and others (Johanson et al., 1972). Nosocomial bacterial pneumonias appear to be associated with anaerobic organisms in 35 % of the cases (Bartlett, 1987), including Fusobacterium nucleatum and black pigmented Bacteroides, Actinomyces, Lactobacillus, Peptostreptococcus, Propriobacterium and Veillonella spp.
It has been shown that oropharyngeal colonization with potential respiratory pathogenic organisms increases during hospitalization (Wahlin and Holm, 1988 ; Sedgley and Samaranayake, 1994). Inadequate oral hygiene, which is common in critically ill patients, may be a predisposing factor in respiratory infection because dental plaque may serve as a potential reservoir of respiratory pathogens. In fact, such pathogens have been found in supragingival dental plaque of 65 % of intensive care unit patients as compared to 16 % of control dental clinic patients (Scannapieco et al., 1992 ; Scannapieco and Mylotte, 1996).
Dental plaque from the intensive care unit patients contained S. sanguis, P. aeruginosa and several genera of aerobic Gram-negative bacteria. These organisms are known to be potential respiratory pathogens. Other reports have presented evidence of well-known periodontal pathogens such as Actinobacillus actinomycetemcomitans, Fusobacterium and Capnocytophaga isolated from pneumonia (Lorenz and Weiss, 1994 ; Venkataramani et al., 1994). Another study has provided evidence, that hospitalized dentate patients aquired aspiration pneumonia more often (6 out of 22 according to 27 %) than edentulous patients (0 out of 12), and a similar trend was found in nursing home residents (19 % of dentate versus 7.6 % of edentulous individuals (Terpenning et al., 1993).
Subgingival bacterial samples may also harbour respiratory pathogens. Isolates from deep periodontal pockets have demonstrated the presence of Enterobacteriaceae (Slots et al., 1988 ; Slots et al., 1990) and others have isolated the same organisms in 8 % and S. aureus from 46 % of periodontitis patients (Dahlen and Wikstrom, 1995). In addition to these findings, it has been reported that the offending organisms in respiratory anaerobic infections with considerable mortality due to suppuration most often originate from the subgingival environment (Bartlett, 1987).
The therapeutic consequence of the present knowledge is increased efforts to prevent gingivitis and periodontitis. In particular, this applies to nursing home residents and hospitalized patients.
Rheumatoid arthritis and periodontitis share several characteristics. An inflammatory resorption of bone is the basic clinical feature which is induced after activation of macrophages and osteoclasts (Nair et al., 1996). Another common factor is the involvement of matrix metallo proteinases in the pathogenesis (Page, 1991). Despite the mutual characteristics of the two diseases, there is, however, no unambiguous evidence at the present stage to support a pathogenic association between the two diseases (Malmstrom and Calonius, 1975 ; Iida and Yamaguchi, 1985 ; Laurell et al., 1989 ; Sjostrom et al., 1989 ; Tolo and Jorkjend, 1990 ; Arneberg et al., 1992 ; Yavuzyilmaz et al., 1992 ; Kasser et al., 1997).
An association between diabetes mellitus (DM) and periodontitis is well-established and characterized by increased susceptibility to loss of periodontal attachment in DM patients (Cianciola et al., 1982 ; Shlossman et al., 1990 ; Grossi et al., 1994). The influence of periodontitis on DM is less approved, but a number of studies have demonstrated that treatment of periodontitis may reduce the need for insulin (Williams and Mahan, 1960 ; Miller et al., 1992) and that severe periodontitis is associated with greater number of cardiovascular complications (Thorstensson et al., 1996) and increased risk of poor glycemic control, which may improve after successful periodontal treatment (Miller et al., 1992 ; Grossi et al., 1996 ; Taylor et al., 1996).
On the basis of a number of case reports it is assumed that brain abscess in some instances may be related to oral infections including periodontitis (Goteiner et al., 1982 ; Marks et al., 1988 ; Anderson and Horton, 1990 ; Andrew and Farnham, 1990). In general, the cause of brain abscesses is polymicrobial and frequently associated with Streptococcus mutans and Streptococcus milleri (Marks et al., 1988 ; Andrew and Farnham, 1990). Anaerobic infections with Peptostreptococcus micros, Fusobacterium nucleatum and Bacteroides species are also reported (Murdoch et al., 1988 ; Sammalkorpi, 1989 ; Anderson and Horton, 1990). Despite the isolation of bacteriae suggestive of periodontitis as a background, there is no clear evidence to definitively link the two diseases.
In summary, an association between periodontal infection and CVD has been found in several epidemiologic studies, but it may be explained by established or non-established risk factors that are shared by the two diseases. On the other hand a number of these risk factors are life style factors for which several of the above mentioned epidemiologic studies were controlled. Future additional studies with further detailed control of confounding factors will add to the understanding of a possible relationship between the diseases. Moreover, the disease mechanisms behind an association between CVD and periodontitis need further elucidation to evidence the association. Infective endocarditis has been shown in some instances to be caused by dental plaque bacteria. Several studies have shown data suggestive of bacterial respiratory infection caused by oral bacteria. It has, on the other hand, not been substantially evidenced, that periodontitis significantly increases the risk of respiratory infection. Additional investigations are therefore needed to conclusively demonstrate an increased risk of respiratory infection in periodontitis patients. The pathogenesis of a number of other diseases, including diabetes mellitus, rheumatoid arthritis and brain abscess, has been associated with periodontitis, but more research is necessary to unravel a possible pathogenic interaction in details.
Demande de tirés à part
Palle HOLMOSTRUP, Department of Periodontology, School of Dentistry, University of Copenhagen, 20 Nørre Allé, DK-2200 COPENHAGEN - DENMARK. Fax : 45 353 266 99. E-mail : palle.holmostrup@odonto.ku.dk.