Clarithromycin reduces recurrent cardiovascular events in subjects without periodontitis

Transcription

1 Atherosclerosis 188 (2006) Clarithromycin reduces recurrent cardiovascular events in subjects without periodontitis Susanna Paju a, Pirkko J. Pussinen a, Juha Sinisalo b, Kimmo Mattila c, Başak Doğan a,d, Jari Ahlberg a, Ville Valtonen c, Markku S. Nieminen b, Sirkka Asikainen e, a Institute of Dentistry, University of Helsinki, and Department of Oral and Maxillofacial Diseases, Helsinki University Central Hospital, Helsinki, Finland b Department of Medicine, Division of Cardiology, Helsinki University Central Hospital, Helsinki, Finland c Department of Medicine, Division of Infectious Diseases, Helsinki University Central Hospital, Helsinki, Finland d Department of Basic Sciences, Faculty of Health Education, Marmara University, Istanbul, Turkey e Institute of Odontology, Division of Oral Microbiology, Umeå University, Tandlakarhogskolan, SE Umeå, Sweden Received 20 May 2005; received in revised form 13 October 2005; accepted 7 November 2005 Available online 4 January 2006 Abstract Inflammation leading to acute coronary syndrome may be triggered by bacteria causing periodontal infection. We investigated if recurrence of cardiovascular events in unstable coronary patients are associated with periodontitis or microbiological/serological markers of it. Periodontitisrelated parameters of 141 patients with acute non-q-wave infarction or unstable angina pectoris, who participated in a double-blind, placebocontrolled study with clarithromycin for 3 months, were adjusted to the occurrence of a recurrent cardiovascular event during a follow-up period (average 519 days). In the age group under 65 years the patients with periodontitis had a univariate odds ratios (OR) 95% confidence intervals (95% CI) of 5.0 ( ) for a recurrent cardiovascular event in comparison with patients without periodontitis. Dental status correlated positively with serum lipopolysaccharide concentrations and combined IgG antibody response to Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis. The end point frequency did not differ between clarithromycin and placebo groups in bacterium-positive, seropositive, or periodontitis patients. Fewer end points in clarithromycin group were seen in bacterium-negative, seronegative, edentulous, and non-periodontitis patients. Periodontitis and edentulousness are associated with recurrent cardiovascular events, especially in younger patients. Long-term clarithromycin therapy seems to be beneficial in prevention of recurrent cardiovascular events in non-periodontitis but not in periodontitis patients Elsevier Ireland Ltd. All rights reserved. Keywords: Inflammation; Coronary disease; Acute coronary syndrome; Periodontitis; Lipopolysaccharide 1. Introduction Inflammation leading to acute coronary syndrome (ACS) may be partly caused by various microorganisms, for example, Chlamydia pneumoniae and periodontal infections have been associated with atherosclerosis and cardiovascular diseases [1,2]. Treatment of these infections with antibiotics is under debate [3,4]. Recently, we reported that a 3-month clarithromycin treatment decreased recurrent cardiovascular Corresponding author. Tel.: ; fax: address: sirkka.asikainen@odont.umu.se (S. Asikainen). (CV) events suggesting contribution of bacterial infections in ACS [5]. Periodontal infections, most commonly occurring as chronic bacterial infections of the gingiva and toothsupporting tissues, are polymicrobial infections which eventually may lead to loss of teeth. These diseases, resulting from the accumulation of bacterial plaque and dental calculus, are often asymptomatic and develop and progress during several years/decades. Soft and hard deposits on teeth may then serve as a persistent reservoir for potential pathogenic bacteria. Professionally conducted mechanical removal of dental plaque and calculus is always required to /$ see front matter 2005 Elsevier Ireland Ltd. All rights reserved. doi: /j.atherosclerosis

2 S. Paju et al. / Atherosclerosis 188 (2006) treat periodontitis. In most severe cases antimicrobial agents can be used in conjunction of mechanical debridement. Currently, however, clarithromycin is not considered the first-choice antibiotics for periodontal treatment. In periodontitis, subgingivally located bacteria, bacterial components and products may enter the blood circulatory system via infected tissues. On the other hand, local inflammation products, such as pro-inflammatory mediators and acute-phase reactants, may give rise to systemic inflammation. Besides the known association between periodontitis and cardiovascular diseases [1,6 10], recent data have additionally revealed a direct association between systemic immune response to periodontal pathogens and coronary heart disease [11] or stroke [12]. However, no data are available on the influence of untreated chronic periodontal infection to the effectiveness of antimicrobial treatment in prevention of recurrent ACS events in cardiovascular patients. The main objective of this study was to estimate the differences in CV outcomes between claritromycin- and placebotreated ACS patients with different periodontal status. The secondary aim was to find out if recurrent CV events are associated with periodontitis or microbiological and/or immunological signs of periodontitis in these patients. We report results on dental status, the presence of two major periodontal pathogens in saliva, the presence of antibodies to these periodontal bacteria in serum, and total bacterial lipopolysaccharide (LPS) concentrations in serum in connection to CV outcomes. 2. Material and methods 2.1. Patients The study population consisted of 141 patients (age [mean ± standard deviations] 64 ± 10 years, range 38 80; females, n = 45; males, n = 96) entering the hospital with acute non-q-wave infarction or unstable angina pectoris. Originally the study comprised 148 patients [5], of whom 7 had to be excluded due to missing data on dental status. The inclusion criteria for recruiting study patients, the symptoms at hospitalization as well as medication, CV status, and pre-existing CV risk factors have been described in detail previously [5]. The study design was approved by the Ethics Committee of Helsinki University Central Hospital, Helsinki, Finland, and an informed consent was obtained from each subject enrolled in the study Study design and cardiovascular end points The study was primarily designed to answer the question if clarithromycin, by affecting infections associated with an increased risk for CVD, will decrease new cardiovascular events. Dental data were collected to find out if periodontal status has a contribution to the outcome of the treatment. Randomization of the study subjects into groups receiving antibiotics or placebo and administration of the antimicrobial treatment have been described previously [5]. Briefly, patients received double-blind treatment with either clarithromycin (Klacid 500 mg OD, Abbott Laboratories) or placebo once a day for 3 months. The observed end points were CV death, myocardial infarction, unstable angina, or ischemic stroke during a follow-up period of the average of 519 days (range days) [5] Oral status The dental status of the patients was assessed from panorama radiographs taken after the admission to the hospital. The dental status included the presence (dentate patients) or absence (edentulous patients) of erupted teeth and the presence or absence of periodontal breakdown. Except for two patients, whose radiographs allowed determining the presence of teeth only, the periodontal status could be determined in all other subjects (n = 139). Patients were divided two ways. Firstly, into three groups: edentulous, without periodontitis (later in the text referred to as non-periodontitis ), and with periodontitis. Secondly, into two groups: patients with no current periodontal infection (including edentulous and non-periodontitis patients) and with periodontitis. Periodontal breakdown was recorded when the distance from the cementoenamel junction to the alveolar bone margin was 4 mm. The non-periodontitis patients had no radiographic evidence of periodontal breakdown Bacterial culture and PCR amplification from salivary samples Salivary samples were taken within 48 h of arrival to the hospital, but before the first dose of study treatment. Paraffin-stimulated saliva was collected for 3 min to a sterile glass tube. An aliquot of 0.3 ml of saliva was transferred to vials containing 2 ml of VMGA III transport medium, and processed for bacterial culture within 24 h. Two major periodontal pathogens, Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis, were cultivated and identified as previously described [13]. Saliva sample in VMGA III medium was diluted to 1:10 and stored at 70 C until DNA was extracted using Chelex 100 resin (Bio-Rad, Helsinki, Finland) [14], and A. actinomycetemcomitans and P. gingivalis were PCR-amplified using species-specific primers as previously described [15]. Chromosomal DNA isolated from A. actinomycetemcomitans ATCC and P. gingivalis W50 strains were used as positive controls and sterile water as negative controls in each series of PCR reactions Serum analysis Serum samples were taken within 48 h of arrival to the hospital, but before the first dose of study treatment. Serum IgG

3 414 S. Paju et al. / Atherosclerosis 188 (2006) and IgA antibody levels to A. actinomycetemcomitans and P. gingivalis were determined with multiserotype ELISA [16]. The subjects were considered IgG-seropositive for A. actinomycetemcomitans and P. gingivalis when the corresponding antibody value was 5.0 ELISA units (EU), and IgAseropositive when the antibody value was 2.0 EU, which represent the mean antibody levels plus 1.5 S.D. of periodontally healthy subjects [16]. Combined antibody response (IgG-class antibody levels to A. actinomycetemcomitans plus P. gingivalis) was considered high when the value was 14.0 EU, which represents the corresponding mean antibody level of periodontally healthy subjects plus 3 S.D. [11,16]. Quantitative measurements of total concentration of LPS in serum was done using Limulus Amoebocyte Lysate assay with a chromogenic substrate (HyCult Biotechnology B.V., The Netherlands) on diluted (1:5, v/v, in endotoxin-free water) samples Statistical analysis Differences in continuous and categorical variables were examined with Mann Whitney U-test and χ 2 -test, respectively. The level of statistical significance was defined as p < Odds ratios (OR) and the corresponding 95% confidence intervals (95% CI) for CV end points were analyzed using a logistic regression model. The models were adjusted for age, gender, current smoking, body mass index, diabetes, and hypercholesterolemia (total cholesterol 5.0 mmol/l or the use of statins), and hypertension (systolic blood pressure 150 mmhg, diastolic blood pressure 90 mmhg, or the use of a beta blocker, a calcium channel blocker, or an ACE inhibitor). Cumulative survival of CV end points in groups of subjects with different oral status was estimated using Cox regression models. The statistical analyses were performed by SPSS Program Version 11.0 for Windows. 3. Results 3.1. Baseline characteristics Table 1 shows results on general and dental variables of 141 patients at baseline. Most (n = 107, 76%) patients were dentate. Based on radiographic alveolar bone loss consistent to periodontal breakdown, periodontitis was found in 54% of all patients and in 71% of dentate patients. No significant differences were seen in any variables between the patient groups assigned to medication with clarithromycin (n = 70) or placebo (n = 71) (Table 1). Table 1 Baseline characteristics of patients (n = 141) at baseline and distribution into groups receiving clarithromycin (n = 70) or placebo (n = 71) All patients Clarithromycin Placebo Mean (S.D.) n Mean (S.D.) n Mean (S.D.) n Age (years) 63.8 (10.1) (9.7) (10.6) 71 Number of teeth per patient 15.1 (11.2) 15.3 (11.4) 14.8 (11.0) Proportion (%) n Proportion (%) n Proportion (%) n Women Current smokers Dental status Edentulous Dentate Non-periodontitis Periodontitis Not known Detection a of salivary A. actinomycetemcomitans P. gingivalis Subjects seropositive b for A. actinomycetemcomitans IgG A. actinomycetemcomitans IgA P. gingivalis IgG P. gingivalis IgA High combined antibody response c Differences between the two treatment groups were not of statistical or clinical significance. a Data available from salivary samples of all patients (n = 140), and clarithromycin (n = 69), and placebo (n = 71) groups. b Seropositive, serum IgG antibody value 5.0 EU, or serum IgA antibody value 2.0 EU. Number of all patients, patients in clarithromycin and placebo groups for IgG measurements: 131, 64, and 67, respectively, and for IgA measurements: 130, 64, and 66, respectively. c High combined antibody response, serum IgG antibodies to A. actinomycetemcomitans plus antibodies to P. gingivalis 14.0 EU [11].

4 S. Paju et al. / Atherosclerosis 188 (2006) Table 2 Association of periodontal status with recurrent cardiovascular end points Patients With end points, n (%) p OR (95% CI) Edentulous (n = 34) 9 (26) ( ) a Dentate (n = 107) 33 (31) ( ) b No current periodontal infection c (n = 63) 15 (23) ( ) a Current periodontitis (n = 76) 26 (34) ( ) b Dentate, non-periodontitis (n = 29) 6 (21) ( ) a Dentate, periodontitis (n = 76) 26 (34) ( ) b Younger than the median age of 65 years Dentate, non-periodontitis (n = 20) 2 (10) ( ) a Dentate, periodontitis (n = 42) 15 (36) ( ) b a Univariate logistic regression. b Multivariate logistic regression, adjusted for age, gender, current smoking, body mass index, diabetes, hypercholesterolemia, and hypertension. c Includes edentulous and non-periodontitis patients Association of radiographic periodontal status with recurrent cardiovascular events During the follow-up period, 34% of patients with periodontitis, 21% without periodontitis, and 26% of edentulous patients met a CV end point (Table 2, Fig. 1). When all patients were analyzed, patients with periodontitis experienced recurrent CV events more frequently than patients with no current periodontal infection or dentate nonperiodontitis patients, but the differences were not statistically significant (Table 2). Periodontitis was a significant risk factor for an end point in dentate patients under the median age of 65 years: when compared with non-periodontitis patients, the periodontitis patients had a univariate OR (95% CI) of 5.0 ( ) for a recurrent cardiovascular event (Table 2). In dentate patients younger than 65 years of age, the cumulative event-free survival rate was 90% in nonperiodontitis, 64% in periodontitis, and 50% in edentulous patients (p for trend 0.092) (Fig. 1). In patients with periodontitis, the end point frequency did not differ (p = 0.333) between clarithromycin and placebo groups, whereas a trend towards fewer end points in clarithromycin group was noted in edentulous (p = 0.052) and non-periodontitis (p = 0.082) groups (Table 3). In the clarithromycin treatment group, cardiovascular end points occurred significantly more frequently (p = 0.048) in patients with periodontitis (29%) than in patients with no current periodontal infection (10%) (includes edentulous and nonperiodontitis patients with CV end points: 3/31, Table 3). The same trend was seen within clarithromycin-treated dentate patients: periodontitis patients (29%) had end points more often (p = 0.081) than non-periodontitis patients (7%) (Table 3). In the clarithromycin group, the cumulative eventfree survival rate was higher (p = 0.067) in patients with no current periodontal infection (90%) than in patients with periodontitis (71%). No significant differences were observed in placebo group in the occurrence of end points nor survival rates between edentulous and dentate, between no current periodontal infection and periodontitis, or between dentate non-periodontitis and periodontitis patients (data not shown) Association of two major periodontal pathogens with recurrent cardiovascular events A. actinomycetemcomitans and P. gingivalis were found in saliva of 15 and 31% of all patients, 16 and 39% of Fig. 1. Cumulative survival without a CV event for days to end point in groups of edentulous, non-periodontitis, and periodontitis patients. The results were estimated using Cox regression model. (A) Results on all patients (n = 139). The cumulative event-free survival rate was 79% in non-periodontitis, 74% in edentulous, and 66% in periodontitis patients (p for trend 0.4). (B) Results on dentate patients younger than 65 years of age (n = 69). The cumulative event-free survival rate was 90% in non-periodontitis, 64% in periodontitis, and 50% in edentulous patients (p for trend 0.092).

5 416 S. Paju et al. / Atherosclerosis 188 (2006) Table 3 The effect of clarithromycin on the occurrence of recurrent cardiovascular events in edentulous (n = 34), non-periodontitis (n = 29), periodontitis (n = 76) patients, and all patients (n = 139) Periodontal status Patients with CV end point, n (%) No. of patients out of all patients in the group (%) p Clarithromycin Edentulous 9 (26) 2/17 (12) 7/17 (41) Dentate, non-periodontitis 6 (21) 1/14 (7) 5/15 (33) Dentate, periodontitis 26 (34) 11/38 (29) 15/38 (39) All patients 41 (29) 14/70 (20) 27/69 (39) Placebo dentate patients, and of 12 and 6% of edentulous patients, respectively. In the group of all patients, subjets culture/pcrnegative for A. actinomycetemcomitans, P. gingivalis, or both, met less frequently recurrent cardiovascular end points in the clarithromycin than in the placebo group (Table 4). The same trend was noted in dentate patients, but the differences were not statistically significant (bacterium-negative for A. actinomycetemcomitans, p = 0.061; P. gingivalis, p = 0.056; both, p = 0.096). No such differences were observed in culture/pcr-positive patients Association of antibody levels to periodontal pathogens with recurrent cardiovascular events IgG antibody levels to P. gingivalis, A. actinomycetemcomitans, and combined IgG antibody response to both bac- Table 4 Differences in meeting a recurrent cardiovascular end point between clarithromycin and placebo groups of bacteria-positive/negative and seropositive/seronegative patients (n = 141, includes edentulous and dentate patients) No. of patients with cardiovascular end points/no. of all patients in the group, n (%) Clarithromycin Placebo p Presence in saliva by culture and/or PCR A. actinomycetemcomitans a Yes 1/8 (13) 3/13 (23) No 13/62 (21) 25/58 (43) P. gingivalis Yes 4/24 (17) 4/21 (19) No 10/46 (22) 24/50 (48) A. actinomycetemcomitans or P. gingivalis or both Yes 4/27 (15) 6/27 (22) No 10/42 (24) 22/44 (50) Serological findings A. actinomycetemcomitans IgG Positive b 0/8 (0) 2/8 (25) Negative 13/56 (23) 26/59 (44) A. actinomycetemcomitans IgA Positive 3/8 (38) 5/11 (45) Negative 9/56 (16) 22/55 (40) P. gingivalis IgG Positive 4/32 (13) 9/30 (30) Negative 8/31 (26) 19/37 (51) P. gingivalis IgA Positive 5/21 (24) 8/26 (31) Negative 7/43 (16) 19/40 (48) A. actinomycetemcomitans IgG and/or P. gingivalis IgG Positive c 4/34 (12) 10/34 (29) Negative 8/29 (28) 18/33 (55) A. actinomycetemcomitans IgA and/or P. gingivalis IgA Positive 6/26 (23) 10/31 (32) Negative 6/38 (16) 17/35 (49) a Number of patients for bacterial cultures/pcr amplification from saliva was 140, for IgG measurements 131, and for IgA measurements 130. b Seropositive, serum IgG antibody value 5.0 EU or serum IgA antibody value 2.0 EU [11]. c Positive, patients seropositive to one or both bacteria. Negative, patients seronegative to both bacteria.

6 S. Paju et al. / Atherosclerosis 188 (2006) Fig. 2. The mean of serum LPS concentrations and combined IgG antibody response to A. actinomycetemcomitans and P. gingivalis [11] in groups of different periodontal status. Serum total-lps concentrations (r = 0.165, p = 0.061) and combined IgG antibody response to P. gingivalis and A. actinomycetemcomitans (r = 0.351, p = 0.000) correlated positively with dentate groups. teria correlated positively with periodontal status (r = 0.306, p = 0.000; r = 0.230, p = 0.009; r = 0.351, p = 0.000, respectively) by being lowest in the edentulous group, higher in the non-periodontitis dentate patients, and highest in the periodontitis patients (Fig. 2). IgA antibody levels to P. gingivalis displayed the same trend (r = 0.166, p = 0.061) but non-periodontitis patients had the lowest levels, edentulous higher, and periodontitis patients the highest antibody levels. Patients seronegative to A. actinomycetemcomitans and P. gingivalis met less frequently end points in clarithromycin than in placebo group (Table 4). Differences were significant for both antibody classes for both pathogens when all patients were analyzed. No such differences were observed in seropositive patients. When only dentate patients (n = 107) were analyzed, patients IgA-seronegative to A. actinomycetemcomitans (18% versus 39%, p = 0.032), P. gingivalis (19% versus 47%, p = 0.023), or both (19% versus 48%, p = 0.029), experienced significantly less frequently cardiovascular end points in clarithromycin than in placebo group Association of serum total-lps concentrations with recurrent cardiovascular events Mean LPS concentrations in serum were significantly higher in patients with periodontitis than with no current periodontal infection (p = 0.013). The concentrations were also higher in dentate than edentulous patients (p = 0.098) as well as in periodontitis patients than in dentate nonperiodontitis patients (p = 0.087). Serum total-lps concentrations correlated positively (r = 0.165, p = 0.061) with dentate groups (Fig. 2). Mean ± S.D. of LPS concentrations were significantly higher in P. gingivalis IgG-seropositive than -seronegative patients: in all seropositive (0.78 ± 2.17 ng/ml) versus seronegative (0.44 ± 1.11 ng/ml) patients (p = 0.042), and in dentate seropositive (0.90 ± 2.35 ng/ml) versus seronegative (0.52 ± 1.33 ng/ml) patients (p = 0.031). The same significant difference was noted in mean LPS concentrations between patients IgG-seropositive for either A. actinomycetemcomitans and/or P. gingivalis when compared with IgG-seronegative to both bacteria: in all seropositive (0.76 ± 2.08 ng/ml) versus seronegative (0.41 ± 1.16 ng/ml) patients (p = 0.033), and in dentate seropositive (0.86 ± 2.24 ng/ml) versus seronegative (0.50 ± 1.42 ng/ml) patients (p = 0.023). No such differences were observed between patients seropositive and seronegative to A. actinomycetemcomitans IgG or IgA, P. gingivalis IgA, or A. actinomycetemcomitans and/or P. gingivalis IgA. Mean ± S.D. of serum LPS concentrations tended to be higher in patients with CV events than in patients without events: in all patient group with a CV event, 0.89 ± 2.90 ng/ml, versus without, 0.47 ± 0.69 ng/ml (p = 0.362), and in dentate patients with an event, 1.10 ± 3.27 ng/ml, versus without, 0.54 ± 0.77 ng/ml (p = 0.336). 4. Discussion As reported earlier, a 3-month clarithromycin treatment may be beneficial in prevention of recurrent cardiovascular events in ACS patients [5]. Based on the present results, this beneficial effect seems to be limited to individuals who do not have periodontitis or microbiological or immunological signs of periodontitis. We have shown earlier that elevated serum IgA antibody levels for a major periodontal pathogen P. gingivalis predicts future stroke incidence in patients with a history of CAD [12]. Presently, we report for the first time that periodontitis, defined by radiographic, microbiological, and/or serological means, may contribute to recurrent cardiovascular events and to the efficacy of a preventive antibiotic treatment of these recurrent events. Since the first report in 1989 [1], the association between periodontitis and CAD has been extensively studied [6 11]. In the present study, the cumulative CV event-free survival rate was the highest in non-periodontitis patients, lower in edentulous, and the lowest in periodontitis patients. Our results support the idea that periodontitis increases the risk of recurrent CV events, which was not found in an earlier study by Hujoel et al. [17]. The difference between these two studies may be due to the inclusion criteria for the pre-existing CAD state; we included only unstable CAD patients and Hujoel et al. [17] included both stable and unstable patients. We also observed that periodontitis was a significant risk factor for a CV end point in dentate patients under 65 years of age. These findings are in accordance with previous studies [8,10] suggesting that the correlation between periodontitis and CAD is age-dependent; the infection as a true CAD risk factor can be seen in younger population. Edentulous patients had lower survival rates than non-periodontitis patients supporting earlier studies suggesting that edentulousness is a risk factor for a CV event[11,18]. Edentulous patients may be former periodontitis patients and their teeth may have been extracted due to past periodontitis [6]. On the other hand, their teeth may have been extracted due to dental caries or for prosthetic indications, which contributed to the majority of tooth loss in Finland 2 decades ago [19]. However, edentulous patients may have high salivary bacterial counts and mucosal lesions

7 418 S. Paju et al. / Atherosclerosis 188 (2006) [18] and therefore harbor an inflammatory stage in the oral cavity despite absence of teeth and lack of periodontium. Interestingly, alike to our previous study [11], some of the present edentulous subjects had high antibody levels against A. actinomycetemcomitans and P. gingivalis. Several studies describing treatment of coronary patients with antibiotics have been published in the past few years. Discrepant results have been shown, and the use of antibiotics in coronary artery disease is under debate (for a review, see Danesh [3], Gelfand and Cannon [4]). Nearly in all of these studies the target pathogen for antibiotic medication was C. pneumoniae. To our knowledge, no information on the dental/periodontal status of the patients in the other antibiotic studies than ours is available. Periodontal infections, which can be found in up to 20% of adults in severe form, may eventually lead to loss of teeth if not treated properly and in time. Even though periodontitis is a bacterial infection, consisting mainly of anaerobic bacterial species, antimicrobial medication is neither the first nor only choice for treatment. Mechanical removal of plaque and dental calculus by scaling and root planing is required to disturb the balance of multispecies biofilm in deepened gingival pockets. Antimicrobial agents, however, can be used in advanced cases of periodontitis in conjunction of mechanical debridement [20]. Clarithromycin was originally chosen to this study as an anti-chlamydia agent, and currently is not a preferred antimicrobial agent for periodontal treatment. Even though both A. actinomycetemcomitans and P. gingivalis are susceptible to clarithromycin in vitro [21,22], a 7-day treatment with clarithromycin caused no significant changes in the anaerobic oropharyngeal microflora of healthy subjects [23]. Single-species biofilm formation by Mycobacterium avium in vitro may be hindered by clarithromycin, but this antimicrobial agent was shown to have no activity against established biofilm [24]. We assume that clarithromycin alone, without mechanical periodontal treatment, does not disturb the balance in established dental biofilm and therefore did not beneficially affect the periodontal conditions in the present study population. Clarithromycin is also anti-inflammatory which may have a beneficial effect [25], although in a placebo-controlled study a slow-release clarithromycin (500 mg) did not influence the levels of inflammatory markers [26]. In the present study, clarithromycin was found efficient in preventing recurrent cardiovascular events in nonperiodontitis patients, patients bacterium-negative to A. actinomycetemcomitans and P. gingivalis, and patients seronegative to IgG and IgA antibodies to these two periodontal pathogens. Although our study population was not originally randomized to be treated based on their periodontal status, the results suggest that patients with no radiographic, microbiological, and/or immunological signs of periodontitis benefit from the long-term antimicrobial treatment, in contrast to periodontitis patients. This also supports the hypothesis that periodontitis serves as a persistent background infection constantly releasing bacteria and bacterial products into the blood stream, and therefore, no beneficial effect of antimicrobial medication was seen. It has been shown earlier that periodontitis increases several systemic markers, such as C-reactive protein (CRP), fibrinogen, and cytokines that are related to cardiovascular diseases [27], and that periodontal treatment decreases CRP-levels [28]. We have no data on the possible dental procedures that the present patients may have undergone during the followup period, but since septicemia can be detected not only after dental procedures but also after daily oral hygiene practices, a long-term bacterial release into the blood circulation in periodontitis patients is likely to occur. Gentle mastication increases the release of bacterial endotoxins into the peripheral blood even in periodontally healthy persons and the release is significantly higher in patients with periodontal disease [29]. Our results support this finding since both serum total-lps and combined IgG antibody response to A. actinomycetemcomitans and P. gingivalis correlate with dental status, the lowest values being in edentulous patients, the middle values in non-periodontitis and highest in periodontitis patients. The combined antibody response reflects well the periodontal status, as shown earlier [11,16]. On the contrary, seropositivity displays current (IgA) or previous (IgG) exposure to the pathogens. Therefore, measuring the systemic antibody levels to periodontal pathogens proved to be a sensitive method for subgrouping the patients into responders and non-responders to the clarithromycin treatment. In conclusion, periodontitis and edentulousness are associated with recurrent cardiovascular events, especially in younger cardiovascular patients. Long-term clarithromycin treatment may be beneficial in prevention of recurrent cardiovascular events in non-periodontitis but not in periodontitis patients. Acknowledgements We thank Ms. Tiina Karvonen for excellent technical assistance. This study was supported by grants from the Academy of Finland (Nos , , 77613, and 75953), the Aarno Koskelo Foundation, the Finnish Foundation for Cardiovascular Research, the Finnish Dental Association, and the Finnish Dental Society Apollonia. References [1] Mattila KJ, Nieminen MS, Valtonen VV, et al. Association between dental health and acute myocardial infarction. BMJ 1989;298: [2] Saikku P, Leinonen M, Mattila K, et al. Serological evidence of an association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet 1988;2: [3] Danesh J. Antibiotics in the prevention of heart attacks. Lancet 2005;365: [4] Gelfand EV, Cannon CP. Antibiotics for secondary prevention of coronary artery disease: an ACES hypothesis but we need to PROVE IT. Am Heart J 2004;147:202 9.

8 S. Paju et al. / Atherosclerosis 188 (2006) [5] Sinisalo J, Mattila K, Valtonen V, et al. Effect of 3 months of antimicrobial treatment with clarithromycin in acute non-q-wave coronary syndrome. Circulation 2002;105: [6] Desvarieux M, Demmer RT, Rundek T, et al. Relationship between periodontal disease, tooth loss, and carotid artery plaque: the Oral Infections and Vascular Disease Epidemiology Study (INVEST). Stroke 2003;34: [7] Desvarieux M, Demmer RT, Rundek T, et al. Periodontal microbiota and carotid intima-media thickness: the Oral Infections and Vascular Disease Epidemiology Study (INVEST). Circulation 2005;111: [8] DeStefano F, Anda RF, Kahn HS, Williamson DF, Russell CM. Dental disease and risk of coronary heart disease and mortality. BMJ 1993;306: [9] Scannapieco FA, Bush RB, Paju S. Associations between periodontal disease and risk for atherosclerosis, cardiovascular disease, and stroke: a systematic review. Ann Periodontol 2003;8: [10] Mattila KJ, Asikainen S, Wolf J, et al. Age, dental infections, and coronary heart disease. J Dent Res 2000;79: [11] Pussinen PJ, Jousilahti P, Alfthan G, et al. Antibodies to periodontal pathogens are associated with coronary heart disease. Arterioscler Thromb Vasc Biol 2003;23: [12] Pussinen PJ, Alfthan G, Rissanen H, et al. Antibodies to periodontal pathogens and stroke risk. Stroke 2004;35: [13] von Troil-Linden B, Torkko H, Alaluusua S, et al. Periodontal findings in spouses: a clinical, radiographic and microbiological study. J Clin Periodontol 1995;22:93 9. [14] Mättö J, Saarela M, Alaluusua S, et al. Detection of Porphyromonas gingivalis from saliva by PCR by using a simple sample-processing method. J Clin Microbiol 1998;36: [15] Ashimoto A, Chen C, Bakker I, Slots J. Polymerase chain reaction detection of 8 putative periodontal pathogens in subgingival plaque of gingivitis and advanced periodontitis lesions. Oral Microbiol Immunol 1996;11: [16] Pussinen PJ, Vilkuna-Rautiainen T, Alfthan G, Mattila K, Asikainen S. Multiserotype enzyme-linked immunosorbent assay as a diagnostic aid for periodontitis in large-scale studies. J Clin Microbiol 2002;40: [17] Hujoel PP, Drangsholt M, Spiekerman C, DeRouen TA. Pre-existing cardiovascular disease and periodontitis: a follow-up study. J Dent Res 2002;81: [18] Ajwani S, Mattila KJ, Närhi TO, Tilvis RS, Ainamo A. Oral health status C-reactive protein and mortality a 10 year follow-up study. Gerodontology 2003;20: [19] Ainamo J, Sarkki L, Kuhalampi ML, Palolampi L, Piirto O. The frequency of periodontal extractions in Finland. Community Dent Health 1984;1: [20] Haffajee AD, Socransky SS, Gunsolley JC. Systemic antiinfective periodontal therapy: a systematic review. Ann Periodontol 2003;8: [21] Goldstein EJ, Citron DM, Merriam CV, Warren Y, Tyrrell K. Activities of telithromycin (HMR 3647, RU 66647) compared to those of erythromycin, azithromycin, clarithromycin, roxithromycin, and other antimicrobial agents against unusual anaerobes. Antimicrob Agents Chemother 1999;43: [22] Piccolomini R, Catamo G, Di Bonaventura G. Bacteriostatic and bactericidal in vitro activities of clarithromycin and erythromycin against periodontopathic Actinobacillus actinomycetemcomitans. Antimicrob Agents Chemother 1998;42: [23] Beyer G, Hiemer-Bau M, Ziege S, et al. Impact of moxifloxacin versus clarithromycin on normal oropharyngeal microflora. Eur J Clin Microbiol Infect Dis 2000;19: [24] Carter G, Young LS, Bermudez LE. A subinhibitory concentration of clarithromycin inhibits Mycobacterium avium biofilm formation. Antimicrob Agents Chemother 2004;48: [25] Labro MT. Anti-inflammatory activity of macrolides: a new therapeutic potential? J Antimicrob Chemother 1998;41(Suppl. B): [26] Berg HF, Maraha B, Scheffer GJ, Peeters MF, Kluytmans JA. Effect of clarithromycin on inflammatory markers in patients with atherosclerosis. Clin Diagn Lab Immunol 2003;10: [27] Noack B, Genco RJ, Trevisan M, et al. Periodontal infections contribute to elevated systemic C-reactive protein level. J Periodontol 2001;72: [28] Mattila K, Vesanen M, Valtonen V, et al. Effect of treating periodontitis on C-reactive protein levels: a pilot study. BMC Infect Dis 2002; 2:30. [29] Geerts SO, Nys M, De MP, et al. Systemic release of endotoxins induced by gentle mastication: association with periodontitis severity. J Periodontol 2002;73:73 8.