SEARCH

SEARCH BY CITATION

Keywords:

  • Periodontal disease;
  • cardiovascular disease;
  • serum antibodies;
  • cytokines;
  • Porphyromonas gingivalis;
  • heat shock protein 60 (HSP60)

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Background:  Previous studies have demonstrated variable effects on systemic inflammatory and immune responses following improved periodontal health. This study examined changes in serum levels of the inflammatory mediators IL-1β, IL-6, TNF-α and sICAM-1, and antibodies to Porphyromonas gingivalis, human heat shock protein (hHSP) 60 and P. gingivalis GroEL following improvement in periodontal health in high cardiovascular (CV) risk and low CV-risk patients.

Methods:  Patients retrospectively selected from a longitudinal study, had undergone yearly periodontal examinations and peripheral blood collections. They had demonstrated a quantifiable improvement in periodontal health (>60% reduction in number of sites with probing depth ≥4 mm from the baseline visit) and could be classified as either high CV-risk (≥6 classical risk factors, n = 13) or low CV-risk (≤1 classical risk factor, n = 14). Serum levels of the cytokines and antibodies were measured using ELISA.

Results:  For sICAM-1 and anti-P. gingivalis GroEL and anti-hHSP60 antibodies, most patients recorded decreased levels. Reductions in serum sICAM-1 levels were more notable in low CV-risk patients (p = 0.006); and reductions in levels of anti-P. gingivalis GroEL and anti-hHSP60 antibodies (p = 0.001 and 0.009 respectively) were more notable in high CV-risk patients.

Conclusions:  This study found that subsequent to improved periodontal health, the anti-HSP (HSP60 and GroEL) antibody response was reduced, particularly for high CV-risk patients. sICAM-1 levels were also lowered, more so for low CV-risk patients.


Abbreviations and acronyms:
BMI

body mass index

CHD

coronary heart disease

CRP

C-reactive protein

CV

cardiovascular

CVD

cardiovascular disease

hHSP

human heat shock protein

MI

myocardial infarction

PPD

peridontal probing depth

s-ICAM

soluble intercellular adhesion molecule

TNF-α

tumour necrosis factor-α

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

It is now well established that advanced chronic periodontitis is associated with cardiovascular disease (CVD)1–3 and a number of different mechanisms have been postulated to explain this association. Two of the suggested models are: (i) an increase in systemic inflammation as a result of the periodontal infection; and (ii) cross-reactivity between human heat shock protein 60 (hHSP60) and Porphyromonas gingivalis GroEL (the bacterial homologue of hHSP60).4–6 Elevated systemic levels of inflammatory mediators occur in chronic periodontitis and it has been proposed that this may lead to endothelial cell activation and atherogenesis.7 At the same time antibodies and T cells specific for P. gingivalis GroEL have been shown to exist in the peripheral blood of patients with atherosclerosis, as well as in the atherosclerotic plaques themselves.6 Cross-reactive immune responses between these antibodies and T cells and the hHSP60 which is expressed on endothelial cells in atherosclerosis5,8 may lead to endothelial dysfunction and result in the initiation and progression of atherosclerosis. In addition, high levels of anti-P. gingivalis GroEL antibodies have been found in the serum of patients with chronic periodontitis, compared with healthy controls.7

The effect of periodontal therapy on the level of systemic inflammation and anti-P. gingivalis GroEL/hHSP60 antibodies is still unclear. A number of studies have shown that treatment of chronic periodontitis results in a reduction in levels of systemic inflammation.9–12 Recent studies have further demonstrated that periodontal therapy also significantly reduces serum levels of a number of factors associated with CVD risk including low density lipoproteins, cholesterol, haptoglobin, interleukin (IL)-18 and interferon-γ (INF-γ),13 as well as sE-selectin, soluble intercellular adhesion molecule (s-ICAM) -1, serum amyloid P,12 IL-6 and C-reactive protein (CRP)11 for periods of up to 12 months post-treatment. Periodontal treatment also significantly reduced the levels of anti-P. gingivalis GroEL antibodies but levels of antibodies to hHSP60 remained unchanged.13,14

However, other studies have demonstrated variable effects on the systemic inflammatory and immune response following periodontal therapy. A recent review15 of studies that have examined the effect of periodontal treatment on surrogate markers of CVD reported that while most showed a reduction in at least one marker, these were often different markers and at least one study showed no effect. The heterogeneity of the results may be due, at least in part, to individuals of different cardiovascular (CV) risk being studied. Therefore, the aim of the present study was to test the hypothesis that improvement in periodontal health leads to a reduction in serum levels of inflammatory mediators and antibodies to hHSP60 and P. gingivalis GroEL, and that the level of this reduction varies with CV-risk.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

This was an exploratory subgroup analysis of participants enrolled in a prospective longitudinal study examining the effect of oral health on general health in a low socio-economic region where the population were considered to be at risk of poor oral and general health. Institutional ethics review committee approval was obtained and all 581 participants provided signed consent according to the Helsinki Declaration after being given a written and verbal explanation of the study. Patients received standard non-surgical periodontal therapy which included extractions where necessary. At baseline and one year later, detailed medical histories were taken, probing depths were measured at six sites per tooth and peripheral blood (20 ml) was collected for analysis of serum antibody and cytokine levels.

The 27 patients selected for the present study had demonstrated at least a 60% reduction in numbers of sites with periodontal probing depth (PPD) ≥4 mm over the study period, and had been identified as either low or high CV-risk. Those with no more than one of the following risk factors were considered to be low CV-risk and those with six or more, or with a history of previous myocardial infarction (MI), to be high CV-risk:16,17 (1) age (>60 years); (2) hypertension (systolic blood pressure >140 mmHg or diastolic blood pressure >90 mmHg); (3) glycaemic control (HbA1c >6%); (4) smoking; (5) body mass index (BMI) of >25 kg/m2 and waist circumference of >88 cm for females or >102 cm for males; (6) total cholesterol >5 mmol/l; (7) triglycerides >2 mmol/l; (8) low density lipoprotein >3.5 mmol/l; (9) high density lipoprotein ≤1 mmol/l; (10) CRP >10 mg/l; (11) erythrocyte sedimentation rate >10 mm/h; and (12) creatinine >0.1 mmol/l. One of the high CV-risk patients had had a previous MI and the other 12 had between six and eight risk factors.

Serum was separated and stored in aliquots at −80 °C until analysis. Commercial high sensitivity ELISA kits (Bender Medsystems, Burlingame, CA, USA) were used to measure serum levels of sICAM-1, IL-6, IL-1β and tumour necrosis factor-α (TNF-α). Serum levels of antibodies to hHSP60, P. gingivalis GroEL and P. gingivalis whole cells were determined using an ELISA technique.18 Each of the antigens (recombinant hHSP60 (2 μg/ml), P. gingivalis GroEL (2 μg/ml), and P. gingivalis ATCC 3277 (1 μg/ml)) was coated onto 96-well high-binding plates (Maxisorb Immunoplates, Nunc, Roskilde, Denmark). Diluted (1:100) serum samples were added followed by peroxidase-conjugated rabbit anti-human IgG (DAKO, Glostrup, Denmark). Substrate containing 0.0075% H2O2 and 2.5 mM o-Tolidine (Eastman Kodak, Rochester, NY, USA) was added, the reaction stopped after 10 minutes with 3M HCl, and the optical density of the wells read at 450 and 655 nm. Antibody levels in the serum samples were determined from a standard curve of dilutions of a known concentration of normal human IgG (Zymed, CA, USA) coated onto each plate. Negative control wells using PBS instead of serum were used to determine background values.

Statistical methods

The Wilcoxon Signed-Rank test, the non-parametric equivalent of the paired t-test, was used to assess the changes in cytokine and antibody levels over the study period. Differences between low and high CV-risk groups with respect to changes in levels were examined using the Mann-Whitney (M-W) test, the non-parametric equivalent of the independent samples t-test. Fisher’s Exact test was used to compare low and high CV-risk groups with respect to proportions of patients with reduced antibody levels. All analyses were performed using SPSS v17.0 and p-values <0.05 were considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

As expected, given the classification criteria, patients in the high CV-risk group were older than those in the low CV-risk group (45 vs. 37 years) and more likely to be male (54% vs. 21%). They were also more likely to be overweight, have higher HbA1c levels and be taking anti-inflammatory medication. Both groups had similar numbers of sites with PPD ≥4 mm at baseline (25 ± 23.3 and 28 ± 22.6 for high and low CV-risk patients respectively), and at the end of the study period 85% of those with high CV-risk and 93% of those with low CV-risk had no more than four sites ≥4 mm (Table 1). The number of sites ≥4 mm remaining at the end of the study period was 2.6 ± 3.9 for high CV-risk patients and 0.7 ± 1.6 for low CV-risk patients.

Table 1.   Participant demographics, periodontal status, systemic inflammatory modifiers and serum cytokine and antibody levels for each of the cardiovascular risk groups at baseline
 Low CV-risk (n = 14)High CV-risk (n = 13)
  1. IQR: Interquartile range.

Age (years) (mean ± SD)37.0 ± 7.245.2 ± 7.1
# (%) males 3 (21.4%)7 (53.8%)
# sites ≥4 mm PPD (mean ± SD, range)28 ± 22.6 (8, 81)25 ± 23.3 (5, 83)
# sites ≥6 mm PPD (mean ± SD, range)1.6 ± 3.1 (0, 11)3.8 ± 8.3 (0, 30)
BMI >30 kg/m2 # (%) patients2 (14.3%)11 (84.6%)
HbA1c >6% # (%) patients0 (0%)3 (23.1%)
Anti-inflammatory medications2 (14.3%)5 (38.5%)
# (%) patients
 Serum sICAM-1 levels (ng/ml) (median, IQR) 540 (463, 604) 697 (623, 799)
 Serum IL-6 levels (pg/ml) (median, IQR)0.69 (0.55, 1.49)1.14 (0.63, 1.88)
 Serum IL-1β levels (pg/ml) (median, IQR)0.00 (0.00, 0.10)0.00 (0.00, 0.15)
 Serum TNF-α levels (pg/ml) (median, IQR)0.00 (0.00, 1.95)0.00 (0.00, 0.00)
 Serum anti-P. gingivalis antibodies (ng/ml) (median, IQR)2457 (1882, 4557)2539 (1867, 5457)
 Serum anti-P. gingivalis GroEL antibodies (ng/ml) (median, IQR)1071 (505, 2270)1310 (540, 2660)
 Serum anti-hHSP60 antibodies (ng/ml) (median, IQR) 969 (479, 2080)2285 (1551, 3168)

Median baseline levels of both sICAM-1 and IL-6 were slightly higher for the high CV-risk than for the low CV-risk patients (697 vs. 540 ng/mL for sICAM, 1.14 vs. 0.7 pg/ml for IL-6) (Table 1). The majority of patients (7/13 high CV-risk and 13/14 low CV-risk) experienced some level of reduction in sICAM over the study period, with the median changes over 12 months being reductions of 62.3 ng/ml (Wilcoxon p = 0.55) and 127.5 ng/ml (Wilcoxon p = 0.006) for the high CV-risk and low CV-risk patients respectively (Fig 1a). There was some evidence of a difference between the risk groups with respect to these changes (M-W p = 0.06). Less than 50% in each group recorded reduced levels of IL-6, the median changes being increases of 0.1 and 0.2 pg/ml in the high CV-risk and low CV-risk patients respectively (M-W p = 0.36). The greatest reductions were observed in the high CV-risk group (Fig 1b). Most patients (10/13 high CV-risk and 8/14 low CV-risk) had no detectable IL-1β at baseline; similarly for TNF-α (11/13 high CV-risk and 8/14 low CV-risk). Of those with detectable IL-1β at baseline, 1/3 high CV-risk and 5/6 low CV-risk patients recorded reduced levels after 12 months; 1/2 high CV-risk and 4/6 low CV-risk patients with detectable TNF-α at baseline also recorded reduced levels.

image

Figure 1.  Changes in (a) serum sICAM-1 and (b) serum IL-6 levels following improvement in periodontal health for low and high CV-risk groups.

Download figure to PowerPoint

Median baseline levels of anti-P. gingivalis, anti-P. gingivalis GroEL and anti-hHSP60 antibodies were higher in the high CV-risk group than in the low CV-risk group (Table 1). Although median anti-P. gingivalis antibody levels were reduced at follow-up (1783 ng/ml and 2240 ng/ml for high and low CV-risk patients respectively), the changes were not significant (Wilcoxon p = 0.22 and 0.43 for high and low CV-risk groups respectively). Median changes in anti-P. gingivalis GroEL antibodies were a reduction of 908.38 ng/ml (Wilcoxon p = 0.001) and an increase of 13.57 ng/ml (Wilcoxon p = 0.83) for high CV-risk and low CV-risk patients respectively. These changes differed significantly (M-W p < 0.001) between the groups (Fig 2a), as did the proportions experiencing reductions (100% and 42.9% in high CV-risk and low CV-risk respectively) (Fisher’s Exact p = 0.002) (Fig 3). The majority of patients (10/13 high CV-risk and 9/14 low CV-risk) experienced some level of reduction in anti-hHSP60 antibody levels. Median changes in anti-hHSP60 antibody levels (reductions of 613.36 ng/ml (Wilcoxon p = 0.009) and 138.41 ng/ml (Wilcoxon p = 0.3) in the high CV-risk and low CV-risk groups respectively) differed significantly (M-W p = 0.03). The greatest reduction in the high CV-risk group (4142 ng/ml) was almost five times that recorded in the low CV-risk group (898 ng/mL) (Fig 2b). By 12 months, anti-P. gingivalis GroEL and anti-hHSP60 antibody levels in the high CV-risk group were reduced to levels which were comparable to those in the low CV-risk group (Fig 3). The majority of the high CV-risk group (10/13 = 76.9%) recorded reductions to both anti-P. gingivalis GroEL and anti-hHSP60 antibodies compared with 14% (2/14) of the low CV-risk group.

image

Figure 2.  Changes in (a) serum anti-P. gingivalis GroEL (PgGroEL) and (b) serum anti-hHSP60 antibody levels following improvement in periodontal health for low and high CV-risk groups.

Download figure to PowerPoint

image

Figure 3.  Baseline and Year 1 serum antibody levels to P. gingivalis GroEL (PgGroEL) and hHSP60 for low and high CV-risk groups.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

The present study has shown that an improvement in periodontal health in individuals with low CV-risk resulted in a significant reduction in the inflammatory mediator sICAM-1 and in those with high CV-risk, and significant reductions in anti-P. gingivalis GroEL and anti-hHSP60 antibodies. In this study, only patients who had demonstrated a quantifiable improvement in periodontal health, i.e. a marked reduction in the numbers of deep pockets, were selected. A 60% reduction in the number of sites ≥4 mm was chosen, not as a therapeutic endpoint, but because it was deemed to reflect a substantial change in the oral inflammatory burden for the individual. While some oral inflammatory burden remained for a proportion of patients, this study aimed to examine changes in the inflammatory burden and of the mediators under study. A further categorization was performed according to CV-risk. It has been shown that the risk of CVD causing mortality is associated with the number of classical CV risk factors.19 Therefore, the patients in this study were clearly defined in terms of their periodontal and their CV parameters. This study analysed the systemic response to improved periodontal health in the context of CV-risk.

Low CV-risk patients in the present study displayed a lower frequency of systemic inflammatory conditions, lower baseline levels of all cytokines and overall, experienced a more notable improvement in these levels following improvement in periodontal health. For these patients, this suggests that periodontal disease was the major contributor to the systemic inflammatory burden, and thus a reduction in periodontal disease had a greater impact in reducing the total inflammatory load.

Higher levels of serum s-ICAM1 have been found in patients with diagnosed coronary heart disease (CHD) and coronary artery atherosclerosis.20 A meta-analysis showed an association between s-ICAM1 and CHD, as well as with classical CVD risk factors.21 sICAM-1 promotes cellular adhesion of leukocytes, and its deficiency in knockout mice has been shown to reduce the development of atherosclerotic plaque.22 sICAM-1 levels were significantly reduced in periodontal patients at four weeks post-treatment.12 The current study confirms this effect over a longer period (12 months), particularly in low CV-risk patients. Increased levels of the inflammatory cytokine IL-6 have been associated with both CV-risk23 and periodontal disease.24,25 In addition, periodontal therapy reduced these levels at three months post-treatment.25 A similar reduction in levels of IL-6 was not found in the current study after 12 months. For the remaining cytokines measured, the current study is in line with the results of previous studies13,24 which concluded that periodontal treatment had minimal impact on levels of TNF-α and IL-1β. The lack of a significant reduction in systemic cytokine levels, especially in the high CV-risk group following periodontal therapy may have been due to the fact that these high risk subjects had other sources of inflammation over and above their periodontal condition. Chronic systemic conditions such as diabetes, obesity and atherosclerotic lesions themselves have been shown to contribute to the total inflammatory burden for an individual,26 which in turn may contribute significantly to the levels of systemic inflammatory mediators.

HSP is expressed on all cell surfaces upon exposure to stimuli including infection, oxidative stress and elevated temperature.27,28 The cross-reactivity of antibodies to P. gingivalis GroEL and hHSP60 has been shown, with the two antigens sharing a high degree of similarity.29 HSP60 has also been shown to elicit a prominent T-cell response in early stage atherosclerosis.30 The current study has shown that high CV-risk patients had higher levels of antibodies to P. gingivalis GroEL and hHSP60 than low CV-risk patients and that these were significantly reduced following improvement in periodontal health. In contrast to the cytokine responses, where the effect was more marked in the low CV-risk patients, reductions in antibody levels were greater and more frequent in high CV-risk patients. This may indicate that high CV-risk patients were more likely to mount an antibody response which was directed towards HSPs and therefore potentially reactive with host antigens. After improvement in periodontal health in high CV-risk patients, antibody levels to both P. gingivalis GroEL and hHSP60 were reduced to levels which were similar to those for low CV-risk patients. Since there is evidence that anti-HSP antibodies have a role in the pathogenesis of CVD, the present study suggests that an improvement in periodontal health could lead to a reduction in CV-risk for high risk patients. The results also indicate that for patients who have other (non-periodontal) sources of inflammation, these antibodies may be better markers of the systemic effects of periodontal improvement than cytokines.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

In conclusion, this study found that subsequent to improved periodontal health, the anti-HSP (HSP60 and GroEL) antibody response was reduced, particularly for high CV-risk patients. sICAM-1 levels were also lowered, more so for low CV-risk patients. These results contribute to the growing understanding of the systemic impact of periodontal disease. In addition, this study provides further evidence for the possible potential benefits of the diagnosis and treatment of periodontal disease as a means of reducing CV-risk.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

The authors acknowledge Vinah Anderson, Anne Carle, staff at the Logan Hospital Dental Clinic and Michelle Sankey for collection of data and samples for the study, and Trevor Holcombe who contributed to the organization of the clinical aspects of the study. This study was supported by unrestricted grants from the National Health and Medical Research Council, HCF Health and Medical Research Foundation, the Australian Dental Research Foundation and Queensland Health.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References