Abstract
Background:
Inflammation is intimately involved in the pathogenesis of atherosclerosis and is accurately measured by high sensitivity C-reactive protein (hs-CRP), a sensitive marker for future risk of cardiovascular disease. The Correlation between Oral Health and Systemic Inflammation (COHESION) trial was designed to test the hypothesis that PlaqueHD, a plaque identifying toothpaste, reduces hs-CRP.
Methods:
COHESION was designed initially to enroll 132 subjects with hs-CRP between 2.0 and 10.0 milligrams per liter but randomized 112 between 0.5 and 10.0, of which 103 had baseline and follow up data and comprise the intention to treat sample. Of these, a pre-specified subgroup analysis included 40 with baseline hs-CRP > 2.0 and all hs-CRP < 10. Since the distribution of hs-CRP was skewed toward higher values, to achieve normality assumptions, the significance of changes in hs-CRP between groups over time was tested on log-transformed data using a mixed effects analysis of variance.
Results:
The intention-to-treat analysis showed no significant differences between the PlaqueHD and placebo group (p=0.615). The pre-specified subgroup analysis showed a significant difference between the PlaqueHD and placebo group (p=0.047). This finding resulted from a reduction in hs-CRP at follow up of 0.58 in the PlaqueHD and an increase of 0.55 in the placebo group.
Conclusions:
These findings are compatible with those of a prior pilot trial which also suggested benefits only in subjects with baseline elevations. These findings suggest that future trials targeting reductions of hs-CRP levels should randomize subjects with baseline hs-CRP between 2.0 and 10.0.
Keywords: plaque identifying toothpaste, Plaque HD, inflammation, hs-CRP
INTRODUCTION
Inflammation is intimately involved in the pathogenesis of atherosclerosis and is accurately measured by high sensitivity C-reactive protein (hs-CRP), a sensitive marker for future risks of cardiovascular disease. (1) The American Heart Association and Centers for Disease Control and Prevention have stratified risk for future CVD as follows based on hs-CRP levels: low: <1.0 milligrams per liter; average: 1.0 to 3.0; and high: > 3.0.(2)
With respect to the interrelationships of periodontal disease, systemic inflammation, and cardiovascular disease (3), speculation began in antiquity. In recent years, systemic inflammation has been proposed as a plausible mechanism to explain any valid statistical association between periodontal disease and cardiovascular disease. The effect size, however, is likely to be small and case-control and cohort studies have inherent confounding by age, gender, socioeconomic status, obesity and perhaps other unknown confounders that may be as large as the effect size. Thus, the most reliable design strategy to test the hypothesis is a randomized trial of sufficient size and duration. (4,5)
The hypothesis that PlaqueHD, a plaque identifying toothpaste, produces significant reductions in hs-CRP had been explored in a previous pilot randomized trial of 61 subjects without any elevated hs-CRP at baseline. (6) In a pre-specified subgroup analysis among 38 subjects treated and followed for 60 days with baseline and follow up hs-CRP > 0.5 and ≤ 10.0. Specifically, subjects randomized to PlaqueHD had a reduction of 0.91, while those assigned to placebo group an increase of 0.57. Thus, PlaqueHD was associated with a significant overall change in 1.46 in level of hs-CRP (p=0.041).
In the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) (7) as well as The Canakinumab Antiinflammatory Thrombosis Outcome Study (CANTOS)(8) the lower limit of hs-CRP was 2.0. Further, values of hs-CRP>10.0 are likely to be due to other systemic causes of inflammation which would bias the results of any small pilot trial.
The Correlation between Oral Health and Systemic Inflammation (COHESION) trial was designed to provide further randomized data concerning this hypothesis.
METHODS
The protocol was approved by the Institutional Review Board of the Marshfield Clinical Research Institute as well as University of Wisconsin School of Medicine and Public Health and the trial was posted on ClinicalTrials.gov. A total of 112 men and women, aged 21 to 84 years, signed informed consents and were randomized. All were given instructions to follow the same brushing protocol and received a 30-day supply of toothpaste containing either the plaque identifying toothpaste or a non-plaque identifying toothpaste identical in fluoride content. Changes in hs-CRP in milligrams per liter were assessed by Quest Laboratories using an enzyme linked immunosorbent assay.
COHESION planned to enroll 132 subjects in the hope to have complete baseline and follow up data on 120 which would achieve adequate statistical power to detect reliably the most plausible small to moderate reductions in hs-CRP. In the initial protocol. hs-CRP was to be between 2.0 and 10.0 mg/L. The lower limit was later reduced to 0.5 mg/L to enhance recruitment. Thus, the intention to treat analysis included all randomized subjects and a prespecified subgroup analysis was performed which included all randomized subjects with baseline hs-CRP > 2.0 and follow hs-CRP < 10.0(2). The significance of any differences in baseline characteristics was tested by t-tests for continuous or chi-square tests for discrete variables. Means and standard deviations were calculated, and the distribution of hs-CRP was skewed toward higher values. Thus, to achieve normality assumptions, the significance of changes in hs-CRP between groups at baseline and follow up was tested on log-transformed data by using mixed effects ANOVA with treatment, time, and treatment:time interaction as fixed effects and subject as a random effect. For ease of interpretation, the actual values are reported but the significance test is based on the logarithmic transformed data. Wilcoxon rank sum tests were also conducted for the comparison of change over time in hs-CRP between groups.
All significance tests were conducted using R version 3.5 at a two-sided significance level of 0.05.
Role of the Funding Source:
This pilot trial was funded as an investigator-initiated research grant by the Marshfield Clinic Research Institute and the University of Wisconsin School of Medicine and Public Health to Amit Acharya, BDS, MS, PhD as Principal Investigator. The toothpaste and matching placebo were provided by TJA Health, LLC, Joliet, Illinois. Neither the funding source nor the distributor of the toothpaste had any role in the design, conduct, analysis, interpretation, preparation of the manuscript, or in the decision where to submit.
RESULTS
Of 112 randomized subjects, 103 had baseline and follow up data so they constituted the intention to treat sample. Of the 103, 54 were randomized to PlaqueHD and 49 to placebo. The average ages were 51.2 years in the PlaqueHD group and 49.3 in the placebo group (p=0.532). In addition, 71 were women (68.9%) and 32 were men. Of the 71 women, 42 (59.2%) were assigned PlaqueHD compared with 12 (37.5%) of the 32 men (p=0.068).
The intention-to-treat analysis (Table 1) showed no statistically significant differences between the two groups (p=0.615). Figure 1 displays the distributions of values at baseline and follow up as well as absolute and percent changes in hs-CRP.
Table 1.
Results of the Intention-to-Treat Analysis of Subjects with Baseline High-Sensitivity C-Reactive Protein >0.5 and <10.0
| Group | n | Baseline Mean (SD) | Follow-up Mean (SD) | Absolute Difference from Baseline to Follow-up | Absolute Difference Between Treatment Groups on Log Scale (95% CI) * | P Value |
|---|---|---|---|---|---|---|
| PlaqueHD | 54 | 3.23 (2.67) | 3.75 (4.59) | 0.52 (3.79) | 1.07 (−1.12, 0.13) | .615 |
| Placebo | 49 | 2.09 (1.79) | 2.46 (2.47) | +0.37 (1.20) |
Absolute difference on arithmetic scale and 95% confidence limits: 0.15 (−0.97, 1.27). CI = confidence interval; SD = standard deviation.
Figure 1.
Changes in high-sensitivity C-reactive protein observed in the intention-to-treat analysis of subjects with baseline >0.5 and <10.0.
The prespecified subgroup analysis consisted of 40 subjects, 15 in PlaqueHD and 25 in placebo. Of the 63 excluded, 60 had were due to baseline hs-CRP<2.0 and 3 had follow up values > 10.0. In this analysis (Table 2 and Figure 2) there was a significant difference between the PlaqueHD and placebo group (p=0.047). This finding resulted from a reduction in hs-CRP at follow up of 0.58 in the PlaqueHD and an increase of 0.55 in the placebo group.
Table 2.
Results of the Prespecified Subgroup Analysis of Subjects with Baseline High-Sensitivity C-Reactive Protein >2.0 and <10.0
| Group | n | Baseline Mean (SD) | Follow-up Mean (SD) | Absolute Difference from Baseline to Follow-up | Absolute Difference on Log Scale (95% CI) * | P Value |
|---|---|---|---|---|---|---|
| PlaqueHD | 25 | 5.00 (2.10) | 4.42 (2.69) | −0.58 (2.26) | −1.51 (1.01–2.27) | .047 |
| Placebo | 15 | 4.33 (1.68) | 4.88 (1.84) | +0.55 (1.36) |
Absolute difference on arithmetic scale and 95% confidence limits: −1.14 (2.45–0.17). CI = confidence interval; SD = standard deviation.
Figure 2.
Change in high-sensitivity C-reactive protein observed in the prespecified subgroup analysis of subjects with baseline >2.0 and <10.0.
CONCLUSIONS
These pilot trial data after 30 days of treatment and follow up are compatible with findings from a previous pilot trial after 60 days of treatment and follow up. (6) Specifically, in subjects without baseline elevations, not surprisingly, there were no significant reductions In pre-specified subgroup analyses of subjects with some elevations at baseline, PlaqueHD was associated in with statistically significant reductions in hs-CRP.
It is intriguing to speculate about the baseline imbalance by gender. In general, women tend to have higher hs-CRPs than men. (7) This may have accounted, at least in part, for the far higher baseline hs-CRPs of 3.23 in the PlaqueHD arm compared with 2.09 in the placebo arm. Women also tend to be less responsive to treatments that affect hs-CRP than men, due principally to confounding by body mass index for which we had no data. Unfortunately, we had no baseline data on variables such as body mass index or use of postmenopausal estrogens to explore effect modification.
With respect to inflammation and decreases risks of myocardial infarction and stroke, in JUPITER, (7) among 17,802 patients, those assigned at random to rosuvastatin experienced reductions of 37% reduction in hs-CRP and 44% (p<0.00001) in clinical cardiovascular events. In CANTOS, (8) among 10,061 patients, those assigned to 150mg canakinumab experienced reductions of 37% in hs-CRP and 20% (p=0.021) in clinical cardiovascular events. Both trials support the hypothesis that inflammation plays a major role in the development of clinical cardiovascular events and, further indicate that decreasing hs-CRP reduces risks of clinical cardiovascular events. Whether this plaque-identifying toothpaste decreases clinical cardiovascular events requires a large-scale randomized trial of sufficient size and duration. (4,5) Such a finding would have major clinical and public health implications.
Both the null intention to treat analyses and positive pre-specified subgroup analyses of this pilot trial are similar to findings from a prior pilot. Both suggest that pilot trials of lowering hs-CRP should randomize only subjects with values >2.0 and <10.0.
ACKNOWLEDGEMENTS
We are indebted to Jennifer Bufford, MS, CCRC, CCRP and Karen Hansen, MD for advice and help.
Funding:
This investigator-initiated research grant was funded by the Marshfield Clinic Research Institute and the University of Wisconsin Institute for Clinical and Translational Research, which is supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1TR002373. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The toothpaste and matching placebo were provided by TJA Health, LLC (Joliet, Ill.). Neither the funding source nor the distributor of the toothpaste had any role in the design, conduct, analysis, interpretation, or preparation of the manuscript. Nor did they have a role in the decision where to submit.
Conflict of Interest:
AA reports that he receives funds from Delta Dental of Wisconsin Foundation for developing and designing public health tools in dentistry, and royalties for authorship and editorship of 2 textbooks. AA does not own any common or preferred stock in any pharmaceutical or medical device company. IG, SH, and MCT report no disclosures. KK reports that he serves as an independent scientist as a member of data monitoring committees for Astellas, AstraZeneca, Actelion, Bayer, Bristol-Myers Squibb, GlaxoSmithKline, Idorsia, Merck, and Sarepta. KK holds no stock in any biopharmaceutical or medical device company. DLD reports that that he serves as an independent scientist in an advisory role to the National Institutes of Health, the US Food and Drug Administration, and the pharmaceutical and medical device industry on the design, monitoring, and analysis of trials; on data monitoring committees for Duke Clinical Research Institute, Population Health Research Institute at University of Hamilton, Actelion, Bristol-Myers Squibb, Medtronic, Boston Scientific, GlaxoSmithKline, Mesoblast, LivaNova, DalCor, and Merck; and to Frontier Science, Intercept, Sanofi, and D L DeMets Consulting. DLD holds no stock in any pharmaceutical or device company. CHH reports that he serves as an independent scientist in an advisory role to investigators and sponsors as chair of data monitoring committees for Amgen, British Heart Foundation, Cadila, Canadian Institutes of Health Research, DalCor, and Regeneron, and to the US Food and Drug Administration, and UpToDate; receives royalties for authorship or editorship of 3 textbooks and as co-inventor on patents for inflammatory markers and cardiovascular disease that are held by Brigham and Women’s Hospital; has an investment management relationship with the West-Bacon Group within SunTrust Investment Services, which has discretionary investment authority; and does not own any common or preferred stock in any pharmaceutical or medical device company.
REFERENCES
- 1.Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997; 336(14):973–979. [DOI] [PubMed] [Google Scholar]
- 2.Pearson T, Mensah GA, Alexander RW, et al. Markers of Inflammation and Cardiovascular Disease Application to Clinical and Public Health Practice A Statement for Healthcare Professionals from the Centers for Disease Control and Prevention and the American Heart Association Circ 2003; 107:499–511. [DOI] [PubMed] [Google Scholar]
- 3.Taylor S, Hennekens CH: Interrelationships of periodontal disease, systemic inflammation, and cardiovascular disease. Ann Dent & Oral Dis, 2018;2:108–110 [Google Scholar]
- 4.Hennekens CH, DeMets D: Statistical association and causation: Contributions of different types of evidence. JAMA, 2011, 306:1134–1136 [DOI] [PubMed] [Google Scholar]
- 5.Hennekens CH, DeMets D The need for large-scale randomized evidence without undue emphasis on small trials, meta-analyses, or subgroup analyses. JAMA. 2009; 302(21):2361–2. [DOI] [PubMed] [Google Scholar]
- 6.Fasula K, Evans C, Boyd L, et al. A randomized trial of plaque identifying toothpaste on reduction of dental plaque and inflammation. AJM, 2017, 130:612–16 [DOI] [PubMed] [Google Scholar]
- 7.Ridker PM, Danielson E, Fonseca F et al. Rosuvastatin to Prevent Vascular Events in Men and Women with Elevated C-Reactive Protein. N Engl J Med. 2008; 359:2195–2207. [DOI] [PubMed] [Google Scholar]
- 8.Ridker PM, Everette BM, Thuren T, et al. for the CANTOS trial group, Antiinflammatory therapy with canakinumab for atherosclerotic disease. NEJM, 377:1119–1131. [DOI] [PubMed] [Google Scholar]