C-Reactive Protein Levels are Elevated in Patients with Periodontitis and Their CRP Levels May Go Down After Periodontal Therapy
Ananda P. Dasanayake BDS, MPH, PhD, FACEa,
aProfessor & Director, Graduate Program in Clinical Research, New York University College of Dentistry, 250 Park Ave South - 6th Floor, New York, NY 10003-1402
J Clin Periodontol 2008;35(4):277-90.
Available online 6 March 2009.
Level of Evidence
Level 1: Good-quality, patient-oriented evidence.
Strength of Recommendation Grade
Grade A: Consistent, good-quality patient-oriented evidence.
Purpose/Question
Are CRP levels significantly increased in periodontitis patients and what is the effect of periodontal therapy on their CRP levels?.
Source of Funding
Industry – Unrestricted grant from Philips Oral Healthcare EMEA.
Type of Study/Design
Systematic review and meta-analysis of observational and experimental studies.
Article Outline
Summary
Subjects
Key Exposure/Study Factor
Main Outcome Measure
Main Results
Conclusions
Commentary and Analysis
References
Summary
Subjects
This is not a single study but a systematic review of 18 studies carefully selected from the literature using objective selection criteria and a comprehensive database (Medline, PubMed, and CENTRAL) search. This also includes a meta-analysis based on 13 of those studies.
In 10 of the cross-sectional studies, there were 702 subjects with periodontitis, and 902 controls. In 4 interventional studies, there were 152 patients who received the intervention and 134 who did not. Data on gender distribution, dates of recruitment, and length of patient follow-up are not sufficiently summarized as it is a difficult task.
Key Exposure/Study Factor
There were 2 primary exposures related to their hypotheses:(1) presence of periodontitis, and (2) treatment of periodontitis. Diagnostic criteria used for periodontitis in each study and the details on the periodontitis treatment are clearly presented in detailed tables.
Main Outcome Measure
The primary outcome measure was high sensitivity C-reactive protein (CRP) levels.
Main Results
In the 18 studies evaluated, subjects with periodontitis had CRP levels that were greater than 2.1 mg/L. Weighted mean difference of CRP levels between subjects with periodontitis and controls was 1.56 mg/L in the meta-analysis, which was statistically significant (P < .05). Treatment studies demonstrated lower levels of CRP after treatment and meta-analysis of treatment studies indicated a weighted mean CRP level difference of 0.5 mg/L after treatment that was also statistically significant (P = .02).
Conclusions
This comprehensive systematic review clearly shows that CRP levels are significantly higher among periodontitis patients compared with controls. Periodontal treatment also seemed to significantly lower the CRP levels; however, the evidence for the latter is based on a smaller number of studies.
Commentary and Analysis
Periodontal disease is known to induce systemic inflammation which is partly mediated through acute phase reactants such as CRP.1 CRP is considered a biomarker of systemic inflammation and a marker of subsequent atherosclerosis and cardiovascular disease (CVD).[2], [3] and [4] However, the proposed link between PD, CRP levels, and CVD is less than conclusive. Within that framework, these authors set out to test 2 specific hypotheses: (1) CRP levels are significantly elevated in periodontitis patients compared with comparable controls, and (2) periodontal treatment significantly reduces the CRP levels. In order to test these hypotheses, one should exclude all other potential factors that are re lated to CRP levels, as CRP is a nonspecific marker of the acute-phase response. They have made a reasonable attempt to control for the heterogeneity of the studies that may at least have partially overcome the above challenge.
Authors have creatively and economically tested their hypotheses by doing a systematic review and a meta-analysis. The study selection criteria, search strategies, data abstraction, analysis, and presentation of results are near perfect. They conclude that the CRP levels in periodontitis patients are significantly elevated compared with controls and periodontitis treatment significantly reduces the CRP levels, even though this reduction is less than 1 mg/L.
Their first conclusion that the subjects with periodontitis have significantly elevated CRP levels may be closer to the truth given the considerable number of studies and sample sizes included in the analyses that led to the above conclusion. However, it is possible that the association is still confounded by unknown and/or uncontrolled confounders. It is difficult to assess the direction of such partial confounding on the observed results (ie, is the true difference greater or lesser than the observed).
Their second conclusion is less robust due to the smaller number of studies and subje cts involved and the variations in the baseline disease characteristics and the treatment protocols. Further experimental studies with adequate statistical power are needed before we draw definite conclusions in relation to this second hypothesis.
CRP levels, although nonspecific, are useful risk indicators for subsequent disease outcomes, and as such, the primary outcome of this study is relevant and significant. This article, however, will be very useful to the reader and the average practitioner. An honest discussion with the patients on the potential link between periodontitis and inflammatory markers that may increase their subsequent risk of events such as atherosclerosis and CVD is warranted and the limitations of the available evidence should be highlighted during that discussion.
References
1 B.G. Loos, Systemic markers of inflammation in periodontitis, J Periodontol 76 (11 Suppl) (2005), pp. 2106–2115. View Record in Scopus | Cited By in Scopus (50)
2 C.M. Albert, J. Ma, N. Rifai, M.J. Stampfer and P.M. Ridker, Prospective study of C-reactive protein, homocysteine, and plasma lipid levels as predictors of sudden cardiac death, Circulation 105 (22) (2002), pp. 2595–2599. View Record in Scopus | Cited By in Scopus (201)
3 J. Danesh, R. Collins, P. Appleby and R. Peto, Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies, JAMA 279 (18) (1998), pp. 1477–1482. View Record in Scopus | Cited By in Scopus (934)
4 P.M. Ridker, N. Rifai, L. Rose, J.E. Buring and N.R. Cook, Comp arison of C-reactive protein and low-density lipoprotein cholesterol levels in the prediction of first cardiovascular events, N Engl J Med 347 (20) (2002), pp. 1557–1565. View Record in Scopus | Cited By in Scopus (1405)