Multilevel modeling of gingival bleeding on probing in young adult carriers of non-JP2-like strains of Aggregatibacter actinomycetemcomitans
- Authors
- Year
- 2009
- Journal
- Clinical Oral Investigations, 13(2), 171-178
- DOI
- 10.1007/s00784-008-0218-4
- Abstract
The influence of Aggregatibacter actinomycetemcomitans on inflammation in subjects with gingivitis has not been studied in great detail. Seventeen healthy young adults with plaque-induced gingivitis or localized mild chronic periodontitis harboring cultivable numbers of A. actinomycetemcomitans were thoroughly examined. Samples of subgingival plaque were obtained from mesial surfaces of all teeth present. In addition, 12 oral mucosal surfaces and unstimulated saliva were sampled. Species identity, presence of the leukotoxin gene, and absence of a specific 530 b deletion in the leukotoxin promoter region indicating non-JP2-like strains were assessed by polymerase chain reaction. Based on a multilevel random intercept model adjusted for probing depth, age, and smoking status, the odds of bleeding on probing was increased by a factor of 1.89 (1.09-3.29, p = 0.024) if, in addition to plaque, A. actinomycetemcomitans could be recovered from the site. At a site without visible supragingival plaque but with cultivable numbers of subgingival A. actinomycetemcomitans the odds ratio of bleeding on probing was 3.37 (0.86-13.2, p = 0.081). Simulating variance partition coefficients revealed that between 1-2% (a clean, shallow site without A. actinomycetemcomitans; a deep site covered by plaque containing A. actinomycetemcomitans) and 6-7% (a moderately deep site with neither visible plaque nor cultivable A. actinomycetemcomitans) of the residual variance was attributable to differences between subjects. The present cross-sectional study indicates that non-JP2-like strains of A. actinomycetemcomitans may enhance gingival bleeding tendency even in the absence of clinically visible supragingival plaque.
- Number of levels
- 2
- Model data structure
- Response types
- Multivariate response model?
- No
- Longitudinal data?
- Yes
- Substantive discipline
- Paper submitted by
- Hans-Peter Müller, Institute of Clinical Dentistry, Tromsø University, hans-peter.muller@uit.no