Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Nov;43(11):909-917.
doi: 10.1111/jcpe.12597. Epub 2016 Aug 12.

Periodontal Diseases and Carotid Intima-Media Thickness in Bangladesh

Free PMC article

Periodontal Diseases and Carotid Intima-Media Thickness in Bangladesh

Fen Wu et al. J Clin Periodontol. .
Free PMC article


Aim: To evaluate the relationship between periodontal diseases and subclinical atherosclerosis in a younger and lean South Asian population.

Methods: We conducted a cross-sectional study in 917 subjects (mean age 46 years and mean body mass index 21.1 kg/m2 ) from the Health Effects of Arsenic Longitudinal Study in Bangladesh. Multivariate linear regression models were used to assess the associations between multiple clinical measures of periodontal diseases and carotid intima-media thickness (IMT).

Results: Mean attachment loss (AL) and percentage of sites with AL ≥ 4 mm (% AL ≥ 4) were associated with increased IMT. The IMT was 20.0-μm (95% CI: 2.2, 37.8) and 26.5-μm (95% CI: 8.9, 44.1) higher in subjects in the top quartile of mean AL (>3.72 mm) and % AL ≥ 4 (>58.4%), respectively, compared to those in the bottom quartile. In a subset of 366 subjects, mean AL was positively associated with plasma levels of matrix metalloproteinase-9 (p < 0.05) and soluble intercellular adhesion molecule-1 (p < 0.01).

Conclusions: Attachment loss was associated with subclinical atherosclerosis in this young and lean Bangladeshi population. Future prospective studies are needed to confirm this association.

Keywords: Bangladesh; atherosclerosis; cardiovascular disease; carotid intima-media thickness; periodontal diseases.


Figure 1
Figure 1
Carotid intima-media thickness (μm) in relation to mean attachment loss (mm) overall and stratified. Cut-off points for mean attachment loss were determined by quartiles (Q1: ≤ 2.30; Q2: 2.31−2.85; Q3: 2.86−3.72; Q4: > 3.72) in the overall population. β represents the adjusted difference in average carotid IMT values between the comparison and reference groups, adjusting for sex, age, BMI, smoking status (never and ever), and betel quid chewing at the time of IMT measurement, as well as educational attainment at baseline (except for the variable stratified by in the stratified analyses). P for trend was based on continuous mean attachment loss. P for interaction was based on the cross-product term between each effect modifier and continuous mean attachment loss.

Similar articles

See all similar articles

Cited by 3 articles