Date of Award


Document Type

Open Access Dissertation


Epidemiology and Biostatistics

First Advisor

Anwar T. Merchant


Periodontal disease is an inflammatory disease caused by polybacterial infection and inflammation plays key role that associates periodontal disease with different systemic diseases including diabetes and cardiovascular diseases. The main objective of this study is to evaluate and explore the association of serum IgG antibodies in four distinct clusters that were formed empirically from species specific 19 periodontal antibody titers with T2DM and pre-diabetes adjusting for known confounders such as age, sex, race-ethnicity, income to poverty ratio, education, smoking and drinking alcohol, BMI, WC, physical activity, missing teeth, dentist visits and other nutritional factors.

From the first aim, we formed 4 clusters empirically using cluster analysis from 19 periodontal antibody titers and named it as Orange-Red, Red-Green, Yellow-Orange, and Orange-Blue based on Socransky’s grouping scheme. Using the priory study model, theoretically we made 3 groups i.e. Etiologic, Putative and Health related groups from 11 periodontal bacteria antibody. On the full adjustments for these known confounders, our finding showed that Orange-Red cluster that included P gingivalis and Prevotella sps were positively associated in moderate level with diabetes but not with pre-diabetes. However, the Red-Green cluster which contained T denticola, T forsythia, A actinomycetemcomitans and others (as shown in Flow chart 1) were inverse but significantly associated to diabetes. The Orange-Blue cluster scores that included A. naeslundii and E. Nodatum showed an inverse relation to diabetes and pre-diabetes in the crude analysis but that was attenuated after adjustment [diabetes: OR 0.935 (0.869- 1.007), pre-diabetes OR: 0.987 (0.944-1.032) . The Yellow-Orange cluster that included majority of Steptococcus sps were not found to be associated with diabetes or pre- diabetes. With the same approach, we found the significant positive association of etiologic group with pre-diabetes among the association with the 3 groups. In our second aim, we explored the possible interaction between serum IgG antibodies and clinical periodontal destruction (as assessed by clinical attachment loss and pocket depth) in association to diabetes. We found that the tertile of pocket depth modified the association of Orange-Blue cluster with diabetes upon the full adjustment for the known confounders.

In our third aim, we measured the extent of the association between empirically defined 4 clusters and theoretically defined 3 groups with each of the five key components of cardiovascular risk factors i.e. hypertension, hypertriglyceridemia, low HDL-cholesterol, central obesity, and elevated plasma glucose and also the MetS. We found no significant association with any of these components and MetS but the elevated plasma glucose. Elevated Orange-Red clusters was positively associated whereas Orange-Blue was inversely but significantly associated with elevated plasma glucose. The association became more prominent when the elevated plasma glucose greater than or equal to 110 mg/dl.

These findings show that specific periodontal antibody serological markers may be used as predictor of systemic health mainly the diabetes status.


© 2014, Deepika Shrestha

Included in

Public Health Commons