P1‐348: Periodontal bacterial antibodies may help discriminate between Alzheimer's disease and normal patients: A pilot study
Year: 2008
Type: article
Study design: Observational Study
Abstract: Biological markers for AD are essential for early treatment intervention. A landmark study (Ray, 2007) developed a plasma molecular signature predicting and classifying AD with high accuracy. Among these proteins, TNF–α and immune molecules were significantly represented. Periodontal disease is a peripheral, chronic infection whose expression depends on the pathogenic bacteria (Porphyromonas gingivalis (Pg), Tenerella forsythensis (Tf), and Aggregatibacter actinomycetemcomitans (Aa)) and the inflammatory-immune response. Since antibody response represents an index of bacterial systemic exposure, host inflammatory-immune capabilities and TNF–α is central to AD molecular signature we hypothesized that AD patients have elevated antibody titers to periodontal bacteria and elevated TNF–α compared to control subjects (NL). 18 patients with probable AD and 16 NL diagnosed using NINCDS-ADRDA criteria were used. Plasma cytokines were assessed by Multiplex and plasma IgG antibodies against Pg, Tf, and Aa by ELISA. Plasma positivity was defined by values in the upper quartile. Mann-Whitney U tests (MWU), χ2 were used for differences between variables. Binary logistic regression determined the independent variables representing the best classifiers of NL and AD. A total of 66.7% of AD subjects were plasma positive for at least one antibody to periodontal bacteria compared to 37.5% NL (χ2=2.9, df=1, p=0.087, OR=3.33, 95% C.I= (0.815–13.637). AD subjects tended to have higher mean number of positive antibody tests compared to NL (mean 1.06±0.94; 0.56±0.89; MWU, z=-1.71, two sided p=0.09). In the binary logistic analysis TNF–α and positive IgG antibody against Pg-Tf-Aa were examined. The OR for TNF–α and positive plasma IgG antibody tests were 1.4 and 3.4 (p=0.007 and 0.037 respectively) and classified AD with 81% specificity and 78% sensitivity. The inclusion of Il-1β into the model resulted in a significant increment in the χ2 to the model (from χ2 =14 to 23, p<.000) and significant combined effect (p=004). Based on the ROC the combined model classified AD and NL with a specificity=93.8% and sensitivity=88.9%. The results of this pilot study suggest that antibodies against periodontal bacteria associated with AD, may constitute an index of immune changes in AD and in combination with TNF–α may be important in discriminating AD patients from normals.
Source: Alzheimer s & Dementia
Institution: New York University
Language: English
Subfield: Periodontics
Field: Dentistry
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W1991327916