Prevalence analysis of putative periodontal pathogens in patients with aggressive periodontitis and healthy elderly. A molecular studyLilian Edesi-Neuss111.121.234561.2.11.2.271.2.381.2.491.2.51.2.6101.3111.4121.4.11.4.21.4.3131.4.41.51.61.6.11415161.6.21.71721733.117183.21920213.33.4222324253.5263.6273.73.7.1283.7.2293.83043055.130315.2325.3604#29933615#30834626#286614#27635614#30736605#6905.4376376.13839406.26.2.141426.2.2436.2.36.2.4446.2.56.2.6456.2.76.2.86.2.9466.2.106.2.1147486.36.4495075085099.1519.29.2.19.2.1.15253549.2.1.29.2.1.39.2.1.455569.2.1.59.2.1.657589.2.29.2.2.1599.2.2.29.2.2.39.2.2.4609.2.2.59.2.2.69.2.2.7619.2.2.89.2.2.99.2.2.10629.2.2.119.2.39.2.3.19.2.3.263References Danksagung Lebenslauf Eidestattliche ErklärungenTable of contentsHelp Statistical analysis

Statistical analyses was supported by Ms. Siebert, Institute of Medical Biometrics, Humboldt University. For all calculations SPSS software v. 10.0 was used.
All 45 GAP patients and 21 elderly subjects were included in the statistical analysis. The patient served as a statistical unit. A difference of p<0.05 was considered statistically significant.
The differences in prevalence (number of positive subjects) between the two groups was computed for each species with chi-square test. A patient was considered positive when at least one sampled site was positive.
When comparing the colonization of periodontal pockets and control sites of the GAP patients, only one arbitrary chosen pocket and one shallow site per patient was analyzed and evaluated using the chi-square test.
Assessment of the prevalence of the species in periodontal pockets of the GAP patients versus sampled sites of the elderly was performed by including four sites per subject and using the chi-square test.
As only one control site was available from each GAP patient, the comparison of the presence of bacteria in similar shallow sites in elders was carried out using a single arbitrary site per elderly subject and the chi-square test.
The analysis of the presence of bacteria at different pockets depths was performed with the data of only 23 GAP patients using the chi-square test.
The non-parametric Mann-Whitney test for two independent samples was used to test the null hypothesis that the load of bacteria per patient, i.e. the number of positive sites per patient, is the same in both subject groups. Four pockets per patient from the GAP group and four sites per subject from the elderly population were analyzed.
The relationship between two bacterial species, i.e. the frequency of co-existence, was analyzed by odds ratio (OR) calculations as suggested by Socransky (177). The odds of an event are defined as the ratio of the probability of the event to the probability of its complement. For each pair of the investigated species within the GAP patient group, 2x2 contingency tables were constructed. The magnitude of the association between two species is indicated by the amount OR differs from 1.0, the indication of no association. OR<0.5 shows a negative association and OR>2 a positive one.