5.1.1 Patient and aneurysm characteristics
Characteristics of 29 patients, 7 autopsy cases, their aneurysms and proportions of PCR-positive samples are shown in Table 2.
Thirteen patients out of 29 were previously healthy, without any regular medication at the time of haemorrhage. Nine patients had hypertension, three had hypercholesterolaemia, three had hypothyreosis, one patient had diabetes, and others had conditions such as allergy, rheumatoid arthritis or a history of previous head trauma. Six patients had a suspicion of excessive alcohol consumption and 16 patients were smokers. Four patients had a positive family history for SAH.
Table 2. Characteristics of the aneurysm patients and autopsy specimens
Surgical patients Autopsy specimens
Number of patients 29 7
Abbreviations: ACA=anterior cerebral artery; ACoA= anterior communicating artery; ICA= internal carotid artery; MCA= middle cerebral artery; VBA= vertebro-basilar aneurysms
sanguinis ATCC 10556, Streptococcus gorgonii ATCC 10558, Aggregatibacter actinomycetemcomitans ATCC 700685, Porphyromonas gingivalis ATCC 33277 and Escherichia coli ATCC 25922 (LGC Standards, Teddington, Middlesex, UK).
4.9 DATA ANALYSES
Differences in the bacterial findings and patient characteristics between ruptured and unruptured groups were assessed with Fisher’s exact test (Studies I, II), and differences in the amount of bacterial DNA between the groups (patients with or without ≥6mm gingival pockets) were assessed with the Mann-Whitney test (Study III).
Fisher’s exact test was used to compare the prevalences of periapical lesions and gingival pockets ≥6mm between smokers and non smokers (Study III).
The distribution of absolute amounts of measured bacterial DNA was skewed, and therefore logarithmic values were used. Mann–Whitney test was used to assess the differences in the amounts of bacterial DNA in the samples between the patients with and without ≥6 mm gingival pockets (Study III). To compare the prevalence of gingival pockets between the study group and the population-based sample (Study III), the open data from a population-based study, the “Health 2000”-study (Suominen-Taipale et al., 2008) was used. The results were calculated using direct age and sex standardization (two-sample z-test) with the Health 2000 survey (Suominen-Taipale et al., 2008) as a reference (control) population.
In study IV, the number of patients was 60. After amplification in duplicates, the quality of 63 (out of 120 possible) samples was sufficient for sequencing. After sequencing, 34 samples from 23 patients were of sufficient quality for the analyses.
Eleven patient samples contained duplicates and 12 samples were single samples.
The replicates were averaged before statistical analyses. To investigate the gingival pocket bacterial communities of the study patients and to assess the alpha diversity (chao1-index) (Chao, 1984), we used the softwares QIIME (Caporaso et al., 2010), trimmomatic (Bolger, Lohse and Usadel, 2014), phyloseq (McMurdie and Holmes, 2013) and DESeq2 (Love, Huber and Anders, 2014). Briefly, quality control was performed with FastQC followed by trimming of the primers and low quality bases (phred score <20) and removal of short (<200bp) sequences. Data preparation was carried out with fasta-join in QIIME. Contaminants, chimera, chloroplasts, archaea and eucaryotic sequences were removed with QIIME and mothur. OTU-clustering was performed with uclust in QIIME. Taxonomy assignment was carried out using Human Oral Microbiome Database (HOMD) (Chen et al., 2010) as a reference database. Alpha diversity and Differential Analysis of Abundance (DAA) were assessed using phyloseq. The regression analyses were made with and without covariants (sex and age). The control samples (n=10) were analyzed by the same protocol. The quality of 4 samples was sufficient for sequencing.
5 RESULTS
5.1 STUDY I
5.1.1 Patient and aneurysm characteristics
Characteristics of 29 patients, 7 autopsy cases, their aneurysms and proportions of PCR-positive samples are shown in Table 2.
Thirteen patients out of 29 were previously healthy, without any regular medication at the time of haemorrhage. Nine patients had hypertension, three had hypercholesterolaemia, three had hypothyreosis, one patient had diabetes, and others had conditions such as allergy, rheumatoid arthritis or a history of previous head trauma. Six patients had a suspicion of excessive alcohol consumption and 16 patients were smokers. Four patients had a positive family history for SAH.
Table 2. Characteristics of the aneurysm patients and autopsy specimens
Surgical patients Autopsy specimens
Number of patients 29 7
Abbreviations: ACA=anterior cerebral artery; ACoA= anterior communicating artery; ICA= internal carotid artery; MCA= middle cerebral artery; VBA= vertebro-basilar aneurysms
5.1.2 Molecular microbiological and immunohistochemical findings in aneurysm walls
Using real time quantitative PCR, candidate bacteria were detected in 21/36 (58%) of aneurysms. Streptococcus and Staphylococcus species were considered as endodontic bacteria. Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, Fusobacterium nucleatum, Prevotella intermedia, Dialister pneumosintes, Parvimonas micra and Treponema denticola were considered as periodontal bacteria. (Williams et al., 1983; Doan et al., 2000; Abusleme et al., 2013)
Altogether, the prevalence of bacteria was 62% in surgical samples and 43% in autopsy samples. In the surgical group, endodontic and periodontal pathogens were identified in 15/29 (52%) and 13/29 (45%) of aneurysms, respectively. Figure 2 (page 49) shows the frequencies of bacterial DNA positive findings in all cases included in Studies I and II.
When comparing the blood samples to the aneurysm tissue samples it was notified that, within the patients, the relative amount of total bacterial DNA (using universal bacterial primers) in the aneurysm tissue samples was 44.5 times higher than that found in their control blood samples (mean; SD 44.5; 62.90). Similar results were observed in autopsy cases. In immunohistochemical studies performed with four aneurysm tissue samples obtained from autopsies, highly intensive staining of bacterial receptors CD14 and TLR-2 was observed in all studied ruptured aneurysm samples, especially at the rupture site (Figure 1).
Figure 1. The presence of receptors recognizing bacteria in aneurysm tissue samples was studied by immunohistochemical staining. This formalin-fixed histological section from an autopsy case was studied using CD14 and toll-like receptor (TLR)-2 antibodies.
Highly intensive staining (brown colour) of CD14 and TLR2 at the rupture site of an intracranial aneurysm was observed.
5.1.2 Molecular microbiological and immunohistochemical findings in aneurysm walls
Using real time quantitative PCR, candidate bacteria were detected in 21/36 (58%) of aneurysms. Streptococcus and Staphylococcus species were considered as endodontic bacteria. Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, Fusobacterium nucleatum, Prevotella intermedia, Dialister pneumosintes, Parvimonas micra and Treponema denticola were considered as periodontal bacteria. (Williams et al., 1983; Doan et al., 2000; Abusleme et al., 2013)
Altogether, the prevalence of bacteria was 62% in surgical samples and 43% in autopsy samples. In the surgical group, endodontic and periodontal pathogens were identified in 15/29 (52%) and 13/29 (45%) of aneurysms, respectively. Figure 2 (page 49) shows the frequencies of bacterial DNA positive findings in all cases included in Studies I and II.
When comparing the blood samples to the aneurysm tissue samples it was notified that, within the patients, the relative amount of total bacterial DNA (using universal bacterial primers) in the aneurysm tissue samples was 44.5 times higher than that found in their control blood samples (mean; SD 44.5; 62.90). Similar results were observed in autopsy cases. In immunohistochemical studies performed with four aneurysm tissue samples obtained from autopsies, highly intensive staining of bacterial receptors CD14 and TLR-2 was observed in all studied ruptured aneurysm samples, especially at the rupture site (Figure 1).
Figure 1. The presence of receptors recognizing bacteria in aneurysm tissue samples was studied by immunohistochemical staining. This formalin-fixed histological section from an autopsy case was studied using CD14 and toll-like receptor (TLR)-2 antibodies.
Highly intensive staining (brown colour) of CD14 and TLR2 at the rupture site of an intracranial aneurysm was observed.