Influence of periodontal treatment on blood microbiotas: a clinical trial

Study design and participants

The present single-blinded randomized controlled trial (RCT) was reviewed and approved by the Research Ethics Committee at the Shandong Stomatological Hospital, China (Ref No: 20191003). The RCT was conducted in accordance with the Helsinki Declaration of 1975 (revised in 2013) and registered in the Clinical Trial Registry (http://www.chictr.org.cn, ChiCTR1900021599). The study participants were recruited from the Department of Periodontology at Qingdao Stomatological Hospital from September 2018 to May 2019 according to the predefined selection criteria (Table 1). An informed written consent form was obtained from all participants.

Fifty-six patients were initially evaluated for eligibility and 29 patients were excluded for various reasons. Twenty-seven patients diagnosed with generalized stage II, III, and IV periodontitis (Papapanou et al., 2018) were enrolled in this three-arm RCT (Fig. 1). Assessment of the sample size was based on power calculation for a two-tailed comparison (G*Power 3.1) (Faul et al., 2007; Wennström, Dahlén & Ramberg, 2011). The results indicated that the inclusion of 27 patients (considering a 15% drop-out risk) would allow for the detection of a mean difference of 0.5 mm in probing depth (PD) reduction between treatments, with 80% power and 5% alpha error.

Process, randomization, and interventions

For each participant, baseline data collection, periodontal charting, and blood sample collection were performed at the first clinical visit before randomization and treatment (Fig. 1). Participants were randomized into a control group (Group A; n = 9) and two experimental groups (Groups B1 and B2; n = 10 for each group) with a random number table. Patients in the control group received supragingival ultrasonic scaling at the baseline visit, followed by subgingival full-mouth SRP 1 week later. The first intervention group (Group B1) received plaque staining, supragingival GAP, and supragingival ultrasonic scaling at the baseline visit, followed by full-mouth SRP and subgingival GAP at the 1-week visit. Participants in the second intervention group (Group B2) received the same therapy as the Group B1 participants; however, subgingival GAP was performed right before full-mouth SRP at the 1-week visit. Subgingival GAP was performed at sites with PD ≥ four mm (Fig. 2).

Supragingival and subgingival air polishing were performed using 65 µm and 25 µm glycine amino acid powder, respectively (Air-Flow Polishing Soft Powder; EMS, Nyon, Switzerland). For subgingival treatment, a specially designed nozzle (Perio-Flow, Nozzle, EMS, Switzerland) was inserted into the periodontal pocket. All periodontal procedures including full-mouth SRP were performed by the same periodontist using manual and ultrasonic instruments.

Periodontal assessment

Clinical periodontal examinations were performed two times—-at the one-week and 6-week follow-up visits after supragingival scaling to collect the following data: (1) PD (Lu et al., 2018), plaque index (PLI) (Silness & Loe, 1964), bleeding index (BI) (Lu et al., 2018), and bleeding on probing (BOP) (Bundidpun, Srisuwantha & Laosrisin, 2018). The examining periodontist was blinded to the patient assignments.

Sample collection and DNA extraction

Peripheral blood samples were obtained by venipuncture in the antecubital fossa before treatment at the baseline as well as the next-day and 6-week follow-up visits. Before sampling, the skin overlying the vein was disinfected using 75% ethyl alcohol and 2% chlorhexidine to minimize skin contaminants. Each sample (3 ml whole blood) was stored at −80 °C immediately after collection until laboratory analysis. Bacterial genomic DNA was extracted from 200 ul of whole blood using a commercial kit (TianGen, TIANamp, China). The concentration and purity of DNA were assessed with the Micro Drop Ultra Micro Spectrophotometer (Bio-DL, TX, USA). Double distilled water was used as the negative control during the extraction process.

Nested polymerase chain reaction (nested PCR) amplification and 16SrRNA Gene Sequencing. Universal bacterial primers (27F 5′AGAGTTTGATCCTGGCTCAG-3′1492R5′-TACGGYTACCTTGTTACGACTT-3′) (Fredriksson, Hermansson & Wilén, 2013) were used to detect bacterial DNA in the blood samples. PCR amplification was carried out in a 25 µl reaction mixture containing 2 × Taq Plus Master Mix (KAPA 2G Robust HotStart ReadyMix, Sigma-Aldrich, USA), 5 µM of the forward and reverse primers, 6 ng bovine serum albumin (BSA), and 30 ng of template DNA. A 2 kb DNA ladder was used as a size marker to assist the analysis of the PCR products. Subsequently, nested PCR was performed using the universal primers (336F 5′-GTACTCCTACGGGAGGCAGCA-3′ 806R 5′-GTGGACTACHVGGGTWTCTAAT-3′) (Wei et al., 2016) targeting the V3-V4 region of the 16SrRNA gene. Doubled distilled water was also used as the negative control.

The final PCR products were separated by 2% agarose gel electrophoresis. The ∼470 bp fragments obtained were purified using a QIAquick Gel Extraction Kit (Qiagen, Hilden, Germany) and sequenced on a MiSeq platform (Allwegene, China).

Statistical analysis

Descriptive analyses of the quantitative variables were performed and the results are presented as the mean ± standard deviation (SD). The median, minimum, and maximum values were calculated for the BI and PLI. All parameters were tested for normality using the Shapiro–Wilk method. Repeated measures analysis of variance (ANOVA) was used to compare longitudinal changes in the patients in the three groups with a normal distribution. If the parameters did not follow a normal distribution, the Friedman non-parametric test was performed. All the statistical analyses were performed using the SPSS ver. 24.0 software (IBM Japan, Tokyo, Japan).

Sequences were analyzed using the pipeline tools QIME v 1.8.0 and Mothur v 1.34.4. Barcodes and primers were trimmed and raw reads with a quality value <20 were removed. The Mann–Whitney U-test was used to compare the bacterial species diversity (α-diversity; measured with the Chao-1 index, observed_species, PD_whole_tree, and Shannon index). Principal component analysis (PCA) based on the community structure (β-diversity) was conducted according to the distance matrix. Analysis of similarities (ANOSIM) was performed to compare the intra- and inter-group similarities based on the UniFrac distance. Microorganism features to distinguish the blood microbiotas specific to periodontitis before and after treatment were identified using a linear discriminant analysis (LDA) effect size (LEfSe) threshold of 2.0. Metastats analysis was performed to compare the relative abundance of bacteria pre- and post-treatment. All tests were two-tailed. A p-value <0.05 was considered to indicate statistical significance.

Clinical data

Of the 29 patients included in the study, 27 completed the 8-week trial (Table 2). Two participants from Group B1 dropped out due to business conflicts. The ages of the 27 patients ranged from 28 to 61 years (mean age, 41 years). Over half (55.6%) of the patients were male and most patients (63%) were diagnosed with stage III periodontitis. There were no significant differences in the study groups in terms of the participants’ age, gender, and periodontitis stage.

Most of the clinical parameters indicated significant improvements in the three groups after treatment (P < 0.05). The percentage of BOP-positive, PD, and BI = 3 and 4 sites decreased sharply and the percentage of BI = 1 and 2 sites increased significantly at the 6-week visit after treatment (Table 3). However, there was no statistically significant difference among the groups in all clinical parameters between the baseline and 6 weeks after full-mouth SRP.

Nine patients (33.3%) complained about systemic symptoms the day after the full-mouth SRP. These symptoms consisted of chills, low grade fever (no higher than 38 °C), headache, fatigue, and drowsiness, which persisted for no longer than 24 h before disappearing completely.

Semiquantitative analysis and general information for 16SrRNA gene sequencing

Eight participants were prescribed antibiotics due to the common cold within 1 week prior to the 6-week follow-up visit and their blood samples from the baseline and the day after treatment were analyzed. The total bacteria in 75 blood samples were evaluated using nested PCR (Fig. 3). Positive PCR gel strips indicated the presence of the 16SrRNA gene in the blood and brighter strips indicated a higher amount. The brightness levels of the gel strip are summarized and the darkest strips were found in Group B2 the day following the full-mouth SRP. Additionally, 70% of the gel strips were obviously less bright 1 day post-treatment (Table 4). The negative control (double distilled water) yielded no PCR products (Fig. 3).

A total of 37 blood samples were selected (12 from Group A, 11 from Group B1, 14 from Group B2) for further high-throughput 16SrRNA gene sequencing. A total of 1,252,175 unique sequences and 383 operational taxonomic units were obtained, with a mean value of 27,826 clean tags/sample. In total, there were 14 phyla, 32 classes, 54 orders, 97 families, 171 genera, and 41 species taxa in these blood samples.

Variations in blood microbiota at 6 weeks after periodontal treatment

The present study revealed that 99% of the bacteria in human blood are Proteobacteria, Firmicutes, Bacteroidetes, and Actinobacteria at the phylum level (Fig. 4A). The 23 most common genera at the baseline and 6 weeks after treatment are shown in Fig. 4B. The Methylobacterium genus had the highest relative abundance in blood relative to periodontitis. A comparison of the relative abundance, α-diversity, and β-diversity at the baseline and 6 weeks after treatment with a threshold filter of 1% revealed no statistical differences in the blood microbiota in the experimental groups. LEfSe analysis indicated that Acinetobacter calcoaceticus was the dominant species 6 weeks after treatment (Fig. 4C). These findings suggest there was no difference in the richness and community structure of blood microbiota at 6 weeks following the local periodontal treatment compared to the baseline levels.

Impact of treatment on the blood microbiota the day after treatment

The day after full-mouth SRP, groups A (conventional full-mouth SRP) and B2 (subgingival GAP right before full-mouth SRP) showed no significant differences in the α-diversity and β-diversity compared to the baseline. On the contrary, Group B1 (subgingival GAP right after full-mouth SRP) exhibited higher α-diversity the day after full-mouth SRP than at the baseline (Chao-1 index, P = 0.03) (Fig. 5A). The relative abundance of the Allicycliphilus genus decreased 13-fold (0.47% vs. 6.1%, P = 0.019) and six-fold (1.5% vs. 0.25% P = 0.028) in Groups A and B1, respectively. The Methylobacterium genus increased nine-fold (1.4% vs. 4.9%, P = 0.049) and the Streptococcus-pneumoniae species decreased 47-fold (7% vs. 0.3%, P = 0.002) in Group B2 the day after full-mouth SRP compared to the baseline values (Fig. 5B). LEfSe analysis indicated that Porphyromonas and Pantoea were the dominant genera in Group B1 the day after full-mouth SRP relative to the baseline (Fig. 5C). These findings suggest that periodontal treatment can alter blood microbiota diversity and relative abundance the day following local periodontal treatment.