Baseline Subject Characteristics and Oral Clinical Indicators
A total of 100 subjects were enrolled in this study. Eight subjects in the CPC group and nine subjects of the PVP-I group dropped out (Figure. 1); thus, samples from 83 subjects, with 42 in the CPC group and 41 in the PVP-I group, were analyzed. Table 1 shows the baseline characteristics and oral clinical indicators of the subjects. None of which was significantly different between the groups (P ≥ 0.05). The results of the examinations at hospital discharge are presented in Table 2. There was no significant difference between the two groups.
Table 1. Baseline characteristics of the subjects.
|
Total
(n = 83)
|
CPC group
(n = 42)
|
PVP-I group
(n = 41)
|
P value
|
Age (years) 1)
|
65.3 ± 12.8
|
66.7 ± 10.8
|
63.8 ± 14.6
|
0.311
|
Sex (male/female) 2)
|
48/35
|
28/14
|
20/21
|
0.099
|
Smoking habit (yes/no)2)
|
57/26
|
26/16
|
31/10
|
0.178
|
Dentures (yes/no)2)
|
67/16
|
34//8
|
33//8
|
0.957
|
Number of teeth3)
|
23.63 ± 0.59
|
23.74 ± 0.74
|
23.51 ± 0.93
|
0.655
|
PPD (mm)3)
|
2.82 ± 0.07
|
2.81 ± 0.11
|
2.84 ± 0.93
|
0.675
|
BOP (%)3)
|
21.8 ± 2.5
|
23.1 ± 3.5
|
20.4 ± 3.6
|
0.402
|
OHI-S3)
|
1.50 ± 0.13
|
1.64 ± 0.18
|
1.36 ± 0.20
|
0.195
|
Tongue coat score*,3)
|
0.98 ± 0.13
|
1.07 ± 0.20
|
0.88 ± 0.16
|
0.620
|
*Calculation method of tongue coat score [tongue coat score] = [spread (score)] × [thickness (score)]
All values are expressed as the mean ± standard deviation (SD).
P-values represent the results of the between-group comparison for each item. Differences were considered statistically significant at P < 0.05: 1) t-test; 2) chi square test; 3) Mann-Whitney U test.
Abbreviations: CPC, cetylpyridinium chloride; PVP-I, povidone iodine; PPD, probing pocket depth; BOP, bleeding on probing; OHI-S, oral hygiene index score.
Table. 2 Examinations at hospital discharge.
|
Total
(n = 83)
|
PVP-I group
(n = 41)
|
CPC group
(n = 42)
|
P value
|
Period of administration of antibiotic drugs (days)2)
|
2.65 ± 0.15
|
2.8 ± 0.22
|
2.49 ± 0.20
|
0.234
|
Types of antibiotic drugs (1st generation cephem/2nd generation cephem/others)1)
|
42/33/7
|
21/17/3
|
21/16/4
|
0.375
|
Number of hospitalization days2)
|
14.5 ± 1.33
|
13.8 ± 1.68
|
15.1 ± 2.06
|
0.848
|
Start date for meals2)
|
3.65 ± 0.462
|
4.07 ± 0.645
|
3.24 ± 0.662
|
0.168
|
Surgery time (min)2)
|
282 ± 20.4
|
291 ± 30.5
|
274 ± 27.5
|
0.838
|
The frequency of mouth rinse use (days)2)
|
5.32 ± 0.180
|
4.97 ± 0.294
|
5.67 ± 0.192
|
0.118
|
All values are expressed as the mean ± standard deviation (SD).
P-values represent the results of the between-group comparison for each item; differences were considered statistically significant at P < 0.05.
1) t-test; 2) chi-square test; 3) Mann-Whitney U test.
Abbreviations: CPC, cetylpyridinium chloride; PVP-I, povidone iodine.
Time Course of Clinical Indicators and Bacteriological Data
Of note, based on the results of 16S rRNA gene sequencing, 15,000 or more sequence reads were used for analysis. After POC, bacterial flora data of the buccal mucosa and tongue coat from one subject of the CPC group were handled as missing data because of lack of the number of sequence reads. Figure 2A shows the changes in the oral clinical indicators of PPD, BOP, OHI-S, and TCI from Bef_POC to 1W_surgery based on oral examinations. In both groups, compared to Bef_POC, the OHI-S and BOP ratio significantly decreased at 1W_surgery (OHI-S: P < 0.001 for both; BOP ratio: CPC group P < 0.001, PVP-I group P = 0.001). In contrast, there was no decrease in PPD and tongue coat score from Bef_POC to 1W_surgery.
Figure 2B shows changes in total bacterial counts (log CFU/mL) in the buccal mucosa, tongue coat, and saliva from Bef_POC to 1W_surgery. Total bacterial counts in the buccal mucosa significantly decreased from Bef_POC to Aft_POC (P = 0.015), and significantly decreased from Bef_POC to 1d_surgery and through 1W_surgery (P < 0.001). For the tongue coat, although there was a significant decrease in bacterial counts from Bef_POC through 1W_surgery (P = 0.001), the change was generally smaller than that in the buccal mucosa. Similar to the buccal mucosa, total bacterial counts in the saliva significantly decreased from Bef_POC to 1d_surgery and through 1W_surgery (P < 0.001). For each site, as compared with the Aft_POC status, a significant decrease was found for the buccal mucosa and saliva on 1d_surgery and 1W_surgery (P < 0.001), whereas no change was observed for the tongue coat. There was no significant difference in bacterial counts between the groups at any sampling time point.
Figure 2C shows the changes in the number of OTUs, an alpha diversity index that represents the number of bacterial strains comprising the flora, from Bef_POC to 1W_surgery. Comparison of the number of OTUs Bef_POC and 1W_surgery showed a significant decrease at all sampling sites (P < 0.001). Similarly, compared with the Aft_POC status, the number of OTUs significantly decreased at all sampling sites (P < 0.001). As with bacterial counts, there was no significant between-group difference at any time point.
Differences in the microbial community structure between CPC and PVP-I
The bacterial taxonomic profiles were compared for each sample and at all time points (Figure 3). Only bacterial genera with an average component ratio ≥ 1% among the samples are labeled in the figure, and those with an average component ratio < 1% are grouped together as “other.” There were few changes in the microbial community structure owing to POC in either group. The use of antibiotics in both groups resulted in a significant decrease in the abundance of the genus Streptococcus on 1d_surgery. The CPC group showed a decrease in the relative abundance of the genus Streptococcus in the buccal mucosa and saliva and an increase in the relative abundance of the genus Haemophilus compared with Bef_surgery levels. This microbial community structure was maintained up to 1W_surgery. In the PVP-I group, the abundance of the genus Streptococcus decreased in the buccal mucosa and saliva on 1d_surgery and increased again at 1W_surgery.
To probe the changes in the microbial community structure in both groups, principal component analysis (PCA) was performed between the Bef_POC and 1W_surgery time points (Figure 4). In the CPC group, the microbial community structure was significantly altered in all samples (PERMANOVA, buccal mucosa: P = 0.001, tongue: P = 0.001, saliva: P = 0.001). In the PVP-I group, the microbial community structure changed significantly in all samples except for the buccal mucosa (PERMANOVA, buccal mucosa: P = 0.086, tongue: P = 0.022, saliva: P = 0.002). LefSe analysis was performed to identify the characteristic bacterial genera between the CPC and PVP-I groups (Figure 5). The genera Streptococcus, Rothia, Porphyromonas, Cryptobacterium, Actinomyces, Peptostreptococcus, Bergeyella, Catonella, and Eikenella in the buccal mucosa; Streptococcus, Actinomyces, Corynebacterium, Bergeyella, and Syntrophomonadaceae_[VIII][G-1] in the tongue; and Streptococcus, Rothia, Porphyromonas, Actinomyces, Peptostreptococcus, Agrobacterium, Bergeyella, Eikenella, Atopobium, Catonella, and Gentipeda in the saliva were significantly enriched in the PVP-I group at 1W_surgery (LDA > 2, P < 0.05). The relative abundances of the genera Streptococcus and Haemophilus were not significantly different between the two groups at baseline. However, after POC, Haemophilus and other bacteria characteristic of the CPC group were detected in the buccal mucosa. In the saliva, the genus Lachnoanaerobaculum was detected in the CPC group and Kingella was detected in the PVP group (Additional file 1: Supplementary Figure 1). From Bef_surgery to 1W_surgery, the most characteristic bacteria observed in the buccal mucosa and saliva were the genus Haemophilus in the CPC group and the genus Streptococcus in the PVP-I group (Additional file 1: Supplementary Figures 2, 3, and 4). Interestingly, tongue samples showed no significant difference in the abundance of Streptococcus between the CPC and PVP-I groups in the Aft_surgery and 1d_surgery comparison.