This study examined the gastric microbiota prepared from the gastric fluid of patients who underwent either BI or RY reconstruction after DG for gastric cancer, using 16S rRNA sequencing analysis. To our knowledge, this is the first report to compare the composition of gastric microbiota prepared from gastric fluid in BI and RY reconstructions. We have experience in gastric fluid research [21], and this study used gastric fluid as the sample. Our results indicate a change in the gastric microbiota before and after gastric resection; interestingly, no significant differences were observed between the two different reconstructive methods in terms of diversity and community composition.
The alpha diversity of the gastric microbiota after DG was significantly lower than that before DG. After DG, the gastric environment changes dramatically; for example, there is reduced gastric acid secretion due to a reduction in the fundic gland region and bile regurgitation. In contrast to our results, Tseng et al. reported that in six cases (deemed to be a relatively small sample size) with Billroth II reconstruction, the gastric microbiota was significantly more diverse after DG [7]. They employed a different methodology from that utilized in this study, in that the sample was retrieved from the gastric mucosa and sample collection was performed 2 years after DG.
It has been recently reported that the diversity and composition of the gastric microbiota differ significantly according to the location: whether around the tumor, peritumor, or in the normal region of the stomach [22]. It was thought that after stomach resection for gastric cancer, whereby the tumor and peritumor area would have been resected, then the remnant stomach would naturally comprise only the normal region. Therefore, only the gastric microbiota of the original normal region would be retained in the microbiota present in the remnant stomach. However, our technique collecting the gastric fluid can represent the inclusive microenvironment of the stomach.
We compared the gastric microbiota after DG with and without postoperative adjuvant chemotherapy. Since there was no significant difference in both alpha and beta diversity (Supplementary Figs. S1, S2), we judged that the effect of postoperative adjuvant chemotherapy on the gastric microbiota was small and included cases with postoperative adjuvant chemotherapy in our study.
In our investigation, the genera Helicobacter, Prevotella, Fusobacterium, and Peptostreptococcus were significantly reduced after DG, and these genera were reported to be relatively abundant in the tumor and peritumor regions of the stomach [22].
The beta diversity exhibited a significant difference before and after DG. As a result of changes in the gastric environment due to the altered gastrointestinal tract, there is an apparent shift in the community composition of the gastric microbiota after DG. Therefore, the gastric microbiota differed before and after surgery. This result is consistent with those of a previous report [7]. The beta diversity of the gastric microbiota significantly increased with the use of acid-suppressive agents, including proton-pump inhibitors [23]. This situation is considered relatively similar to the environment after DG due to reduced gastric acid secretion. Therefore, it is in agreement with our beta diversity results.
In our study, the microbiota of the remnant stomach was dominated by four phyla: Proteobacteria, Firmicutes, Bacteroidetes, and Actinobacteria (Fig. 1), with a significantly higher abundance of Firmicutes (Fig. 5b). Tseng et al. also reported that the stomach was dominated by the same four phyla after DG, with a significantly higher abundance of Firmicutes [7]. Our LEfSe analyses revealed that the proportions of genera Rothia and Lactobacillus were significantly higher after DG (Fig. 5c). Both genera, Rothia and Lactobacillus, are oral commensals, and it is believed that the oral microbiota moves into the remnant stomach and coexists.
No significant differences in diversity and community composition between BI and RY reconstructions were observed in this study. Many reports have compared BI and RY reconstruction after DG in terms of quality of life and dysfunction [24–33]. The incidence of remnant gastritis and bile regurgitation is higher after BI than after RY reconstruction [24,25,27,28,30–32]; the procedures are generally equivalent regarding postoperative quality of life [26,27,30,31,33].
It has been established that there is a discrepancy between endoscopic findings and a patient’s symptoms. The gastric microbiota of patients with functional dyspepsia is altered compared to that in healthy controls [34]. Alterations in the gastric microbiota are thought to cause postprandial distress and epigastric pain syndrome. Our finding of no difference in the bacterial microbiota between BI and RY reconstruction means that the gastric environment was comparable, possibly suggesting that both reconstructive methods were generally equivalent in terms of postoperative quality of life. However, since there are few published reports in this field and much remains unknown, further research on this topic is needed.
The present study has several limitations. First, we aspirated gastric fluid and examined the gastric microbiota, but the amount of gastric fluid that could be aspirated was limited, which could have caused individual differences. Although some reports collected gastric mucosal tissue, the gastric mucosal microbiota differs based on the collection site (e.g., the normal, peritumor, and tumor sites) [6,7,22]. Gastric fluid is spread evenly throughout the stomach. In this respect, gastric fluid aspiration has the advantage of evenly reflecting the gastric environment. In addition to the difference in the collected samples between this study and others, the measurement times are different. It is unknown when the bacterial microbiota changes; therefore, following the alterations in the bacterial microbiota over time is necessary. Knowledge of these changes and how they are affected by fluctuations in the gastric environment is of importance because it may potentially hold the keys to improved treatment approaches. Finally, in approximately half the patients, H. pylori was detected in the gastric fluid and was not unified. However, the percentage of H. pylori in the gastric fluid after DG is extremely low; hence, we believe that it has little effect on the comparison of gastric microflora with stomach microflora based on the method of reconstruction.