Patients with esophageal cancer who undergo esophagectomy have various risk factors that can negatively impact the intestinal environment, such as malnutrition due to cancer stenosis, preoperative therapy, surgical stress, use of antibiotics, postoperative fasting, and parenteral nutrition. If the preoperative intestinal environment is maintained, postoperative infectious complications may be reduced. This study revealed that concentrations of preoperative fecal SCFAs were significantly associated with the occurrence of postoperative infectious complications. Acetic acid, propionic acid, and butyric acid make up the majority of SCFAs (>95%) . In this study, the patients with complication demonstrated significantly lower concentrations of acetic acid and propionic acid and nonsignificantly lower concentrations of butyric acid than those without complications. The concentration of lactic acid was nonsignificantly higher in the patients with complication, and the APB-L gap was a significant independent risk factor for postoperative infectious complications. These results are consistent with those of the previous study by Yokoyama et al. . In their study of patients undergoing major hepatectomy with extrahepatic bile duct resection, the preoperative fecal concentrations of acetic acid and butyric acid were significantly lower and the concentration of lactic acid was nonsignificantly lower in patients who developed postoperative infectious complications, and the acetic acid plus butyric acid minus lactic acid gap was an independent risk factor for postoperative infectious complications. Lactic acid is digested by obligate anaerobes to produce SCFAs . Yokoyama et al. speculated that a low acetic acid plus butyric acid minus lactic acid gap may reflect an impaired intestinal environment and thus lead to postoperative infectious complications .
SCFAs play various roles in maintaining the intestinal environment, and contribute to the prevention of bacterial translocation. SCFAs stimulate the proliferation and differentiation of intestinal epithelial cells and the secretion of mucus by goblet cells [16, 17]. They also maintain the acidity of the intestine and suppress the growth of harmful bacteria. Acetic acid, which is the most common organic acid, has an antimicrobial effect on harmful bacteria  and promotes the defensive functions of host epithelial cells . Furthermore, Hsieh et al. reported that acetic acid strengthened tight junctions in Caco–2 monolayers . Ohata et al. reported that propionic acid and butyric acid induced tight junction permeability in Caco–2 monolayer cells . Butyric acid is the major source of colonic epithelial cells, and induces the differentiation of regulatory T cells and ameliorates chronic intestinal inflammation in mice . Thus, maintaining the concentration of SCFAs prior to iatrogenic gastrointestinal tract damage is useful for preventing bacterial translocation.
Organic acids are mainly produced in the large intestine by obligate anaerobes. In this study, there was no significant difference between the two groups in the number of representative fecal obligate anaerobes, as shown in Table 3. This result is inconsistent with that of the prior study mentioned above . We previously reported that in patients with advanced esophageal cancer, neoadjuvant chemotherapy altered the intestinal microbiota and concentrations of organic acids . In this study, about two-thirds of patients received preoperative chemotherapy or preoperative chemoradiotherapy. When we analyzed the correlation between the preoperative number of representative fecal obligate anaerobes or the fecal concentrations of organic acids and the incidence of postoperative infectious complications only in patients who received preoperative therapy, no significant difference was found in the number of representative fecal obligate anaerobes between the two groups, while the concentrations of acetic acid and propionic acid were significantly lower in the patients with complication than in those without complication (p = 0.025 and 0.022, respectively, data not shown). The concentration of SCFAs is considered to reflect the function of the intestinal microbiota, which may be affected by preoperative therapy. Thus, the concentration of SCFAs but not the number of representative fecal obligate anaerobes was useful for predicting postoperative infectious complications in this study.
Acetic acid, which is the most common SCFA, is mainly generated by Bifidobacterium [19, 23]. In this study, the concentration of acetic acid was significantly lower in the patients with complication than in those without complication. This may be because while the number of Bifidobacterium did not differ between the two groups, the number of Bifidobacterium breve strain Yakult, which was the administered probiotics,was nonsignificantly higher in the complication (-) group. Since Bifidobacterium breve strain Yakult can selectively metabolize galacto-oligosaccharide, the production of acetic acid may be increased . In this study, although all patients received synbiotics, including Bifidobacterium breve strain Yakult, for at least 5 days before surgery, the number of preoperative fecal Bifidobacterium breve strain Yakult was nonsignificantly lower in the patients with complication than in those without complication. It is considered that different patients had different responses to synbiotics. The mechanism of this variation is unclear. Currently, we administer perioperative synbiotics to all patients with esophageal cancer scheduled to undergo esophagectomy. If a patient has low concentrations of SCFAs after receiving synbiotics, more careful attention should be paid during the perioperative period to prevent postoperative infectious complications.
Our study has several limitations. First, the number of enrolled patients was small. Second, about two-thirds of patients received preoperative chemotherapy or chemoradiotherapy, both of which alter the intestinal environment, while about one-third of patients received no preoperative therapy. In addition, the duration of synbiotic administration varied among individuals. This heterogeneity may have affected the results of our study. Third, there are no data concerning intestinal permeability, which correlates with the degree of bacterial translocation.