Portal hypertension can be treated with transjugular intrahepatic portal shunt (TIPS). Hepatic encephalopathy (HE) is one of the main complications after TIPS with an average incidence rate of 20–50% [1–6]. The occurrence of HE after TIPS increases the readmission rate and the healthcare costs, impacting patient and family quality of life [7]. There is currently no effective strategy to prevent HE after TIPS.
Previous studies have reported that HE after TIPS is associated with changes in gut microbiota [8] and sarcopenia [9,10]. Probiotics may mitigate muscle loss and modulate gut microbiota [11,12]. Clostridium butyricum is a probiotic that produces butyric acid, which can improve cognitive status [13], repair intestinal mucosal barriers [14], lower endotoxin levels in the body [15], and alleviate sarcopenia [8]. There is currently a lack of research on probiotics for preventing HE after TIPS. Therefore, we aimed to explore the efficacy of Clostridium butyricum in preventing postoperative HE in patients with portal hypertension undergoing TIPS.
2Materials and Methods2.1PatientsFrom April 2023 to November 2024, 98 liver cirrhosis patients undergoing TIPS were enrolled. Inclusion criteria: 18–70 years old, liver cirrhosis, TIPS for variceal bleeding and/or refractory ascites. Exclusion criteria: prior HE, mental disorders, cancer, uncontrolled infections, recent gastrointestinal surgery (within 3 months), other digestive diseases affecting microbiota, previous TIPS.
2.2Grouping and treatmentBased on the previous studies, the incidence rate of HE was assumed to be 40% in the control group and 15% in the experimental group [1,16]. Using PASS 15.0 software with a two-sided α of 0.05 and a β of 0.2 (power=80%), and accounting for a 5% dropout rate, a sample size of 49 patients per group (98 in total) was calculated. A total of 98 patients were randomized in a 1:1 ratio using simple randomization. Patients in the experimental group received Clostridium butyricum and lactulose, whereas the control group received only lactulose. Clostridium butyricum tablets (3.5 × 105–3.5 × 108 CFU/tablet, Miyalsan Pharmaceutical Co., Ltd., Japan, Approval Number: SJ20140080) 2 tablets, 3 times per day. Lactulose (667 mg/mL, Abbott Laboratories, Approval Number: HJ20171057) 15 mL, 3 times per day. Oral administration started 24 hours after TIPS and continued until 12 weeks after TIPS. Monthly telephone follow-ups were conducted to evaluate medication adherence, with verification by reconciling medication counts at in-person clinic visits.
2.3Data collection and follow-upPatients were followed up at baseline, 1 month, and 3 months after TIPS to collect data on laboratory indicators and the third lumbar skeletal muscle index (L3-SMI). The incidence of hepatic encephalopathy (HE) will be monitored, encompassing both minimal hepatic encephalopathy (MHE) and overt hepatic encephalopathy (OHE). OHE was defined as HE grade 2 or higher according to West Haven modified criteria [17], with grade 2 including isolated asterixis. The measurement of MHE was based on the number connection test-A (NCT-A) and the digit symbol test (DST), which use reference values standardised among the Chinese population [18]. Two researchers will measure the skeletal muscle area at the lumbar level on CT scans to calculate the L3-SMI (muscle area at the third lumbar vertebra divided by height squared). Sarcopenia is defined as L3-SMI < 42cm²/m² for men and < 38cm²/m² for women [19].
2.4Ethical considerationsThis study has been approved by the Ethics Committee of Beijing Youan Hospital Affiliated to Capital Medical University (LL-2023-022-K) and registered for clinical trials (ChiCTR2300070406) .
2.5ELISA testPeripheral blood samples were collected from the Clostridium butyricum plus lactulose group (n = 14) and the lactulose group (n = 10) at baseline, 1 month, and 3 months after TIPS. We employed ELISA kits JM-03204H1 for Interleukin-6 (IL-6), JM-03277H1 for tumor necrosis factor-α (TNF-α), EU3126 for lipopolysaccharides (LPS), EH1889 for D-lactate dehydrogenase (D-LDH), and EH1546 for diamine oxidase (DAO) to quantify the levels of these biomarkers.
2.616S rRNA sequence and short-chain fatty acid detectionFecal samples were collected from patients in the Clostridium butyricum plus lactulose group (n = 14) and the lactulose group (n = 10) at baseline, as well as 1 month and 3 months after TIPS. While patient participation was voluntary, the final sample sizes were determined by the availability of complete datasets for all scheduled follow-up visits. Genomic DNA from fecal samples was extracted and tested for concentration and purity. The subsequent library construction and sequencing process was commissioned to Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China). Based on the sequencing of fecal microbiota, ASV representative sequences and abundance information were analyzed on the Majorbio platform, including Alpha diversity analysis, Beta diversity analysis, community composition analysis, and species diversity analysis. P<0.05 indicates statistical significance differences. We commissioned Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China) to detect short-chain fatty acids in fecal samples using gas chromatography-mass spectrometry from the Clostridium butyricum plus lactulose group (n = 14) and the lactulose group at baseline and 3 months post-TIPS (n = 10).
2.7Data analysisThe data was analyzed using SPSS 23.0. The primary endpoint was the occurrence of HE, for which there were no missing data. A few missing laboratory values were imputed using values from adjacent time points. Based on data type and characteristics, we used mean, standard deviation, and median quartiles for representation. For quantitative data between two groups, we applied independent t-tests or non-parametric tests. For categorical variables, when the T-value was greater than or equal to 5, chi square test was used. When at least one T-value was greater than or equal to 1 and less than 5, Fisher's exact test was appplied. Within-group quantitative data at different time points were analyzed by repeated measures ANOVA. The cumulative incidence of HE three months post-TIPS was assessed using Kaplan-Meier analysis and Cox regression analysis. The analysis was performed on all included patient data according to the intention-to-treat principle. To evaluate the effectiveness of randomization, we conducted inter group balance tests on all baseline features. If the test results showed a statistically significant difference (P<0.05) in a baseline feature between two groups, it was considered a confounding factor and adjusted for analysis in subsequent multivariate models.
3Results3.1Patient characteristicsFrom April 2023 to November 2024, 98 patients who underwent TIPS met the inclusion and exclusion criteria. These patients were randomly divided into two groups. The patient screening process was shown in Fig. 1. The age, gender, liver disease etiology, TIPS surgical indications, and surgical procedures of the two groups of patients were comparable between the two groups. The general information of the two groups was shown in Table 1. All patients received covered stents with a diameter of 8 mm. No significant infections were observed before or after the TIPS. The stents remained patent during the three months after TIPS, and patients adhered to a low-protein diet for two weeks. For patients with spontaneous portal shunt, occlusion has been performed during TIPS.
Demographics and clinical characteristics of patients at baseline
Note: HBV, hepatitis B virus; HCV, hepatitis C virus; PBC, primary biliary cholangitis; NAFLD, non-alcoholic fatty liver disease; WBC, white blood cell; HGB, red blood cell; PLT, platelet; ALT, alanine aminotransferase; AST, aspartate aminotransferase; γ-GT, glutamyl transpeptidase; ALP, alkaline phosphatase; ALB, albumin; TBIL, total bilirubin; eGFR, estimated glomerular filtration rate; NH3, plasma ammonia; PTA, prothrombin activity; NLR, Neutrophil-lymphocyte Ratio; NCT-A, number connection test A; DST, digit symbol test.
MHE incidence was the same between groups at 1 and 3 months after TIPS. However, OHE incidence differed significantly, with the Clostridium butyricum plus lactulose group exhibiting a significantly lower incidence compared to the lactulose group (8.2% vs. 26.5% at 1 month, P = 0.016; 10.2% vs. 34.7% at 3 months, P = 0.004), as depicted in Fig. 2. The severity and frequency of HE in Clostridium butyricum plus lactulose group was significantly lower than those treated with lactulose alone, as shown in Table 2. Kaplan-Meier analysis revealed a higher probability of OHE in the lactulose group compared with Clostridium butyricum plus lactulose group (HR=3.867, P = 0.008), as shown in Fig. 2. After adjusting for relevant factors using Cox regression, HR=4.819, as shown in Table 3.
The occurrence of OHE in Clostridium butyricum plus lactulose group and the lactulose group after TIPS
The Cox regression analysis of OHE within 3 months after TIPS
There were no significant differences in common laboratory indicators between Clostridium butyricum plus lactulose group and lactulose group at baseline, 1 month, and 3 months after TIPS, as shown in Supplementary Table 1. However, at 3 months, the L3-SMI of the Clostridium butyricum plus lactulose group was significantly higher than that of the lactulose group (42.85 ± 10.66 vs. 38.38 ± 8.61 cm²/m2, P = 0.024), as shown in Supplementary Fig. 1. Repeated measures ANOVA revealed significant intra-group effects and interaction, but not inter-group effects for L3-SMI, as shown in Supplementary Table 2. L3-SMI in Clostridium butyricum plus lactulose group gradually increased over time and was significantly higher than the lactulose group at 3 months, as shown in Supplementary Fig. 1. No adverse drug reactions were reported in either group. When comparing within the group, we also found that the levels of albumin and prealbumin in the lactulose group after TIPS were significantly reduced compared to baseline. Although the Clostridium butyricum plus lactulose group also showed a decreasing trend, it did not reach statistical differences, as shown in the Supplementary Figs. 2 and 3.
3.4Correlation between sarcopenia and overt hepatic encephalopathy occurrence after transjugular intrahepatic portal shuntAccording to L3-SMI at 3 months after TIPS, patients were divided into those with sarcopenia and those without sarcopenia. The incidence of OHE after TIPS within 1 month and 3 months in patients with sarcopenia were both significantly higher than patients without sarcopenia, as shown in Supplementary Fig. 4. We further analyzed the incidence of OHE between sarcopenia and without sarcopenia on the basis of different interventions as shown in Supplementary Table 3. We found that in the lactulose group the incidence of OHE in patients with sarcopenia was significantly higher compared to the patients without sarcopenia at 3 months after TIPS (48.3% vs. 15%, P = 0.016). Among patients with sarcopenia at baseline, the incidence of OHE at 3 months after TIPS was significantly lower in the Clostridium butyricum plus lactulose group than in the lactulose group (15.0% vs. 48.3%, P = 0.016).
3.5Changes in gut microbiota and short-chain fatty acidsWe performed 16S rRNA detection on the feces of 14 Clostridium butyricum plus lactulose group and 10 lactulose patients at baseline, 1 month and 3 months after TIPS. The baseline data of patients were comparable, and the incidence of OHE after TIPS and the dynamic changes in clinical indicators were consistent with the overall patient data trend, as shown in Supplementary Table 4. There was no significant difference in alpha diversity and beta diversity between the two groups of patients at different time points as shown in Fig. 3. When analyzing the changes in the patient's own microbiota through paired testing, it was found that compared to baseline, the abundance of Veillonellaceae and Clostridiaceae in the postoperative feces of patients treated with Clostridium butyricum and lactulose increased, while the abundance of Enterococcus and Enterobacteriaceae decreased compared to baseline. In the lactulose group, the abundance of Bifidobacterium and Lactobacillus in postoperative patients' feces increased compared to baseline as shown in Fig. 4. The changes of gut microbiota in the level of genus between two groups were shown in the Supplementary Fig. 5.
When comparing the levels of butyric acid in the feces of two groups at different time points, it was found that there was no significant change in the levels of each butyric acid in the feces of both groups at the same time as shown in the Supplementary Table 5. However, the butyric acid content in the Clostridium butyricum plus lactulose group was significantly higher than the baseline level at 3 months after TIPS. At the same time, the content of isobutyric acid in the lactulose group was significantly lower than the baseline level at 3 months after TIPS, as shown in Fig. 5. Afterwards, several indicators related to intestinal barrier and inflammation were tested, and it was found that at 3 months after TIPS, the levels of TNF-α and D-LDH in the Clostridium butyricum plus lactulose group were significantly lower than those in the lactulose group as shown in Supplementary Table 6 and Fig. 6.
TIPS effectively reduces portal vein pressure and improves outcomes for patients with refractory ascites or acute esophageal variceal bleeding, especially in advanced liver cirrhosis. It significantly increases 1-year transplant-free survival compared to traditional treatments and is widely used for drug-resistant portal hypertension complications [20,21]. The main complication after TIPS is the occurrence of HE, but currently, there is a lack of ideal prevention strategies. The 2022 BAVENO VII consensus on portal hypertension [22] and the 2022 EASL clinical guidelines for the management of HE [23] both recommend lactulose as the first-line drug for the treatment of HE.
The occurrence of HE after TIPS is to some extent related to the establishment of diversion channels. Blood from the intestinal tract that has not been detoxified by the liver directly enters the systemic circulation, so the gut microbiota plays an important role in the occurrence of HE after TIPS. A meta-analysis of 21 studies [24] showed that the probiotic group had certain advantages in treating and improving HE symptoms compared to the control group, which could reduce the blood ammonia concentration of subjects. However, it is still uncertain whether it is superior to lactulose. Sharma P [25] found no significant difference in efficacy among the three groups when comparing lactulose, probiotics, and their combination therapy for MHE.
At present, the main drugs for preventing HE after TIPS are lactulose and rifaximin. Research on probiotics for preventing HE after TIPS is limited, and the efficacy is still unclear. In the study by O Riggio et al. [26], no significant differences were found in the incidence of HE and blood ammonia levels among the lactulose, rifaximin and control groups at one month after TIPS (P = 0.97). In Leon Louis Seifert's study [27], it was found that there was no significant difference in the incidence of HE between patients receiving lactulose and those without prophylaxis after TIPS. Compared with patients who received only lactulose or no prophylaxis, patients who received both lactulose and rifaximin simultaneously had a significantly lower incidence of HE within 12 months (28.1% vs. 52.1%, P = 0.004). In a study by Christophe Bureau [28], rifaximin was found to prevent HE for 6 months post-TIPS (35.3% vs. 55.5%, P = 0.008).
Lactulose, an unabsorbed disaccharide, acidifies the intestine, reduces ammonia absorption, promotes beneficial bacteria growth, and acts as a laxative. Clostridium butyricum produces short-chain fatty acids, primarily butyric acid, energizing intestinal mucosal cells, repairing the intestinal barrier, reducing inflammation, and regulating intestinal immune homeostasis. The relationship between gut microbiota and HE has received increasing attention [29,30]. Therefore, this study used lactulose as the control group to explore the preventive effect of Clostridium Butyricum combined with lactulose on HE after TIPS. Our research results found the incidence of MHE did not differ significantly between the two groups at 1 month and 3 months following TIPS. However, the incidence of OHE was significantly lower in the Clostridium Butyricum plus lactulose group compared to the lactulose group. Additionally, the severity and frequency of HE episodes in the Clostridium Butyricum plus lactulose group were reduced relative to the lactulose group, aligning with the findings reported by Manish Kumar Lunia [31].
There was no significant difference in common clinical indicators between the experimental group and the control group, except for the L3-SMI. Therefore, the decrease in OHE incidence in the experimental group was not only related to changes in gut microbiota, but also to an increase in L3-SMI. Because muscle tissue can participate in ammonia metabolism and transport [32]. A study has found that a decrease in L3-SMI increases the risk of OHE (40.8% versus 18.6%, P = 0.011) and death (log-rank P = 0.001) after TIPS [33]. In our study, we also found that patients with sarcopenia had a higher incidence of OHE than those without sarcopenia. The increase in L3-SMI may be related to the intervention of butyrate-producing probiotics [34]. Research has determined the level of butyrate in patients with sarcopenia was reduced and further confirmed that butyrate may stimulate C2C12 muscle cell proliferation by promoting G1/S cell cycle transition through activating the ERK/MAPK pathway [35]. Another randomized controlled study conducted in the elderly found that dietary supplementation with butyrate significantly increased L3-SMI (P = 0.001) and improved muscle function in the intervention group compared to the control group [36].
To further investigate the effects of combining lactulose with Clostridium butyricum on the composition and function of gut microbiota. According to the order of enrollment, fecal samples were collected from 14 patients in the Clostridium butyricum plus lactulose group and 10 patients in the lactulose group at baseline, 1 month and 3 months after TIPS, and 16S rRNA gene Sequencing was performed. The results showed that there was no significant difference in Alpha diversity and Beta diversity between the two groups of patients, which may be related to the limited number of samples sent for testing in each group. On the other hand, lactulose, as a prebiotic, may also affect the gut microbiota, making it difficult to detect the additional effects of Clostridium butyricum. A study on animals found that Clostridium butyricum had no significant effect on the Alpha and Beta diversity of gut microbiota, which was consistent with the results of this study [37].
Pairing tests within the same group revealed that after TIPS, the abundance of Enterococcaceae, Enterobacteriaceae, and Enterococcus in the Clostridium butyricum plus lactulose treatment group was significantly lower than baseline levels, whereas the abundance of Clostridiaceae, Lachnospira, Ruminococcus, and Clostridium_sensus_stricico_1 was significantly higher. Changes in the abundance of Enterococcus and Enterobacteriaceae can affect the production and metabolism of ammonia and are believed to be associated with impaired cognitive function. Previous studies on treating MHE due to hepatitis B cirrhosis with Clostridium butyricum and bifidobacteria [38] found a reduced abundance of Enterococcus and Enterobacteriaceae, consistent with the findings of this study. Lachnospira and Ruminococcus belong to Lachnospiraceae and Ruminococcus, respectively, and both belong to the Firmicutes phylum. They are also butyrate producing bacteria and are believed to be associated with good cognitive function. In a study on mice, it was found that supplementing with Clostridium butyricum can increase the abundance of Ruminococcaceae[39], which was consistent with the findings of this study. In the lactulose group, the abundance of Bifidobacterium and Lachnospira increased after TIPS compared to baseline, but the abundance of some genera in Ruminococcaceae decreased significantly compared to baseline, and no changes in the abundance of Clostridium_sensus_strico_1 were observed. The changes in Bifidobacterium may be related to the use of lactulose [40]. Therefore, the above research results indicate that supplementing Clostridium butyricum with lactulose can to some extent increase the abundance of butyrate producing bacteria in feces.
Further testing of fecal samples revealed that supplementation with Clostridium butyricum significantly increased butyric acid content compared to baseline, consistent with the increased abundance of Clostridium butyricum. At the same time, it was found that at 3 months after TIPS, the levels of TNF-α and D-lactate dehydrogenase in the group treated with Clostridium Butyricum plus lactulose were significantly lower than those in the lactulose group. Clostridium butyricum can reduce inflammatory levels to a certain extent [14], which may be related to the repair of intestinal barriers and regulation of immune function. Although current research suggests that the effect of supplementing Clostridium butyricum is related to butyrate, there is still a lack of direct evidence.
There were also some shortcomings in this study: The number of fecal samples submitted for testing was limited. The accuracy of 16S rRNA detection is limited and cannot be analyzed at the species level. No further analysis was conducted on the relationship between microbiota and patient prognosis due to the lack of long-term follow-up for the patient. Finally, for ethical considerations, this study did not establish a blank control group that did not receive any intervention or a treatment group that only used Clostridium butyricum.
5ConclusionsThe combination of lactulose and Clostridium butyricum can significantly reduce the incidence of HE after TIPS, increase the abundance of beneficial bacteria in feces and reduce the abundance of harmful bacteria. It can also increase the level of butyric acid in feces. To some extent, improve the inflammatory and nutritional status of liver disease patients (Fig. 7).
Patients with cirrhosis who received TIPS were randomly divided into two groups. One group received Clostridium butyricum and lactulose (experimental group), while the other group received only lactulose (control group). The incidence of OHE in the experimental group was significantly lower than that in the control group at 1 and 3 months after TIPS. Moreover, the experimental group showed an increase in beneficial bacteria in their feces 3 months after TIPS, had a lower peripheral blood inflammatory level than the control group, and a higher L3-SMI compared to the control group.
The authors declare that financial support was provided by the Scientific Research Project of Beijing Youan Hospital, CCMU. 2022. (no. BJYAYY-YN2022-14) as well as the development project of nutritional intervention products based on the assessment of cancer patients' nutritional status in the 2023 Beijing Municipal Health Commission Science and Technology Promotion Plan Matching Project (3-1-03-152-01).
Author contributionsThe surgical procedure was performed by TZ, CJ, KZ and JL. XX, TZ and CJ contributed to the data collection and patient follow-up. XX contributed to the data analysis and writing of the manuscript. The topic was conceptualized by XX, JL and QM. The article was reviewed by JL and QM.
None.
