This prospective cohort study included five institutional cohorts of patients referred to outpatient clinics (Yokohama City University Hospital, Kanagawa Dental University Yokohama Clinic, International University of Health and Welfare Atami Hospital, Yokohama Sakae Kyosai Hospital, and Namiki Koiso Clinic) for unexplained abdominal symptoms between February 6, 2021, and May 31, 2022. The study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the local ethics committee of Yokohama City University Hospital (approval number: B200800055, October 15, 2020). It was also registered with the University Hospital Medical Information Network (UMIN000043262) on February 5, 2021. All participants provided written informed consent before enrollment. Because this is a cross-sectional study aimed at exploratory investigation of the relationship between BAs in SUDD and healthy subjects, the planned enrollment was set at the number of subjects that could be enrolled within the enrollment period.

Patients with chronic abdominal symptoms were considered to have unexplained abdominal symptoms upon visiting the gastroenterology clinic at least three times during a 6-month period and did not obtain a clear diagnosis despite having undergone various tests. The patients were examined in outpatient clinics specializing in the diagnosis of chronic abdominal symptoms to rule out organic diseases such as inflammatory bowel disease or colorectal cancer. All patients underwent colonoscopy, and the presence and site of diverticulosis were assessed. Colonoscopy findings were then double-checked by authors who are expert colonoscopists. Questionnaires were used to investigate age, sex, body mass index (BMI), waist circumference, abdominal surgery history, and pain level. Abdominal pain severity was assessed using an 11-point numerical rating scale (NRS), with 0 representing no pain and 10 representing the most severe pain (17). In order to assess the shape of stool, the Bristol Stool Form Scale (BSFS) was used to grade stool consistency. Types 1 and 2 suggest constipation, types 3 and 4 indicate normal defecation, and types 5, 6, and 7 indicate diarrhea (18). Blood and fecal samples were collected, with medications pertaining to BA metabolism withdrawn for at least 1 month before sample collection.

With reference to previous reports (1), SUDD was diagnosed as follows: (1) persistence of symptoms for more than 24 h; (2) pain in the abdomen consistent with the site of the diverticulum (3) absence of signs and symptoms and clinical and microscopic evidence of acute diverticulitis; and (4) no criteria for IBS (1, 16). In Europe, left-sided diverticula are predominant, whereas right-sided diverticula are more common in Asia (16, 19). The definition of SUDD in the previous report (1) is limited to left-sided lower abdominal pain, but since the existence of right-sided SUDD has been reported in Asia (2), we modified the definition of SUDD in this report to include abdominal pain consistent with the site of the diverticulum, so that right-sided SUDD can also be diagnosed. CT or lower gastrointestinal endoscopy confirmed the absence of obvious intestinal inflammatory findings. IBS was defined according to the Rome IV criteria as recurrent abdominal pain, averaging at least 1 day per week over the last 3 months, meeting two or more of the following criteria: (1) abdominal pain associated with defecation, (2) abdominal pain associated with changes in defecation frequency, and (3) abdominal pain associated with changes in fecal shape (20). Conversely, patients with SUDD are characterized by abdominal pain that is not associated with defecation and persists for long durations, such that they do not satisfy the Rome IV criteria (1). Probiotics, antidiarrheals, antimicrobials, and ursodeoxycholic acid and cholestimide, which may affect BA kinetics, were discontinued at least 4 weeks before specimen collection. Laxatives continued to be taken internally, but patients taking elobixibat were excluded from the study because elobixibat affects BA kinetics.

Patients who had diverticula but did not fulfill the criteria for SUDD were referred to as non-SUDD patients. Patients with post-diverticulitis SUDD (PD-SUDD) are those who have experienced at least one episode of acute diverticulitis with or without complications and have abdominal pain despite resolution of macroscopic findings of diverticulitis (1); in this study, PD-SUDD patients were included in SUDD patients. SUDD patients with no history of diverticulitis were described as non-PD-SUDD. The healthy control (HC) group included individuals with no history of medications or evidence of colonic diverticula. In total, 34 patients met the eligibility criteria for the present study and had available BA measurements (UMIN000020917, approval number: B160101015, February 5, 2016).

Information pertaining to this is described in the Supplementary Methods.

Information pertaining to this is described in the Supplementary Methods.

Bile acid was measured in accordance with a previous report (21). The following is a description of the method used to measure fecal BAs. To 20 mg of fecal sample, 250 μL of 1 M potassium hydroxide solution was added, vortexed, and incubated at 80°C for 20 min. To the 5-ml tube, 2 ml of 0.5 M Potassium Phosphate Buffer (PPB) was added, and 2 μL each of 10 internal standard solutions were added. 100 μL of stool sample volume was added to the 5-mL tube containing PPB, vortexed, and placed in a Bond Elut C18 Cartridge to extract BA. The sample was evaporated to dryness in EYELA blown concentrator for 40 min, 200 μl of 0.05% acetic acid solution and acetonitrile were added, and centrifuged. 40 μL of the supernatant was quantified by liquid chromatography-mass spectrometer (Shimadzu Corporation, Kyoto, Japan). The following is a description of the method used to measure serum BAs. Ten μl of sample serum was vortexed with 2 μL each of 10 different internal standard solutions, 30 μL of 1 M hydrochloric acid, and 1000 μL of acetonitrile, and then centrifuged. The 1000 μL supernatant was evaporated to dryness for 1 h, and 200 μL of methanol was added and centrifuged. Fifty microliters of the supernatant was quantified by liquid chromatography-mass spectrometer (Shimadzu Corporation, Kyoto, Japan).

Data were analyzed using JMP software (version 15.0; SAS Institute Inc., Cary, NC, USA) and are expressed as mean ± standard deviation unless otherwise indicated. Student’s t-tests were used to compare continuous variables or univariate means between two groups; for the F-test, P-values were calculated using one-way analysis of variance across the three groups. According to Tukey’s honest significant difference post-hoc test to correct for multiple testing, two-sample t-tests were conducted for each pairwise comparison among the three groups if the P-value for the F-test was significant at a two-sided significance level of 5%. The χ² test was used to compare binary variables. All t-tests, F-tests, and χ² tests were two-sided with a significance level of 5% (P < 0.05). The graphs were prepared using GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, USA); data are presented as means ± standard errors.

In total, 361 participants with abdominal complaints were screened. In 253 individuals, including 54 with chronic intestinal pseudo-obstruction, 128 with functional constipation, 12 with functional diarrhea, and 25 with IBS, diverticulosis was ruled out. Of the 361 patients, 108 (30%) had diverticulosis, and 33 (9%) presented with SUDD. Of the 108 patients with diverticulosis, 33 (31%) had SUDD, while 75 (69%) had non-SUDD. The non-SUDD group included patients with diverticula but not meeting the diagnostic criteria for SUDD. Of the 75 patients with non-SUDD, 39 had asymptomatic diverticulosis. There were 26 patients who did not meet the criteria for SUDD but had abdominal pain symptoms, 5 had functional constipation, 2 had IBS, and 19 were other. The HC included 34 samples for which BAs were measurable from previous study samples (Figure 1).

Patients with SUDD had a mean age of 62 years, BMI of 23.6 kg/m², and waist circumference of 83.4 cm. There were no differences in age, sex, BMI, or waist circumference among HC, non-SUDD, and SUDD groups. No patient had a history of ileal terminal surgery. No one in the SUDD group or the healthy group had ever had cholelithiasis or cholecystectomy. Among the patients in the non-SUDD group, two patients had cholelithiasis and two patients had a history of cholecystectomy surgery. Of the 33 patients with SUDD, 39% (13 patients) were right-sided, 45% (15 patients) were left-sided, and 15% (5 patients) were bilateral. The location of the diverticulum did not differ significantly between non-SUDD and SUDD groups. NRS scores were significantly higher in the SUDD group than in the non-SUDD group (non-SUDD 3.6 vs. SUDD 5.8, P < 0.0001). BSFS scores were significantly higher in the SUDD group than in the non-SUDD group (non-SUDD 3.8 vs. SUDD 5.5, P < 0.0001). The number of patients with a BSFS score of 6 or 7 was also significantly higher in the SUDD group than in the non-SUDD group (non-SUDD 12% vs. SUDD 63%, P < 0.0001). The total DICA classification score was significantly higher in the SUDD group than in the non-SUDD group (non-SUDD 1.6 vs. SUDD 2.4, P < 0.0001) (Table 1).

The total BA concentration did not differ between the HC and non-SUDD groups; however, it was significantly higher in the SUDD group than in the other two groups (Figure 2a). Most fecal BAs were secondary BAs such as deoxycholic acid (DCA) and lithocholic acid (LCA) (Figure 2b). Total cholic acid (CA), chenodeoxycholic acid (CDCA), DCA, LCA, hyodeoxycholic acid (HDCA), and ursodeoxycholic acid (UDCA) levels were not different between the HC and non-SUDD groups; however, they were significantly higher in the SUDD group than in the HC and non-SUDD groups (Figures 2c, d). The same results were obtained for total unconjugated, primary, and secondary BA levels (Figures 3a, b) and the primary/secondary BA ratio (Figure 3c and Supplementary Table 1). A sensitivity analysis was performed after excluding two patients with cholelithiasis and two patients who had undergone cholecystectomy, and the results were similar to those of the main analysis. In a comparison of asymptomatic diverticula and non-SUDD with abdominal pain and SUDD, the concentrations of fecal total BA, unconjugated BA, primary BA, secondary BA, Total CA, CDCA, DCA, LCA, HDCA, and UDCA were significantly higher in the SUDD group than in the other two groups. There was no significant difference between the asymptomatic diverticula group and the non-SUDD group with abdominal pain (Supplementary Table 2). In a comparison of non-diverticula with abdominal pain, non-SUDD with abdominal pain, and SUDD, there was no significant difference in total BA concentration between the non-diverticula with abdominal pain group and the non-SUDD with abdominal pain group, but it was significantly higher in the SUDD group than in the other two groups. Total CA, total CDCA, total DCA, total LCA, Total UDCA concentrations were significantly higher in the SUDD group than in the non-diverticula with abdominal pain group (Supplementary Table 3).

The levels of unconjugated CA, CDCA, DCA, LCA, HDCA, and UDCA were higher in the SUDD group than in the HC and non-SUDD groups, with no difference between the HC and non-SUDD groups. Conjugated levels of CA and UDCA did not differ between the HC and non-SUDD groups; however, they were significantly increased in the SUDD group. Conversely, the levels of conjugated DCA, LCA, and HDCA did not differ across the three groups (Supplementary Table 4).

Among the 33 SUDD patients, 23 had non-PD-SUDD and 10 had PD-SUDD. Fecal BA concentrations were analyzed between the non-PD-SUDD and PD-SUDD groups. Total fecal BA, CA, CDCA, DCA, LCA, HDCA, and UDCA levels were not different between the non-PD-SUDD and PD-SUDD groups. Total conjugated, unconjugated, primary, secondary BA levels and the primary/secondary BA ratio did not differ between the non-PD-SUDD and PD-SUDD groups (Supplementary Table 5).

Total serum BA levels did not differ across the three groups (Supplementary Figure 1). Most serum BAs were UDCA (Supplementary Figure 2a). CA, CDCA, DCA, LCA, HDCA, and UDCA levels were not different between the non-SUDD and SUDD groups (Supplementary Table 6). Total conjugated, unconjugated, primary, and secondary BA levels did not differ between the non-SUDD and SUDD groups (Supplementary Figures 2b, c). Conjugated and unconjugated levels of CA, CDCA, DCA, LCA, HDCA, and UDCA did not differ between the non-SUDD and SUDD groups (Supplementary Table 7). The serum primary/secondary BA ratio did not differ between the HC and non-SUDD groups; however, it was significantly lower in the SUDD group than in the HC group (Supplementary Figure 2d).

No differences in fecal BA concentrations were found among patients with right-sided SUDD, left-sided SUDD, and bilateral SUDD (Supplementary Table 8). Serum BA concentrations also did not differ according to site (Supplementary Table 9).

Serum high-sensitivity C-reactive protein (hsCRP) levels were significantly higher in the SUDD group than in the HC and non-SUDD groups; however, they did not differ between the non-SUDD and HC groups (Figure 4a and Supplementary Table 10). Fecal calprotectin levels were significantly higher in the SUDD and non-SUDD groups than in the HC group, and in the SUDD group than in the non-SUDD group (Figure 4b). Blood Endotoxin Activity Assay (EAA) levels were significantly higher in the SUDD group than in the other groups, and higher in the non-SUDD group than in the HC group (Figure 4c). Serum C4 and FGF-19 levels did not differ across the three groups (Figure 5).

The fecal total BA concentration was mildly positively correlated with fecal calprotectin and blood EAA (vs fecal calprotectin; r = 0.28, vs blood EAA; r = 0.23). Inflammatory markers were more positively correlated with secondary BAs than with primary BAs and more positively correlated with unconjugated BAs than with conjugated BAs (Supplementary Figures 3a, b, c, d and Supplementary Table 11). Among the unconjugated secondary BAs, UDCA most strongly correlated with inflammatory markers (Supplementary Table 12).

Fecal total BA concentration showed a mild positive correlation with BSFS score and a very weak positive correlation with NRS score and DICA total score. Primary BA concentrations showed a mild positive correlation with NRS score, and primary, secondary, and unconjugated BA concentrations showed a mild positive correlation with BSFS score. Fecal CA, HDCA and UDCA were mildly positively correlated with NRS score; and fecal CDCA, DCA, LCA, HDCA and UDCA were mildly positively correlated with BSFS score; and Fecal CA, LCA and UDCA were mildly positively correlated with DICA total score (Supplementary Table 13).