Natasha A. Koloski

EClinicalMedicine • September 08, 2024

Ayesha Shah, Liam Spannenburg, Parag Thite, Mark Morrison, Thomas Fairlie, Natasha Koloski, Purna Kashyap, Mark Pimentel, Ali Rezaie, Gregory Gores, Michael Jones, Gerald Holtmann

Background Small Intestinal Bacterial Overgrowth (SIBO) has been implicated in the pathophysiology of chronic liver disease (CLD). We conducted a systematic review and meta-analysis to assess and compare the prevalence of SIBO among CLD patients (with and without with complications of end stage liver disease) and healthy controls. Methods Electronic databases were searched from inception up to July-2024 for case–control studies reporting SIBO in CLD. Prevalence rates, odds ratios (ORs), and 95% confidence intervals (CIs) of SIBO in patients with CLD and controls were calculated utilizing a random-effects model. The protocol was prospectively registered with PROSPERO (CRD42022379578). Findings The final dataset included 34 case–control studies with 2130 CLD patients and 1222 controls. Overall, the odds for SIBO prevalence in CLD patients compared to controls was 6.7 (95% CI 4.6–9.7, p < 0.001). Although the prevalence of SIBO among patients with CLD with cirrhosis was higher at 42.9% (95% CI: 35.9–50.2) compared to 36.9% (95% CI: 27.4–47.6) in those without cirrhosis, this difference failed statistical significance. However, CLD patients with decompensated cirrhosis had a significantly higher prevalence of SIBO compared to those with compensated cirrhosis, with an OR of 2.6 (95% CI: 1.5–4.5, p < 0.001). Additionally, the prevalence of SIBO was significantly higher in CLD patients with portal hypertension (PHT) than in those without PHT, with an OR of 2.1 (95% CI: 1.4–3.1, p < 0.001). The highest prevalence of SIBO was observed in patients with spontaneous bacterial peritonitis (SBP) (57.7%, 95% CI 38.8–74.5), followed by patients with hepatic encephalopathy (41.0%, 95% CI 16.0–72.3) and patients with variceal bleed (39.5%, 95% CI 12.1–75.6). Interpretation Overall, there is a significantly increased prevalence of SIBO in CLD patients compared to controls. The prevalence is even higher in CLD patients with PHT, especially those with SBP. This meta-analysis suggests that SIBO is associated with complications of CLD and potentially linked to the progression of CLD.

Neurogastroenterology And Motility • July 25, 2024

Michael Jones, Gerald Holtmann, Jan Tack, Florencia Carbonne, William Chey, Natasha Koloski, Ayesha Shah, Shrikant Bangdiwala, Ami Sperber, Olafur Palsson, Nicholas Talley

Objective: The group of disorders known as Disorders of Gut Brain Interaction (DGBI) were originally labeled functional GI disorders and were thought to be disorders of the gastrointestinal tract that had several psychological conditions as comorbidities. Despite mounting evidence that psychological morbidity plays an innate role in the etiology and maintenance of DGBI, none of the Rome IV criteria include any measure of psychological symptoms. This study tested the hypothesis that individuals would cluster differently if GI symptoms alone were considered versus GI symptoms combined with measures of psychological symptoms. Methods: Data were obtained from the Rome Foundation Global Epidemiology Study measuring Rome IV GI symptoms, psychological measures and demographic characteristics. Latent profile models were used to cluster individuals based on (i) GI symptoms only (GI only) and then (ii) GI and psychological measures (GI + Psych). Results: Individuals clustering into the same group of individuals whether formed via GI only or GI + Psych, ranged from 96% for a 2-class solution (the most simplistic) to 76% with 6 classes (the parsimonious system) and 59% with twenty-two classes (mimicking Rome IV). The generalisability of this finding between six geographic regions was confirmed with agreement varying between 95%-97% for 2 clusters and 71-79% for 6 classes and 51%-63% for 22 classes. These findings were also consistent between DGBI (range 94% with 2 classes to 50% with 22 classes) and non-DGBI (range 97% with 2 clusters to 65% with 22 classes) groups. Conclusions: Our data suggest that considering psychological as well as gastrointestinal symptoms would lead to a different clustering of individuals in more complex, and accurate, classification systems. For this reason, future work on DGBI classification should consider inclusion of psychological traits.

Frontiers In Microbiology

Peter Sternes, Ayesha Shah, Camila Ayala Pintos, Thomas Fairlie, Natasha Koloski, Seungha Kang, Kaylyn Tousignant, Simon Mcilroy, Mark Morrison, Gene Tyson, Gerald Holtmann

Little is known about the biogeography of the mucosa associated microbiome (MAM) in patients with inflammatory bowel disease (IBD) versus controls in different segments of the gastrointestinal tract, as well as the links between the MAM, gastrointestinal symptoms, and use of proton pump inhibitors (PPI). We recruited 59 controls (without structural abnormalities and gastrointestinal symptoms), 44 patients with ulcerative colitis (UC) and 31 with Crohn's disease (CD). Biopsies from various segments of the upper and lower gastrointestinal tract were collected. Microbial composition was assessed via 16S rRNA gene amplicon analysis and the bacterial load of the mucosal biopsies were assessed via qPCR. The MAM was examined in the context of disease status, PPI usage, the severity of gastrointestinal symptoms, and the symptom response to a standardised nutrient challenge (SNC). Microbial communities of the MAM in the upper and lower gastrointestinal tract differed. IBD patients were characterised by relative and absolute depletion of numerous genera known to produce butyrate and/or propionate, with the largest differentiation being the depletion of Faecalibacterium in the lower gastrointestinal tract of CD patients. Notably, PPI users exhibited an enrichment of Faecalibacterium in the lower gastrointestinal tract. The severity of gastrointestinal symptoms, as well as the symptom response to the SNC, were significantly associated with MAM composition in the gastrointestinal tract. The absolute and relative composition of the MAM is variable across different segments of the gastrointestinal tract. These quantitative changes indicates that MAM can be targeted in specific segments of the GI tract to improve patient outcomes.

Journal Of Gastroenterology And Hepatology • June 17, 2024

Sahar Pakneshan, Naomi Moy, Ayesha Shah, Natasha Koloski, Mike Jones, Nicholas Talley, Gerald Holtmann

Objective: In immunochemical fecal occult blood test (iFOBT) positive subjects, colonoscopy screening can detect colorectal cancers and advanced adenomas, yet most iFOBT-positive subjects find no relevant lower gastrointestinal lesions. Limited data are available on upper gastrointestinal (UGI) cancer risk in iFOBT-positive patients. This study investigated the incidence of UGI malignancies diagnosed within 3 years post-colonoscopy after a positive iFOBT. Methods: Retrospective analysis of iFOBT-positive patients aged 50-75 years who underwent a colonoscopy at a single institution. All patients with a diagnosis of UGI cancer within 3 years post-colonoscopy were identified by linking with the Queensland Cancer Register. This was used to compare to the geographical population aged 50-74 years based on the Australian Bureau of Statistics and Queensland Cancer Council data. Results: From 1748 eligible participants, 0.23% (95% confidence interval [CI] 0.06-0.58) were diagnosed with UGI cancer within 3 years post-colonoscopy. This indicates an esophageal cancers rate of 114.42 per 100 000 (95% CI 100.56-298.28) and gastric cancer rate of 57.21 per 100 000 (95% CI 55.76-261.12). Of the patients with a UGI cancer, 75% would have had an unexplained iFOBT. Annual incidence for the same geographic region, ages, and period for the combined esophageal and gastric cancer was 36.08 per 100 000 (95% CI 32.87-39.52). Conclusions: Among individuals with a positive iFOBT in a bowel cancer screening program, the rates of gastric and esophageal cancers were 2.7 and 7.5 times higher than the general population. Adding gastroscopy to a colonoscopy for iFOBT-positive patients in cancer surveillance programs may be justifiable.

Journal Of Gastroenterology And Hepatology • March 06, 2024

Ayesha Shah, Thomas Fairlie, Mark Morrison, Neal Martin, Karin Hammer, Johann Hammer, Natasha Koloski, Ali Rezaie, Mark Pimentel, Purna Kashyap, Michael Jones, Gerald Holtmann

Objective: Patients with intestinal failure (IF) have abnormal intestinal anatomy, secretion, and dysmotility, which impairs intestinal homeostatic mechanisms and may lead to small intestinal bacterial overgrowth (SIBO). We conducted a systematic review and meta-analysis to determine the prevalence of SIBO in patients with IF and to identify risk factors for SIBO. Methods: MEDLINE (PubMed) and Embase electronic databases were searched from inception to December 2023 for studies that reported the prevalence of SIBO in IF. The prevalence rates, odds ratio (OR), and 95% confidence intervals of SIBO in IF and the risk factors for SIBO in IF were calculated using random effects model. Results: Final dataset included nine studies reporting on 407 patients with IF. The prevalence of SIBO in IF was 57.5% (95% CI 44.6-69.4), with substantial heterogeneity in this analysis (I2 = 80.9, P = 0.0001). SIBO prevalence was sixfold higher in patients with IF who received parenteral nutrition (PN) compared with IF patients not on PN (OR = 6.0, 95% CI 3.0-11.9, P = 0.0001). Overall, the prevalence of SIBO in patients with IF using PPI/acid-suppressing agents (72.0%, 95% CI 57.5-83.8) was numerically higher compared with IF patients not using these agents (47.6%, 95% CI 25.7-70.2). Conclusions: This systematic review and meta-analysis suggests that there is an increased risk of SIBO in patients with IF and that PN, and potentially, the use of PPI/acid-suppressing agents is risk factors for SIBO development in patients with IF. However, the quality of evidence is low and can be attributed to lack of case-control studies and clinical heterogeneity seen in the studies.