BMJ Open Gastroenterology最新文献

Objectives: Uncertainty remains about many aspects of first-line treatment of autoimmune hepatitis (AIH).

Design: Systemic review with meta-analysis (MA).

Data sources: Bespoke AIH Endnote Library, updated to 30 June 2024.

Eligibility criteria: Randomised controlled trials (RCTs) and comparative cohort studies including adult patients with AIH, reporting death/transplantation, biochemical response (BR) and/or adverse effects (AEs).

Data extraction and synthesis: Data pooled in MA as relative risk (RR) under random effects. Risk of bias (ROB) assessed using Cochrane ROB-2 and ROBINS-1 tools.

Results: From seven RCTs (five with low and two with some ROB) and 18 cohort studies (12 moderate ROB, six high for death/transplant), we found lower death/transplantation rates in (a) patients receiving pred+/-aza (vs no pred): overall (RR 0.38 (95% CI 0.20 to 0.74)), in patients without symptoms (0.38 (0.19-0.75)), without cirrhosis (0.30 (0.14-0.65)), and with decompensated cirrhosis (RR 0.38 (0.23-0.61)), and (b) patients receiving pred+aza (vs pred alone) (0.38 (0.22-0.65)). Patients receiving higher (vs lower) initial pred doses had similar BR rates (RR 1.07 (0.92-1.24)) and mortality (0.71 (0.25-2.05)) but more AEs (1.73 (1.17-2.55)). Patients receiving bud (vs pred) had similar BR rates (RR 0.99 (0.71-1.39)), with fewer cosmetic AEs (0.46 (0.34-0.62)). Patients receiving mycophenolate mofetil (MMF) (vs aza) had similar BR rates (RR 1.32 (0.73-2.38)) and fewer AEs requiring drug cessation (0.20 (0.09-0.43)).

Conclusions: Mortality is lower in pred-treated (vs untreated) patients, overall and in several subgroups, and in those receiving pred+aza (vs pred). Higher initial pred doses confer no clear benefit and cause more AEs. Bud (vs pred) achieves similar BR rates, with fewer cosmetic AEs. MMF (vs aza) achieves similar BR rates, with fewer serious AEs.

Objective: Colorectal cancer (CRC) is the third most common cancer in Denmark, with a 5-year mortality of 40%. To reduce CRC incidence and mortality, a faecal immunochemical test (FIT)-based screening programme was introduced in 2014. Adenoma detection rate (ADR) is an established quality marker inversely associated with post-colonoscopy CRC (PCCRC), but evidence mainly stems from non-FIT populations. Using ADR in a FIT-based setting may be costly due to histopathological examination. Alternative markers like polyp detection rate (PDR) and sessile serrated lesion detection rate (SDR) could be viable but lack evidence for their association with PCCRC.

Methods: We conducted a nationwide cohort study of 77 009 FIT-positive participants undergoing colonoscopy (2014-2017). National registry data on CRC outcomes were linked, and endoscopy units were grouped by ADR, PDR and SDR levels. Poisson regression adjusted for age, sex and comorbidities was used to assess PCCRC risk.

Results: Among 70 009 colonoscopies within 6 months of FIT positivity, 4401 (92.7%) had CRC, while 342 (7.2%) were PCCRC cases. PCCRC risk was inversely associated with ADR, PDR and SDR. High ADR endoscopy units had a 35% lower PCCRC risk than low ADR units. Similar associations were found for PDR and SDR, with high SDR units showing a 33% lower PCCRC risk than low SDR units.

Conclusions: ADR, PDR and SDR predict PCCRC risk, with SDR emerging as a feasible, cost-efficient quality marker in FIT-based screening. This study supports SDR as a primary performance indicator in future protocols.

Objective: Non-operative management (NOM) of acute cholecystitis (ACC) may be preferable in patients with advanced inflammation, long duration of symptoms or severe comorbidities. This study aims to investigate time to recurrence and patient factors predicting relapse in gallstone complications after NOM.

Methods: Records of 1634 patients treated for ACC at three Swedish centres between 2017 and 2020 were analysed, with 909 managed non-operatively. Data were linked to the National Gallstone Surgery registry for those who later underwent surgery. The time to relapse of gallstone complications was calculated and Cox proportional hazards regression was used to analyse new gallstone complications and adjust for multiple variables.

Results: Of the 909 non-operatively managed patients, 348 patients suffered a new gallstone complication. The median time to recurrence was 82 days. Of those who recurred, 27% did so within 30 days, 17% between 31 and 60 days, 27% between 61 days and 6 months, 16% between 6 months and 1 year and 13% later than 1 year. Younger patients with their first gallstone complication had a lower risk of new complications compared with those with previous gallstone complications. In older individuals, there was no difference in the risk of relapse regardless of previous gallstone complications, but they were more likely to be readmitted than younger patients.

Conclusion: Delayed cholecystectomy should be prioritised for younger patients with a history of gallstone disease if early cholecystectomy is not feasible. Delayed cholecystectomy should be scheduled without a prior outpatient clinic visit to minimise delays.

Objective: Despite guidelines indicating no contraindications for contraceptives in women with inflammatory bowel disease (IBD), this population shows increased voluntary childlessness and lower contraceptive use. Knowledge gaps among healthcare providers on IBD's impact on fertility and contraception may drive these trends. This survey assessed knowledge discrepancies among IBD patients, gastroenterologists (GEs), and women's healthcare providers (WHPs) regarding fertility and contraception.

Methods: An anonymous survey was conducted between August and December 2023, targeting IBD patients of childbearing age, GEs and WHPs. The questionnaire was offered consecutively to all patients consulting or hospitalised in our department. Additionally, the survey link was shared with healthcare professionals during dedicated training sessions. It assessed awareness of IBD-related fertility and contraception impacts.

Results: Two hundred twenty-two participants fulfilled the survey (100 patients, 50 GEs and 72 WHPs). Among patients (63% with Crohn's disease), 95% were on biologic or immunosuppressant therapy. Nearly half (47%) of women had not discussed fertility or contraception with their GE, and only 22% had done so on request. A majority (80% of women, 54% of GEs) were unsure if IBD affects contraception efficacy, and 50% of WHPs believed oral contraceptives to be less effective for IBD patients. Key concerns influencing patients' fertility decisions included the impact of IBD medication on pregnancy (51%), risk of passing IBD to offspring (47%) and potential flare-ups during pregnancy (39%).

Conclusion: Significant knowledge gaps on fertility and contraception in IBD persist among patients, GEs and WHPs.

Objective: Granulomatous liver disease (GLD) is a rare condition with various aetiologies and is characterised by the formation of hepatic granulomas. A comprehensive evaluation of GLD from a broad perspective is lacking. We aimed to investigate the aetiology and the clinicoradiopathological characteristics of patients with GLD in recent decades in Thailand.

Methods: This retrospective study was conducted at a tertiary care centre in Thailand. All patients who underwent liver biopsy between 2003 and 2023 were reviewed. Patients with a histopathological report of granulomas in liver specimens were included. Clinical presentations, radiological data, and laboratory data closest to the procedure date were also collected.

Results: Of the 4384 liver biopsy specimens collected during the study period, 89 (2%) had GLD. Of these, 58.4% were men, with the following aetiologies: 61 (68.5%) infectious, 16 (18%) non-infectious, and 12 (13.5%) undetermined. Common presentations included abnormal liver test results (81.4%) and fever (56.1%). Among infectious granulomas, mycobacterial infections (tuberculosis: 28; non-tuberculous mycobacteria (NTM): 11) were predominant. Compared with other causes, NTM was associated with a significantly lower body mass index, more extragastrointestinal involvement, and lower serum albumin levels. Caseating-type granulomas were also observed in 16% of non-mycobacterial cases. Nearly 40% of patients with GLD demonstrated no focal lesions on liver imaging, whereas multifocal lesions were found in a third of patients.

Conclusions: Infectious causes, especially mycobacterial infections, remain the primary aetiology of GLD in Thailand. Granuloma types are not pathognomonic of specific diseases, emphasising the need for extensive evaluation beyond liver biopsy to determine the underlying aetiology.

Introduction: Large learning models (LLMs) such as GPT are advanced artificial intelligence (AI) models. Originally developed for natural language processing, they have been adapted for multi-modal tasks with vision-language input. One clinically relevant task is scoring the Boston Bowel Preparation Scale (BBPS). While traditional AI techniques use large amounts of data for training, we hypothesise that vision-language LLM can perform this task with fewer examples.

Methods: We used the GPT4V vision-language LLM developed by OpenAI, via the OpenAI application programming interface. A standardised prompt instructed the model to grade BBPS with contextual references extracted from the original paper describing the BBPS by Lai et al (GIE 2009). Performance was tested on the HyperKvasir dataset, an open dataset for automated BBPS grading.

Results: Of 1794 images, GPT4V returned valid results for 1772 (98%). It had an accuracy of 0.84 for two-class classification (BBPS 0-1 vs 2-3) and 0.74 for four-class classification (BBPS 0, 1, 2, 3). Macro-averaged F1 scores were 0.81 and 0.63, respectively. Qualitatively, most errors arose from misclassification of BBPS 1 as 2. These results compare favourably with current methods using large amounts of training data, which achieve an accuracy in the range of 0.8-0.9.

Conclusion: This study provides proof-of-concept that a vision-language LLM is able to perform BBPS classification accurately, without large training datasets. This represents a paradigm shift in AI classification methods in medicine, where many diseases lack sufficient data to train traditional AI models. An LLM with appropriate examples may be used in such cases.

Introduction: Primary sclerosing cholangitis (PSC) is a rare immune-mediated hepatobiliary disease, characterised by progressive biliary fibrosis, cirrhosis, and end-stage liver disease. As yet, no licensed pharmacological therapy exists. While significant advancements have been made in our understanding of the pathophysiology, the exact aetiology remains poorly defined. Compelling evidence from basic science and translational studies implicates the role of T helper 17 cells (Th17) and the interleukin 17 (IL-17) pro-inflammatory signalling pathway in the pathogenesis of PSC. However, exploration of the safety and efficacy of inhibiting the IL-17 pathway in PSC is lacking.

Methods and analysis: This is a phase 2a, open-label, multicentre pilot study, testing the safety of brodalumab, a recombinant human monoclonal antibody that binds with high affinity to interleukin-17RA, in adults with PSC. This study will enrol 20 PSC patients across five large National Health Service tertiary centres in the UK. The primary outcome of the study relates to determining the safety and feasibility of administering brodalumab in early, non-cirrhotic PSC patients. Secondary efficacy outcomes include non-invasive assessment of liver fibrosis, changes in alkaline phosphatase values and other liver biochemical readouts, assessment of biliary metrics through quantitative MR cholangiography+, and quality of life evaluation on completion of follow-up (using the 5D-itch tool, the PSC-patient-reported outcome and PSC-specific Chronic Liver Disease Questionnaire).

Ethics and dissemination: Ethical approval for this study has been obtained from the London Bridge Research Ethics Committee (REC23/LO/0718). Written informed consent will be obtained from all trial participants prior to undertaking any trial-specific examinations or investigations. On completion of the study, results will be submitted for publication in peer-reviewed journals and presented at national and international hepatology meetings. A summary of the findings will also be shared with participants and PSC communities.

Trial registration number: ISRCTN15271834.

Objective: We provide an overview of the latest estimates of five gastrointestinal (GI) cancers in the Eastern Mediterranean Region (EMR) countries to guide cancer control policy.

Methods: We extracted the number of cases and deaths for oesophageal, gastric, liver, colorectal and pancreatic cancers from the GLOBOCAN database produced as estimated by the International Agency for Research on Cancer for the year 2022. Age-standardised incidence and mortality rates (ASR) per 100 000 person-years were estimated for the 22 EMR countries, cancer site and sex.

Results: The estimated 173 000 new cancer cases and 139 000 deaths from the five GI cancers corresponded to 22.2% of the incidence and 28.7% of the mortality burden in the EMR. Across all cancers (for both sexes combined), colorectal cancer ranked third (6.9%; ASR 8.9), followed by liver cancer (6.2%; ASR 8.4) in terms of incidence, while liver cancer (9.6%; ASR 8.1) and gastric cancer (6.4%; ASR 5.5) were the third and fourth leading causes of cancer-related mortality in the region, respectively. Marked differences in cancer incidence and mortality rates were observed between the 22 countries, particularly the 10-fold variations seen in liver cancer incidence.

Conclusion: GI cancers currently account for an important fraction of the cancer burden in the EMR; the present analysis seeks to inform tailored decision-making according to the country-specific GI cancer profiles.

Objectives: Atypical ulcerative colitis (UC) presenting reverse gradient colitis, backwash ileitis, or rectal sparing and/or positive atypical antineutrophil cytoplasmic antibody serology is often associated with primary sclerosing cholangitis (PSC) and can be resistant to conventional medical therapies (CMT) for inflammatory bowel diseases. We report short-term and long-term outcomes of oral vancomycin therapy (OVT) in children with atypical UC and confirmed PSC in imaging/biopsy (PSC-UC) or treatment-resistant atypical UC without detectable PSC (aUC-non-PSC).

Methods: In this retrospective real-world observational study from a tertiary paediatric centre in Brisbane, Australia, 44 children with aUC (29 PSC-UC, 15 aUC-non-PSC) received 79 OVT courses between 2014 and 2023. Pre-post-OVT characteristics were compared and relapses/repeated courses were recorded.

Results: Pre-OVT, all had active colitis by Paediatric Ulcerative Colitis Activity Index (PUCAI), Feacal Calprotectin (FC) and/or colonoscopy. Post-OVT, PUCAI reduced from 15 (IQR 5-33) to 0 (IQR 0-5); 85% of children with pre-OVT PUCAI ≥10 achieved clinical remission (100% PSC-UC vs 64% aUC-non-PSC, p=0.019). FC reduced from 995 (IQR 319-1825) to 44 (IQR 16-79) µg/g; 83% of children with pre-OVT FC ≥100 µg/g achieved biochemical remission (92% PSC-UC vs 64% aUC-non-PSC, p=0.063). Colonoscopy confirmed Mayo 0 healing in 62% (67% PSC-UC vs 54% aUC-non-PSC, p=0.443) and 46% achieved pan-colonic histological remission (54% PSC-UC vs 31% aUC-non-PSC, p=0.173). All pre-post-OVT changes in these four markers were significant in both groups. After ceasing first OVT, 25/44 relapsed within 8.2 (IQR 1.9-14.5) months. Recommencing OVT regained biomarker remission in 13/25. During 3.8 (IQR 2.0-5.3) years of follow-up, 79 OVT courses in conjunction with CMT maintained deep remission in 67%. Routine stool testing (n=138) detected no vancomycin-resistant Enterococcus (VRE).

Conclusions: OVT induced and reinduced remission in children with atypical UC. Relapse often followed ceasing vancomycin, half responded to reinduction. No VRE was developed.

Introduction: Chronic hepatitis C virus (HCV) infection is prevalent in prisons. Universal reception HCV testing is recommended, but acceptance can be suboptimal. To detect and treat missed HCV infections, a high-intensity test and treat (HITT) programme was introduced to rapidly test entire prisons. It remains unknown whether regular HITTs will be required to maintain prison microelimination. We aimed to assess the outcomes of HITTs conducted in a prison 4 years apart with ongoing reception testing.

Methods: A prospective observational evaluation of the impact of HITTs was conducted in January 2020 and February 2024 at Low Newton, a female prison. The outcomes of the reception testing were reviewed in the intervening period to determine the number of newly identified HCV infections.

Results: HITTs were successful in testing almost all residents (305/307) in 2020 and (296/296) in 2024. The number of newly diagnosed HCV individuals fell from 6.6% in 2020 to 0.3% in 2024. One new HCV case was identified in the second HITT. In between the HITTs, 89% of receptions had HCV testing, increasing from 83% in 2020 to 95% in 2023. Overall, 81% (144/178) of active HCV infections received antiviral treatment, and 89% achieved sustained virological response. The proportion of active HCV infections between the HITTs was 7.2% through reception testing.

Conclusion: A follow-up HITT after 4 years yielded only 0.3% active HCV infection in a high HCV prevalence prison and a reasonably good reception testing and treatment programme. Therefore, resources should be focused on optimising reception testing rather than undertaking repeated HITTs.