Introduction

Chat Generative Pre-Trained Transformer (Chat-GPT) by OpenAI, released in November 2022, is an artificial intelligence (AI) language model that was designed to generate human-like text based on inputs provided. Given the high potential of a pre-trained technology, it has recently gained attention among healthcare professionals for its capabilities of communicating information. This technology has, indeed, proven to be effective in providing information to patients regarding specific queries and doubts in common gastrointestinal diseases, ensuring high-quality answers. Gastrointestinal (GI) endoscopy is perceived by patients as uncomfortable and worrying, despite the advancements regarding instrumentation and sedation.

Aims & Methods

We aimed to assess the effectiveness of Chat-GPT in answering common patients’ queries on GI endoscopy in three domains (reliability, accuracy, and comprehensibility). Two healthcare providers selected the 18 most common patient queries regarding GI endoscopy. Questions, concerning pre-procedural, intraprocedural or post-procedural queries, were administered to Chat-GPT 4.0 with the following prelude “As a physician, how would you answer to a patient asking…”. Afterwards, ten healthcare professionals, including eight physicians and two nurses, along with two non-healthcare professionals were asked to evaluate the AI-based answers. All 12 participants had expertise in GI endoscopy and were asked to rank all 18 answers according to a Likert scale for each of the following features: reliability (1 to 6 points), accuracy and comprehensibility (1 to 3 points). The one-way non-parametric ANOVA (Kruskal-Wallis test) was applied to test the distribution of respondents’ ranks.

Results

The mean overall reliability was 5.2±1.7 (Figure 1), with questions 3 and 8 receiving the highest score (5.7±0,7 and 5.7±0.7, respectively), and questions 13 and 15 ranking the lowest (4.6±0.9 and 4.8±0.4, respectively). The mean overall accuracy was 2.7± 0.4, with most of the questions ranking more than 2,5, except for questions 1, 13 and 14, which all had 2.5±0.5 scores. Finally, the mean overall comprehensibility was 2.9± 0.2: regarding this specific feature, question 18 scored had the lower mean rank (2.5±0.5). The mean ranks with standard deviations for each feature of all 18 answers are reported in Table 1.

Conclusion

The findings of our study show that Chat-GPT can deliver reliable, accurate and easy-to-understand responses to a broad range of common queries from patients undergoing GI endoscopy. The use of AI-based conversational programs, like Chat-GPT, holds promise as a valid option for physicians to support daily clinical practice, also within the setting of GI endoscopy.

Introduction

Dientamoeba fragilis is a protozoan of the human digestive tract, yet its pathogenic role remains poorly understood. Fecal microbiota transplantation (FMT) provides a useful model for investigating the potential pathogenicity and transmission of this protist from donor to recipient. Donor selection is highly stringent, and the presence of D. fragilis is subject to controversy as an exclusion criterion [1-4].

Aims & Methods

This study aimed to determine the prevalence of D. fragilis in both FMT donors and recipients and to evaluate its potential impact on the efficacy and safety of FMT as a treatment for recurrent Clostridioides difficile infection (rCDI). This single-center, longitudinal cohort study analyzed stool samples from FMT donors and recipients at three time points: pre-treatment and at 2 and 8 weeks post-FMT. All samples were negative for D. fragilis on direct microscopy and were retrospectively tested using real-time (RT-PCR). CDI cure rates, gastrointestinal adverse events (AE), and serious adverse events (SAE) were assessed prospectively.

Results

A total of 53 FMT were administered to 47 different patients, with 23 of 53 (43%) derived from D. fragilis-positive donor stool (4 out of 10 donors, 40%). One recipient tested positive before FMT (1/47, 2.1%). Among 22 patients initially negative for D. fragilis who received stool from D. fragilis-positive donors, four (18.2%) tested positive on day 15 post-FMT and remained positive at week 8. None of the recipient with D. fragilis-negative FMT was tested positive. No significant differences in efficacy, AE, or SAE were observed between recipients of FMT from D. fragilis-positive versus -negative donors, even in severely immunocompromised patients. No SAE could be attributed to FMT.

Conclusion

This study suggests that D. fragilis can be transmitted through FMT without significant clinical impact in terms of efficacy and safety. These findings argue for reconsidering the exclusion of positive donors and support the lack of pathogenicity of D. fragilis.

References

1. Hall LM, Munasinghe VS, Vella NGF, Ellis JT, Stark D. Observations on the transmission of Dientamoeba fragilis and the cyst life cycle stage. Parasitology 2024 ; 151:337–345.
2. Cacciò SM. Molecular epidemiology of Dientamoeba fragilis. Acta Trop 2018 ; 184:73–77.
3. Shasha D, Grupel D, Treigerman O, et al. The clinical significance of Dientamoeba fragilis and Blastocystis in human stool—retrospective cohort study. Clin Microbiol Infect 2024 ; 30:130–136.
4. Terveer EM, Gool T van, Ooijevaar RE, et al. Human Transmission of Blastocystis by Fecal Microbiota Transplantation Without Development of Gastrointestinal Symptoms in Recipients. Clin Infect Dis 2019 ; 71:2630–2636.

Introduction

Subcutaneous infliximab (SC-IFX) has demonstrated efficacy and safety in maintaining remission in inflammatory bowel disease (IBD) patients. However, its role in reinducing remission in patients with active disease who lost response to intravenous infliximab (IV-IFX) remains poorly explored.

Aims & Methods

Evaluate the utility of switching from IV-IFX to SC-IFX in patients with IBD and active disease.
A retrospective and multicenter study was conducted using data from the ENEIDA registry. Patients with Crohn’s disease (CD), ulcerative colitis (UC) or unclassified IBD (IBDU) who were switched from IV-IFX to SC-IFX due to clinically active IBD were included. Clinical activity was defined as Harvey Bradshaw Index (HBI) ≥5 for CD or Partial Mayo Index (PMI) >2 for UC. Clinical response was defined as a ≥3-point reduction in HBI/PMI and remission as HBI <5 or PMI ≤2. Demographics characteristics and rates of response were analyzed. Statistical analysis was performed using Chi-square or Fisher’s exact test, as appropriate. A p-value <0.05 was considered statistically significant.

Results

A total of 94 patients were included (64 CD (66.0%), 25 UC (26.6%), 5 IBDU (2.1%)). 48.9% (n=46) were female and median age was 45.8 ±14.8. Baseline characteristics are described in Table 1.
Most patients (n=62; 65.9%) required intensified IV-IFX before the switch, an 81.9% (n=77) switched to standard SC-IFX dose of 120mg every 2 weeks. Treatment was withdrawn in 13.8% (13/94) patients before week 8 because of lack of short term response. Clinical response/remission at week 8 was observed in 70.3% (66/94) and 46.8% (44/94) of patients, at week 16 in 69.1% (64/94) and 55.3% (52/94), at 6 months in 69.1% (64/94) and 54.3% (51/94), and at 12 months in 45.7% (43/94) and 37.2% (35/94) of patients. Corticosteroid-free clinical remission was reached in 95.4% (42/44) and 91.4% (32/35) of patients in clinical remission at week 8 and 12 months, respectively. Mean IV-IFX levels previous to switch were 7.6 ± 7.1µg/mL. After switch, SC-IFX levels at week 8, 16, 6 month and 12 months were 13.9 ± 6.3, 17.4 ± 10.8, 15.1 ± 8.3 and 14.8 ± 6.1µg/mL.
When comparing CD and UC, clinical response at week 8 was significantly higher in CD patients (81.2% vs 52.0%, p=0.011), while no significant differences were observed at week 16 (p=0.350), month 6 (p=0.687) or month 12 (p=0.456). Regarding clinical remission, no statistically significant differences were found (week 8 (53.1% vs 40.0%, p=0.380), week 16 (57.8% vs 60.0%, p=1.000), month 6 (54.7% vs 60.0%, p=0.829), and month 12 (37.5% vs 40.0%, p=1.000)).
Adverse events were reported in 2.5% of patients, and no serious adverse events occurred.

Conclusion

Switching from IV-IFX to SC-IFX could be an effective and safe strategy to recapture clinical remission in IBD patients with active disease that lose response to IV-IFX.