Development and Verification of Time-Series Deep Learning for Drug-Induced Liver Injury Detection in Patients Taking Angiotensin II Receptor Blockers: A Multicenter Distributed Research Network Approach / 대한의료정보학회지

Objectives@#The objective of this study was to develop and validate a multicenter-based, multi-model, time-series deep learning model for predicting drug-induced liver injury (DILI) in patients taking angiotensin receptor blockers (ARBs). The study leveraged a national-level multicenter approach, utilizing electronic health records (EHRs) from six hospitals in Korea. @*Methods@#A retrospective cohort analysis was conducted using EHRs from six hospitals in Korea, comprising a total of 10,852 patients whose data were converted to the Common Data Model. The study assessed the incidence rate of DILI among patients taking ARBs and compared it to a control group. Temporal patterns of important variables were analyzed using an interpretable timeseries model. @*Results@#The overall incidence rate of DILI among patients taking ARBs was found to be 1.09%. The incidence rates varied for each specific ARB drug and institution, with valsartan having the highest rate (1.24%) and olmesartan having the lowest rate (0.83%). The DILI prediction models showed varying performance, measured by the average area under the receiver operating characteristic curve, with telmisartan (0.93), losartan (0.92), and irbesartan (0.90) exhibiting higher classification performance. The aggregated attention scores from the models highlighted the importance of variables such as hematocrit, albumin, prothrombin time, and lymphocytes in predicting DILI. @*Conclusions@#Implementing a multicenter-based timeseries classification model provided evidence that could be valuable to clinicians regarding temporal patterns associated with DILI in ARB users. This information supports informed decisions regarding appropriate drug use and treatment strategies.

Risk Factors for Surgery in Patients with Intestinal Behçet’s Disease During Anti-Tumor Necrosis Factor-Alpha Therapy

Purpose@#Behçet’s disease (BD) is a chronic inflammatory immune-mediated disease involving multiorgan systems. Gastrointestinal (GI) manifestations of BD include abdominal pain, vomiting, GI bleeding, fistula formation, obstruction, and perforation that might require surgery. Recently, anti-tumor necrosis factor-alpha (anti-TNF-α) therapy has been shown to have favorable outcomes in patients with intestinal BD who are refractory to conventional therapy. This study sought to figure out the risk factors for undergoing surgery during anti-TNF-α therapy in patients with intestinal BD. @*Materials and Methods@#In this retrospective analysis of intestinal BD patients who were treated with anti-TNF-α, we collected the baseline patient data including comorbidities, clinical, endoscopic, and radiologic characteristics, and the Disease Activity Index for Intestinal Behçet’s Disease at the time of anti-TNF-α initiation. Each potential risk factor was compared. For multivariate analysis, Cox regression was used. @*Results@#A total of 62 patients were considered eligible for analysis, and 15 of them (24.1%) underwent surgery. In univariate analysis, the presence of extraintestinal manifestation, such as joint symptoms and erythrocyte sedimentation rate (ESR), were significantly associated with surgery during therapy. In multivariate analysis, drug response within 4 weeks [hazard ratio (HR), 64.59], skin and joint manifestation (HR, 10.23 and HR, 6.22), geographic ulcer (HR, 743.97), and ESR >42.5 mm/h (HR, 9.16) were found to be factors predictive of undergoing surgery during anti-TNF-α therapy. @*Conclusion@#We found five risk factors predictive of surgery in patients with intestinal BD receiving anti-TNF-α therapy, which can guide physicians in selecting appropriate patients between anti-TNF-α therapy and early surgery.

Radiology plus ileocolonoscopy versus radiology alone in Crohn’s disease: prognosis prediction and mutual agreement

Background/Aims@#The optimal tools for monitoring Crohn’s disease (CD) are controversial. We compared radiology plus ileocolonoscopy and radiology alone in terms of prognosis prediction and evaluated the agreement between radiologic and ileocolonoscopic findings in patients with CD. @*Methods@#Patients with CD who were followed up with computed tomography enterography (CTE) or magnetic resonance enterography (MRE) alone or CTE or MRE plus ileocolonoscopy were retrospectively recruited. Time to relapse was investigated to evaluate the difference in prognosis using the log-rank and Cox regression tests, and the agreement between radiologic and ileocolonoscopic findings was determined using a kappa value. @*Results@#A total of 501 patients with CD in clinical remission who underwent CTE or MRE and/or ileocolonoscopy were analyzed. Of these, 372 (74.3%) patients underwent CTE or MRE alone and 129 (25.7%) patients underwent CTE or MRE plus ileocolonoscopy. The cumulative maintenance rate of clinical remission between the two groups was not significantly different (p = 0.526, log-rank test). In multivariate analysis, age <40 years (hazard ratio [HR], 2.756; 95% confidence interval [CI], 1.263 to 6.013) and a history of steroid use (HR, 2.212; 95% CI, 1.258 to 3.577) were found to independently predict an increased risk for clinical relapse in patients with CD in clinical remission. Radiologic and ileocolonoscopic findings had a moderate degree of agreement (κ = 0.401, –0.094 to 0.142). The comparison of agreement between radiologic and ileocolonoscopic findings was the highest in the anastomotic site (κ = 0.749, –0.168 to 0.377). @*Conclusions@#Radiology plus ileocolonoscopy was not superior to radiology alone in predicting the prognosis of CD.

Continued Postoperative Use of Tumor Necrosis Factor-αInhibitors for the Prevention of Crohn’s Disease Recurrence

Background/Aims@#Many patients with Crohn’s disease (CD) undergo intestinal resection during the disease course. Despite surgery, postoperative recurrence (POR) commonly occurs. Although postoperative use of tumor necrosis factor α (TNF-α) inhibitors is known to be effective in preventing POR, few studies have evaluated the effectiveness of continuing the same TNF-α inhibitors postoperatively in patients who received TNF-ɑ inhibitors before surgery. @*Methods@#This retrospective observational study was performed in a single tertiary medical center. We retrospectively reviewed patients who had undergone the first intestinal resection due to CD and divided them into two groups: TNF-α inhibitor users in both the preoperative and postoperative periods, and TNF-α inhibitor users in only the preoperative period. We compared the clinical outcomes between these two groups. @*Results@#In total, 45 patients who used TNF-α inhibitors preoperatively were recruited. Among them, TNF-α inhibitors were used postoperatively in 20 patients (44.4%). The baseline characteristics except age at diagnosis were similar in both groups. The rates of surgical and endoscopic recurrence were not different between the two groups, but the cumulative clinical recurrence rate was significantly lower in the postoperative TNF-α inhibitors group (log-rank p=0.003). In multivariate Cox regression analysis, postoperative TNF-α inhibitors use was significantly associated with a decreased risk of clinical recurrence (adjusted hazard ratio, 0.204; 95% confidence interval, 0.060 to 0.691; p=0.011). @*Conclusions@#Continuing TNF-α inhibitors postoperatively in patients who were receiving TNF-α inhibitors before surgery significantly reduced the rate of clinical recurrence. For patients with CD who received TNF-α inhibitors preoperatively, continuing their use after surgery could be recommended.

Prognostic Value of Terminal Ileal Inflammation in Patients with Ulcerative Colitis

Background/Aims@#Few studies have investigated terminal ileal lesions and their prognostic value in patients with ulcerative colitis (UC). We evaluated the clinical significance of these lesions as a prognostic factor in patients with UC who were in clinical remission. @*Methods@#We retrospectively selected 567 of 4,066 colonoscopy reports that included positive findings from orificial observations of the terminal ileum (TI) and appendix in patients with UC. We finally recruited patients who were in clinical remission (n=204). We compared the clinical courses, including relapse and other prognostic parameters associated with UC, between the groups. @*Results@#The baseline patient characteristics were not significantly different between patients with (n=69, TI+ group) and without TI lesions (n=135, TI– group), although there were more never-smokers in the TI+ group (n=57 [82.6%] in the TI+ group; n=86 [63.7%] in the TI– group; p=0.005). Of note, appendiceal orifice inflammation (AOI) was less frequently found in the TI+ group (14.5%) than in the TI– group (71.9%, p<0.001). The cumulative relapse rate was numerically higher in the TI– group, but it was not significantly different according to the Kaplan-Meier analysis (p=0.116). Multivariate Cox regression analysis also revealed advanced age at diagnosis as the most significant factor (adjusted hazard ratio, 0.964; 95% confidence interval, 0.932 to 0.998; p=0.037), but neither TI inflammation nor AOI were significantly associated with the cumulative relapse rate in patients with UC in clinical remission. @*Conclusions@#For patients with UC in clinical remission, neither terminal ileal lesions nor AOI had significant clinical or predictive value for future relapse.

Risk of Diabetes in Subjects with Positive Fecal Immunochemical Test: A Nationwide Population-Based Study

Background@#Positive fecal immunochemical test (FIT) results have been recently suggested as a risk factor for systemic inflammation. Diabetes induces inflammation in the gastrointestinal tract via several ways. We investigated the association between FIT results and the incidence of diabetes. @*Methods@#A total of 7,946,393 individuals aged ≥50 years from the National Cancer Screening Program database who underwent FIT for colorectal cancer (CRC) screening from 2009 to 2012 were enrolled. The primary outcome was newly diagnosed diabetes based on the International Classification of Disease 10th revision codes and administration of anti-diabetic medication during the follow-up period. @*Results@#During a mean follow-up of 6.5 years, the incidence rates of diabetes were 11.97, 13.60, 14.53, and 16.82 per 1,000 personyears in the FIT negative, one-positive, two-positive, and three-positive groups, respectively. The hazard ratios (HRs) for the incidence of diabetes were 1.14 (95% confidence interval [CI], 1.12 to 1.16; HR, 1.21; 95% CI, 1.16 to 1.27; and HR, 1.40; 95% CI, 1.28 to 1.55) in the one-positive, two-positive, and three-positive FIT groups compared with the FIT negative group, respectively. The effect was consistent in individuals with normal fasting blood glucose (adjusted HR 1.55 vs. 1.14, P for interaction <0.001). @*Conclusion@#Positive FIT results were associated with a significantly higher risk of diabetes, suggesting that the FIT can play a role not only as a CRC screening tool, but also as a surrogate marker of systemic inflammation; thus, increasing the diabetes risk.

Long-term Outcomes after the Discontinuation of Anti-TumorNecrosis Factor-α Therapy in Patients with Inflammatory BowelDisease under Clinical Remission: A Korean Association for the Study of Intestinal Disease Multicenter Study

Background/Aims@#Our study aimed to evaluate the long-term outcomes and risk factors forrelapse after anti-tumor necrosis factor (TNF)-α cessation in inflammatory bowel disease (IBD) patients because they are not well established. @*Methods@#A retrospective multicenter cohort study was conducted involving patients with Crohn’s disease (CD) or ulcerative colitis (UC) from 10 referral hospitals in Korea who discontinued firstline anti-TNF therapy after achieving clinical remission. @*Results@#A total of 109 IBD patients (71 CD and 38 UC) with a median follow-up duration of 56months were analyzed. The cumulative relapse rates at 1, 3, and 5 years were 11.3%, 46.7%, and 62.5% for CD patients and 28.9%, 45.3%, and 60.9% for UC patients. Multivariable Coxanalysis revealed that discontinuation owing to the clinician’s decision was associated with lower risk of relapse (vs patient’s preference: hazard ratio [HR], 0.13; 95% confidence interval [CI], 0.04 to 0.48; p=0.002) and adalimumab use was associated with higher risk of relapse (vs infliximab: HR, 4.42; 95% CI, 1.24 to 17.74; p=0.022) in CD patients. Mucosal healing was associated with lower risk of relapse (vs nonmucosal healing: HR, 0.12; 95% CI, 0.02 to 0.83; p=0.031) in UC patients. Anti-TNF re-induction was provided to 52 patients, and a response was obtained in 50 patients. However, 25 of them discontinued retreatment owing to a loss of response (n=15), the patient’s preference (n=6), and other factors (n=4). @*Conclusions@#More than 60% of IBD patients in remission under anti-TNF therapy relapsed within 5 years of treatment cessation. Anti-TNF re-induction was effective. However, half of the patients discontinued anti-TNF therapy, and 50% of these patients discontinued treatment owing to loss of response.

Long-term Outcomes after the Discontinuation of Anti-TumorNecrosis Factor-α Therapy in Patients with Inflammatory BowelDisease under Clinical Remission: A Korean Association for the Study of Intestinal Disease Multicenter Study

Background/Aims@#Our study aimed to evaluate the long-term outcomes and risk factors forrelapse after anti-tumor necrosis factor (TNF)-α cessation in inflammatory bowel disease (IBD) patients because they are not well established. @*Methods@#A retrospective multicenter cohort study was conducted involving patients with Crohn’s disease (CD) or ulcerative colitis (UC) from 10 referral hospitals in Korea who discontinued firstline anti-TNF therapy after achieving clinical remission. @*Results@#A total of 109 IBD patients (71 CD and 38 UC) with a median follow-up duration of 56months were analyzed. The cumulative relapse rates at 1, 3, and 5 years were 11.3%, 46.7%, and 62.5% for CD patients and 28.9%, 45.3%, and 60.9% for UC patients. Multivariable Coxanalysis revealed that discontinuation owing to the clinician’s decision was associated with lower risk of relapse (vs patient’s preference: hazard ratio [HR], 0.13; 95% confidence interval [CI], 0.04 to 0.48; p=0.002) and adalimumab use was associated with higher risk of relapse (vs infliximab: HR, 4.42; 95% CI, 1.24 to 17.74; p=0.022) in CD patients. Mucosal healing was associated with lower risk of relapse (vs nonmucosal healing: HR, 0.12; 95% CI, 0.02 to 0.83; p=0.031) in UC patients. Anti-TNF re-induction was provided to 52 patients, and a response was obtained in 50 patients. However, 25 of them discontinued retreatment owing to a loss of response (n=15), the patient’s preference (n=6), and other factors (n=4). @*Conclusions@#More than 60% of IBD patients in remission under anti-TNF therapy relapsed within 5 years of treatment cessation. Anti-TNF re-induction was effective. However, half of the patients discontinued anti-TNF therapy, and 50% of these patients discontinued treatment owing to loss of response.

Endoscopic Features of Upper Gastrointestinal Tract in Patients with Systemic Sclerosis Compared to the Healthy Control

OBJECTIVE: To characterize the endoscopic features of upper gastrointestinal tract in patients with systemic sclerosis (SSc) compared with those in the healthy controls. METHODS: Data on esophagogastroduodenoscopy (EGD) in 180 patients with SSc (SSc group) were compared with that from the 181 age- and sex-matched healthy control who underwent EGD for routine check-up (control group). Clinical data of participants at the time of EGD (defined as baseline) were collected from electric medical record. Endoscopic findings were evaluated by two experts with blinded to their clinical features. Primary outcome of the study was prevalence of each endoscopic lesion between the two groups. RESULTS: The mean±standard deviation age and disease duration in the SSc group at baseline were 55.3±11.8 and 2.9±3.7 years, respectively. Compared to the control group, SSc group more frequently showed reflux esophagitis (32.8% vs. 9.4%, p < 0.001). In contrast, prevalence of atrophic gastritis was significantly lower in the SSc group (8.3% vs. 29.3%, p < 0.001). This result was consistent in the multivariable analysis where patients' age and concomitant proton pump inhibitor use were adjusted. There was no case of gastric antral vascular ectasia (GAVE) in both groups. However, 29 (16.1%) patients in SSc group showed a clinically significant anemia (hemoglobin < 10 mg/dL), with none of the endoscopic features showed significant associations with the outcome. CONCLUSION: Patients with SSc showed significantly lower prevalence of atrophic gastritis. There was no case of GAVE, which suggests that clinical phenotype of the SSc could be different according to the ethnicity or geographic region.

Erratum: Endoscopic Features of Upper Gastrointestinal Tract in Patients with Systemic Sclerosis Compared to the Healthy Control

The original version of this article contained an error of the acknowledgement (funding source).

Multidrug-resistant Disseminated Tuberculosis Related to Infliximab in a Patient with Ulcerative Colitis and Negative Evaluation for Latent Tuberculosis / 대한소화기학회지

Anti-tumor necrosis factor (anti-TNF) is an effective biological agent for the treatment of moderate-to-severe active ulcerative colitis (UC) refractory to conventional therapy. On the other hand, anti-TNF therapy is strongly associated with a potential risk of tuberculosis (TB). Active TB is a critical complication that makes it difficult to treat patients who require anti-TNF for the treatment of UC refractory to conventional therapy. Based on the clinical guidelines, patients with inflammatory bowel disease (IBD) are strongly recommended to screen for latent TB before anti-TNF administration. Considering the possibility of active or reactivated TB related to anti-TNF therapy, all patients with IBD should be monitored closely for TB during anti-TNF therapy, irrespective of the screening results for latent TB. In particular, the risk of anti-TNF-related multidrug-resistant TB (MDR-TB) in patients with IBD has not been elucidated. This paper reports the first case of disseminated MDR-TB that developed in a UC patient receiving infliximab despite the negative evaluation for latent TB screening.

Multidrug-resistant Disseminated Tuberculosis Related to Infliximab in a Patient with Ulcerative Colitis and Negative Evaluation for Latent Tuberculosis / 대한소화기학회지

Anti-tumor necrosis factor (anti-TNF) is an effective biological agent for the treatment of moderate-to-severe active ulcerative colitis (UC) refractory to conventional therapy. On the other hand, anti-TNF therapy is strongly associated with a potential risk of tuberculosis (TB). Active TB is a critical complication that makes it difficult to treat patients who require anti-TNF for the treatment of UC refractory to conventional therapy. Based on the clinical guidelines, patients with inflammatory bowel disease (IBD) are strongly recommended to screen for latent TB before anti-TNF administration. Considering the possibility of active or reactivated TB related to anti-TNF therapy, all patients with IBD should be monitored closely for TB during anti-TNF therapy, irrespective of the screening results for latent TB. In particular, the risk of anti-TNF-related multidrug-resistant TB (MDR-TB) in patients with IBD has not been elucidated. This paper reports the first case of disseminated MDR-TB that developed in a UC patient receiving infliximab despite the negative evaluation for latent TB screening.

Clinical features and outcomes in spontaneous intramural small bowel hematoma: cohort study and literature review

BACKGROUND/AIMS: Spontaneous intramural small bowel hematoma (SISBH) is an extremely rare complication of anticoagulant or antiplatelet therapy. We assessed the clinical characteristics and outcomes of patients with SISBH according to the anatomical location of the hematoma. METHODS: From January 2003 to February 2016, medical records for all patients hospitalized for SISBH at 2 tertiary referral hospitals were retrospectively reviewed. The primary outcome was requirement for surgery. RESULTS: A total of 37 patients were enrolled. The mean age was 74.1 years. Among them, 33 patients (89.2%) were taking anticoagulant and/or antiplatelet agents. Duodenal intramural hematoma was detected in 4 patients (10.8%), jejunal in 16 (43.2%), and ileal in 17 (45.9%). Compared to jejunal and ileal involvement, duodenal intramural hematoma was significantly associated with high Charlson comorbidity index and low levels of white blood cells, hemoglobin, and platelets in the blood. SISBH in the duodenum was related to thrombocytopenia in 3 patients following systemic chemotherapy for malignancy. All patients with SISBH showed clinical improvement with conservative therapy. Mean length of hospital stay was 9.35 days. Independent predictors of a hospital stay of more than 7 days were body weight less than 60 kg (odds ratio [OR], 12.213; 95% confidence interval [CI], 1.755–84.998; P=0.011) and a history of cerebrovascular accidents (OR, 6.667; 95% CI, 1.121–39.650; P=0.037). CONCLUSIONS: Compared to jejunal and ileal involvement, thrombocytopenia may result in spontaneous duodenal intramural hematoma among patients who are treated with systemic chemotherapy for malignancies. Patients with SISBH have excellent clinical outcomes with conservative therapy regardless of the anatomical location of the hematoma.

A Case of Idiopathic Severe Acute Pancreatitis following Cesarean Section Delivery / 대한소화기학회지

Acute pancreatitis rarely occurs in the postpartum period. Furthermore, there are very few reports of it after cesarean section delivery. A 35-year-old woman presented with dyspnea and abdominal distension on the third day after cesarean section delivery. Under a suspicion of acute pancreatitis, she was initially managed with conservative treatment. However, she developed intra-abdominal fluid collections and gastric bleeding, which were managed with percutaneous drainage, endoscopic hemostasis, and angiographic embolization. She was discharged with good clinical recovery. Postpartum pancreatitis, especially after cesarean section, is rare; however, its management is not different from that for usual pancreatitis.