Risk factors and outcomes of peptic ulcer bleed in a Pakistani population: A single-center observational study.

BACKGROUND: Peptic ulcer disease (PUD) remains a significant healthcare burden, contributing to morbidity and mortality worldwide. Despite advancements in therapies, its prevalence persists, particularly in regions with widespread nonsteroidal anti-inflammatory drugs (NSAIDs) use and Helicobacter pylori infection. AIM: To comprehensively analyse the risk factors and outcomes of PUD-related upper gastrointestinal (GI) bleeding in Pakistani population. METHODS: This retrospective cohort study included 142 patients with peptic ulcer bleeding who underwent upper GI endoscopy from January to December 2022. Data on demographics, symptoms, length of stay, mortality, re-bleed, and Forrest classification was collected. RESULTS: The mean age of patients was 53 years, and the majority was men (68.3%). Hematemesis (82.4%) and epigastric pain (75.4%) were the most common presenting symptoms. Most patients (73.2%) were discharged within five days. The mortality rates at one week and one month were 10.6% and 14.8%, respectively. Re-bleed within 24 h and seven days occurred in 14.1% and 18.3% of patients, respectively. Most ulcers were Forrest class (FC) III (72.5%). Antiplatelet use was associated with higher mortality at 7 and 30 d, while alternative medications were linked to higher 24-hour re-bleed rates. NSAID use was associated with more FC III ulcers. Re-bleed at 24 h and 7 d was strongly associated with one-week or one-month mortality. CONCLUSION: Antiplatelet use and rebleeding increase the risk of early mortality in PUD-related upper GI bleeding, while alternative medicines are associated with early rebleeding.

Metabolic puzzle: Exploring liver fibrosis differences in Asian metabolic-associated fatty liver disease subtypes.

BACKGROUND: Metabolic-associated fatty liver disease (MAFLD) is a liver condition marked by excessive fat buildup in the absence of heavy alcohol use. It is primarily linked with metabolic issues like insulin resistance, obesity, and abnormal lipid levels, and is often observed with other conditions such as type 2 diabetes and cardiovascular disease. However, whether the subtypes of MAFLD based on the metabolic disorder differentially impact liver fibrosis is not well explicated, especially in the Asian population. AIM: To compare the severity of liver fibrosis among different MAFLD subtypes. METHODS: A total of 322 adult patients of either gender with fatty liver on ultrasound were enrolled between January to December 2021. MAFLD was defined as per the Asian Pacific Association for the Study of the Liver guidelines. Fibrosis-4 index (Fib-4) and nonalcoholic fatty liver disease fibrosis score (NFS) were employed to evaluate liver fibrosis. RESULTS: The mean age was 44.84 ± 11 years. Seventy-two percent of the patients were female. Two hundred and seventy-three patients were classified as having MAFLD, of which 110 (40.3%) carried a single, 129 (47.3%) had two, and 34 (12.5%) had all three metabolic conditions. The cumulative number of metabolic conditions was related to elevated body mass index, triglyceride (TG) levels, and glycated hemoglobin, lower high-density lipoprotein (HDL) levels, higher liver inflammation (by aspartate aminotransferase and γ-glutamyl transferase), and higher likelihood of fibrosis (by NFS and Fib-4 scores) (P < 0.05 for all). The proportion of advanced fibrosis also increased with an increase in the number of metabolic conditions (4.1%, 25.5%, 35.6%, and 44.1% by NFS and 6.1%, 10.9%, 17%, and 26.5% by Fib-4 for no MAFLD and MAFLD with 1, 2, and 3 conditions, respectively). Among MAFLD patients, those with diabetes alone were the eldest and had the highest mean value of NFS score and Fib-4 score (P < 0.05), while MAFLD patients diagnosed with lean metabolic dysfunction exhibited the highest levels of TG and alanine aminotransferase but the lowest HDL levels (P < 0.05). CONCLUSION: The study suggests that the severity of liver fibrosis in MAFLD patients is influenced by the number and type of metabolic conditions present. Early identification and management of MAFLD, particularly in patients with multiple metabolic conditions, are crucial to prevent liver-related complications.

Trajectories of patients relisted for liver transplantation.

INTRODUCTION AND OBJECTIVES: Recurrent cirrhosis complicates 10-30% of Liver transplants (LT) and can lead to consideration for re-transplantation. We evaluated the trajectories of relisted versus primary listed patients on the waitlist using a competing risk framework. MATERIALS AND METHODS: We retrospectively examined 1,912 patients listed for LT at our centre between from 2012 to 2020. Cox proportional hazard models were used to assess overall survival (OS) by listing type and competing risk analysis Fine-Gray models were used to assess cumulative incidence of transplant by listing type. RESULTS: 1,731 patients were included (104 relisted). 44.2% of relisted patients received exception points vs. 19.8% of primary listed patients (p<0.001). Patients relisted without exceptions, representing those with graft cirrhosis, had the worst OS (HR: 4.17, 95%CI 2.63 - 6.67, p=<0.0001) and lowest instantaneous rate of transplant (HR: 0.56, 95%CI 0.38 - 0.83, p=0.006) than primary listed with exception points. On multivariate analysis listing type, height, bilirubin and INR were associated with cumulative incidence of transplant, while listing type, bilirubin, INR, sodium, creatinine were associated with OS. Within relisted patients, there was a trend towards higher mortality (HR: 1.79, 95%CI 0.91 - 3.52, p=0.08) and low transplant incidence (HR: 0.51, 95%CI 0.22 - 1.15, p=0.07) for graft cirrhosis vs other relisting indications. CONCLUSIONS: Patients relisted for LT are carefully curated and comprise a minority of the waitlist population. Despite their younger age, they have worse liver/kidney function, poor waitlist survival, and decreased transplant incidence suggesting the need for early relisting, while considering standardized exception points.

Development of a Radiomics-Based Model to Predict Graft Fibrosis in Liver Transplant Recipients: A Pilot Study.

Liver Transplantation is complicated by recurrent fibrosis in 40% of recipients. We evaluated the ability of clinical and radiomic features to flag patients at risk of developing future graft fibrosis. CT scans of 254 patients at 3-6 months post-liver transplant were retrospectively analyzed. Volumetric radiomic features were extracted from the portal phase using an Artificial Intelligence-based tool (PyRadiomics). The primary endpoint was clinically significant (≥F2) graft fibrosis. A 10-fold cross-validated LASSO model using clinical and radiomic features was developed. In total, 75 patients (29.5%) developed ≥F2 fibrosis by a median of 19 (4.3-121.8) months. The maximum liver attenuation at the venous phase (a radiomic feature reflecting venous perfusion), primary etiology, donor/recipient age, recurrence of disease, brain-dead donor, tacrolimus use at 3 months, and APRI score at 3 months were predictive of ≥F2 fibrosis. The combination of radiomics and the clinical features increased the AUC to 0.811 from 0.793 for the clinical-only model (p = 0.008) and from 0.664 for the radiomics-only model (p < 0.001) to predict future ≥F2 fibrosis. This pilot study exploring the role of radiomics demonstrates that the addition of radiomic features in a clinical model increased the model's performance. Further studies are required to investigate the generalizability of this experimental tool.

A deep learning framework for personalised dynamic diagnosis of graft fibrosis after liver transplantation: a retrospective, single Canadian centre, longitudinal study.

BACKGROUND: Recurrent graft fibrosis after liver transplantation can threaten both graft and patient survival. Therefore, early detection of fibrosis is essential to avoid disease progression and the need for retransplantation. Non-invasive blood-based biomarkers of fibrosis are limited by moderate accuracy and high cost. We aimed to evaluate the accuracy of machine learning algorithms in detecting graft fibrosis using longitudinal clinical and laboratory data. METHODS: In this retrospective, longitudinal study, we trained machine learning algorithms, including our novel weighted long short-term memory (LSTM) model, to predict the risk of significant fibrosis using follow-up data from 1893 adults who had a liver transplantation between Feb 1, 1987, and Dec 30, 2019, with at least one liver biopsy post transplantation. Liver biopsy samples with indefinitive fibrosis stage and those from patients with multiple transplantations were excluded. Longitudinal clinical variables were collected from transplantation to the date of last available liver biopsy. Deep learning models were trained on 70% of the patients as the training set and 30% of the patients as the test set. The algorithms were also separately tested on longitudinal data from patients in a subgroup of patients (n=149) who had transient elastography within 1 year before or after the date of liver biopsy. Weighted LSTM model performance for diagnosing significant fibrosis was compared against LSTM, other deep learning models (recurrent neural network and temporal convolutional network), and machine learning models (Random Forest, Support vector machines, Logistic regression, Lasso regression, and Ridge regression) and aspartate aminotransferase-to-platelet ratio index (APRI), fibrosis-4 index (FIB-4), and transient elastography. FINDINGS: 1893 people who had a liver transplantation (1261 [67%] men and 632 [33%] women) with at least one liver biopsy between Jan 1, 1992, and June 30, 2020, were included in the study (591 [31%] cases and 1302 [69%] controls). The median age at liver transplantation was 53·7 years (IQR 47·3-59·0) for cases and 55·3 years (48·0 to 61·2) for controls. The median time interval between transplant and liver biopsy was 21 months (5 to 71). The weighted LSTM model (area under the curve 0·798 [95% CI 0·790 to 0·810]) consistently outperformed other methods, including unweighted LSTM (0·761 [0·750 to 0·769]; p=0·031) Recurrent Neural Network (0·736 [0·721 to 0·744]), Temporal Convolutional Networks (0·700 [0·662 to 0·747], and Random Forest 0·679 [0·652 to 0·707]), FIB-4 (0·650 [0·636 to 0·663]) and APRI (0·682 [0·671 to 0·694]) when diagnosing F2 or worse stage fibrosis. In a subgroup of patients with transient elastography results, weighted LSTM was not significantly better at detecting fibrosis (≥F2; 0·705 [0·687 to 0·724]) than transient elastography (0·685 [0·662 to 0·704]). The top ten variables predictive for significant fibrosis were recipient age, primary indication for transplantation, donor age, and longitudinal data for creatinine, alanine aminotransferase, aspartate aminotransferase, total bilirubin, platelets, white blood cell count, and weight. INTERPRETATION: Deep learning algorithms, particularly weighted LSTM, outperform other routinely used non-invasive modalities and could help with the earlier diagnosis of graft fibrosis using longitudinal clinical and laboratory variables. The list of most important predictive variables for the development of fibrosis will enable clinicians to modify their management accordingly to prevent onset of graft cirrhosis. FUNDING: Canadian Institute of Health Research, American Society of Transplantation, Toronto General and Western Hospital Foundation, and Paladin Labs.

Divergent trajectories of lean vs obese non-alcoholic steatohepatitis patients from listing to post-transplant: A retrospective cohort study.

BACKGROUND: Non-alcoholic steatohepatitis (NASH) cirrhosis is the second most common indication for liver transplantation (LT). The role of body mass index (BMI) on outcomes of NASH cirrhosis has been conflicting. AIM: To compare the longitudinal trajectories of patients with lean vs obese NASH cirrhosis, from listing up to post-transplant, having adjusted their BMI for ascites. METHODS: We retrospectively reviewed all adult NASH patients listed for LT in our program from 2012 to 2019. Fine-Gray Competing Risk analyses and Cox Proportional-Hazard Models were performed to examine the cumulative incidence of transplant and survival outcomes respectively. RESULTS: Out of 265 NASH cirrhosis listed patients, 176 were included. Median age was 61.0 years; 46% were females. 111 patients underwent LT. Obese robust patients had better waitlist survival [hazard ratio (HR): 0.12; 95%CI: 0.05-0.29, P < 0.0001] with higher instantaneous rate of transplant (HR: 5.71; 95%CI: 1.26-25.9, P = 0.02). Lean NASH patients had a substantially higher risk of graft loss within 90 d post-LT (1.2% vs 13.8%, P = 0.032) and death post-LT (2.4% vs 17.2%, P = 0.029). 1- 3- and 5-year graft survival was poor for lean NASH (78.6%, 77.3% and 41.7% vs 98.6%, 96% and 85% respectively). Overall patient survival post-LT was significantly worse in lean NASH (HR: 0.17; 95%CI: 0.03-0.86, P = 0.0142) with 83% lower instantaneous rate of death in obese group. CONCLUSION: Although lean NASH is considered to be more benign than obese NASH, our study suggests a paradoxical correlation of lean NASH with waitlist outcomes, and graft and patient survival post-LT.

Long-term mortality risk stratification of liver transplant recipients: real-time application of deep learning algorithms on longitudinal data.

BACKGROUND: Survival of liver transplant recipients beyond 1 year since transplantation is compromised by an increased risk of cancer, cardiovascular events, infection, and graft failure. Few clinical tools are available to identify patients at risk of these complications, which would flag them for screening tests and potentially life-saving interventions. In this retrospective analysis, we aimed to assess the ability of deep learning algorithms of longitudinal data from two prospective cohorts to predict complications resulting in death after liver transplantation over multiple timeframes, compared with logistic regression models. METHODS: In this machine learning analysis, model development was done on a set of 42 146 liver transplant recipients (mean age 48·6 years [SD 17·3]; 17 196 [40·8%] women) from the Scientific Registry of Transplant Recipients (SRTR) in the USA. Transferability of the model was further evaluated by fine-tuning on a dataset from the University Health Network (UHN) in Canada (n=3269; mean age 52·5 years [11·1]; 1079 [33·0%] women). The primary outcome was cause of death, as recorded in the databases, due to cardiovascular causes, infection, graft failure, or cancer, within 1 year and 5 years of each follow-up examination after transplantation. We compared the performance of four deep learning models against logistic regression, assessing performance using the area under the receiver operating characteristic curve (AUROC). FINDINGS: In both datasets, deep learning models outperformed logistic regression, with the Transformer model achieving the highest AUROCs in both datasets (p<0·0001). The AUROC for the Transformer model across all outcomes in the SRTR dataset was 0·804 (99% CI 0·795-0·854) for 1-year predictions and 0·733 (0·729-0·769) for 5-year predictions. In the UHN dataset, the AUROC for the top-performing deep learning model was 0·807 (0·795-0·842) for 1-year predictions and 0·722 (0·705-0·764) for 5-year predictions. AUROCs ranged from 0·695 (0·680-0·713) for prediction of death from infection within 5 years to 0·859 (0·847-0·871) for prediction of death by graft failure within 1 year. INTERPRETATION: Deep learning algorithms can incorporate longitudinal information to continuously predict long-term outcomes after liver transplantation, outperforming logistic regression models. Physicians could use these algorithms at routine follow-up visits to identify liver transplant recipients at risk for adverse outcomes and prevent these complications by modifying management based on ranked features. FUNDING: Canadian Donation and Transplant Research Program, CIFAR AI Chairs Program.

Impact of sepsis and non-communicable diseases on prognostic models to predict the outcome of hospitalized chronic liver disease patients.

AIM: To evaluate the impact of sepsis and non-communicable diseases (NCDs) on the outcome of decompensated chronic liver disease (CLD) patients. METHODS: In this cross-sectional study, medical records of patients with CLD admitted to the Gastroenterology unit at the Aga Khan University Hospital were reviewed. Patients older than 18 years with decompensation of CLD (i.e., jaundice, ascites, encephalopathy, and/or upper gastrointestinal bleed) as the primary reason for admission were included, while those who were admitted for reasons other than decompensation of CLD were excluded. Each patient was followed for 6 wk after index admission to assess mortality, prolonged hospital stay (> 5 d), and early readmission (within 7 d). RESULTS: A total of 399 patients were enrolled. The mean age was 54.3 ± 11.7 years and 64.6% (n = 258) were male. Six-week mortality was 13% (n = 52). Prolonged hospital stay and readmission were present in 18% (n = 72) and 7% (n = 28) of patients, respectively. NCDs were found in 47.4% (n = 189) of patients. Acute kidney injury, sepsis, and non-ST elevation myocardial infarction were found in 41% (n = 165), 17.5% (n = 70), and 1.75% (n = 7) of patients, respectively. Upon multivariate analysis, acute kidney injury, non-ST elevation myocardial infarction, sepsis, and coagulopathy were found to be statistically significant predictors of mortality. While chronic kidney disease (CKD), low albumin, and high Model for End-Stage Liver Disease (MELD)-Na score were found to be statistically significant predictors of morbidity. Addition of sepsis in conventional MELD score predicted mortality even better than MELD-Na (area under receiver operating characteristic: 0.735 vs 0.686; P < 0.001). Among NCDs, CKD was found to increase morbidity independently. CONCLUSION: Addition of sepsis improved the predictability of MELD score as a prognostic marker for mortality in patients with CLD. Presence of CKD increases the morbidity of patients with CLD.