Desalegn, Hailemichael

INTRODUCTION: Nosocomial infections (NIs) in cirrhosis are associated with high mortality but could be preventable. Logistic regression (LR) models have failed to identify high-risk patients. We aimed to develop machine learning (ML) models to predict NI. METHODS: The CLEARED consortium consists of prospectively enrolled cirrhosis inpatients from >120 centers. Using day-of-admission clinical data, 3 ML approaches (random forest [RF], extreme gradient boosting, and neural networks [NNs]) were used to predict NI. Data were split 80:20 for training and testing stratified by the outcome. Models were compared using area under the receiver operating characteristic curve (AUC). RESULTS: In total, 8,263 patients (55.90 ± 13.34 years; 64.1% men) from 127 centers in 37 countries were included. NI developed in 869 (10.5%), a median of 6 (4–11) days of postadmission. Major NIs were respiratory (29.6%) and urinary tract infection (15.7%), and spontaneous bacterial peritonitis (13.5%). NIs occurred more frequently in patients from low/low-middle income countries and those with severe liver disease, alcohol etiology, and admission infections. NIs were associated with inpatient mortality (31.9% vs 8.1%, P < 0.001) and liver transplantation (4.8% vs 1.9%, P < 0.001). Although the RF model (AUC 0.69) showed good calibration (Brier score 0.09), outperforming extreme gradient boosting, neural network, and LR models (AUC 0.66 for all; LR comparison P = 0.043), no model achieved AUC ≥0.80 for clinical utility. At 10% predicted probability threshold, the RF model demonstrated only 75.4% sensitivity, 52.9% specificity, and 15.9% positive predictive value (PPV). DISCUSSION: NIs cannot be accurately predicted from day-of-admission data using ML models, even in a large, prospective, global cirrhosis cohort. Every hospitalized patient with cirrhosis should receive protocolized infection control measures.

PURPOSE Hepatocellular carcinoma (HCC), the fourth most common cancer in Africa, has a dismal overall survival of only 3 months like in sub-Saharan Africa. This is affected by the low gross domestic product and human development index, absence of coherent guidelines, and other factors. METHODS An open forum for HCC-experienced health care workers from Africa and the rest of the world was held in October 2021. Participants completed a survey to help assess the real-life access to screening, diagnoses, and treatment in the North and Southern Africa (NS), East and West Africa (EW), Central Africa (C), and the rest of the world. RESULTS Of 461 participants from all relevant subspecialties, 372 were from Africa. Most African participants provided hepatitis B vaccination and treatment for hepatitis B and C. More than half of the participants use serum alpha-fetoprotein and ultrasound for surveillance. Only 20% reported using image-guided diagnostic liver biopsy. The Barcelona Clinic Liver Cancer is the most used staging system (52%). Liver transplant is available for only 28% of NS and 3% EW. C reported a significantly lower availability of resection. Availability of local therapy ranged from 94% in NS to 62% in C. Sorafenib is the most commonly used systemic therapy (66%). Only 12.9% reported access to other medications including immune checkpoint inhibitors. Besides 42% access to regorafenib in NS, second-line treatments were not provided. CONCLUSION Similarities and differences in the care for patients with HCC in Africa are reported. This reconfirms the major gaps in access and availability especially in C and marginally less so in EW. This is a call for concerted multidisciplinary efforts to achieve and sustain a reduction in incidence and mortality from HCC in Africa.

Background and Aims: Fatty liver disease is a major public health threat due to its very high prevalence and related morbidity and mortality. Focused and dedicated interventions are urgently needed to target disease prevention, treatment, and care. Approach and Results: We developed an aligned, prioritized action agenda for the global fatty liver disease community of practice. Following a Delphi methodology over 2 rounds, a large panel (R1 n = 344, R2 n = 288) reviewed the action priorities using Qualtrics XM, indicating agreement using a 4-point Likert-scale and providing written feedback. Priorities were revised between rounds, and in R2, panelists also ranked the priorities within 6 domains: epidemiology, treatment and care, models of care, education and awareness, patient and community perspectives, and leadership and public health policy. The consensus fatty liver disease action agenda encompasses 29 priorities. In R2, the mean percentage of “agree” responses was 82.4%, with all individual priorities having at least a super-majority of agreement (> 66.7% “agree”). The highest-ranked action priorities included collaboration between liver specialists and primary care doctors on early diagnosis, action to address the needs of people living with multiple morbidities, and the incorporation of fatty liver disease into relevant non-communicable disease strategies and guidance. Conclusions: This consensus-driven multidisciplinary fatty liver disease action agenda developed by care providers, clinical researchers, and public health and policy experts provides a path to reduce the prevalence of fatty liver disease and improve health outcomes. To implement this agenda, concerted efforts will be needed at the global, regional, and national levels.