Background and aims Anti-neoplastic immunotherapy has revolutionized cancer management; however, its safety profile with respect to liver-related injury remains largely unexplored. Herein, we analyzed a United States national database to determine the incidence, mortality, and predictors of hepatotoxicity in the setting of anti-neoplastic immunotherapy. Methods This was a nationwide retrospective study of hospital encounters from 2011 to 2014 using the National Inpatient Sample (NIS) database. We utilized the International Classification of Diseases, Ninth Revision (ICD-9) coding system to identify all adult patients who underwent anti-neoplastic immunotherapy during hospitalization. The primary outcome was the incidence of hepatotoxicity during the same hospitalization. Secondary outcomes included in-hospital mortality as well as socioeconomic and ethno-racial predictors of hepatotoxicity. Analyses were performed using IBM SPSS Statistics 23.0. Results The sample included 3002 patients who underwent inpatient anti-neoplastic immunotherapy. The incidence of hepatotoxicity was 10.1%, which was significantly higher as compared to a matched inpatient population (adjusted odds ratio (aOR) 4.93, 95% confidence interval (CI): 3.80–6.40, P = 0.001). No significant mortality difference was seen in those that developed hepatotoxicity compared to those who did not (aOR 0.47, 95% CI: 0.03–8.03, P = 0.612). Age under 60 (aOR 1.56, 95% CI: 1.23–1.78, P = 0.050) and white race (aOR 1.85, 95% CI: 1.35–2.04, P<0.010) were independent risk factors for developing immunotherapy-associated hepatotoxicity. Conclusions In this large, nationwide database analysis, we found that anti-neoplastic immunotherapy was associated with a nearly five-fold risk of in-hospital hepatotoxicity as compared to a matched inpatient population, though without an associated mortality difference. Additionally, younger age and white race were identified as predictors of immunotherapy-associated hepatotoxicity. Heightened vigilance and prospective investigation of the risk factors and liver-related adverse effects of anti-neoplastic immunotherapy are warranted.

Background and aims Noninvasive predictors of choledocholithiasis have generally exhibited marginal performance characteristics. We aimed to identify noninvasive independent predictors of endoscopic retrograde cholangiopancreatography (ERCP)-confirmed choledocholithiasis and accordingly developed predictive machine learning models (MLMs). Methods Clinical data of consecutive patients undergoing first-ever ERCP for suspected choledocholithiasis from 2015 to 2019 were abstracted from a prospectively-maintained database. Multiple logistic regression was used to identify predictors of ERCP-confirmed choledocholithiasis. MLMs were then trained to predict ERCP-confirmed choledocholithiasis using pre-ERCP ultrasound (US) imaging only as well as using all available noninvasive imaging (US, computed tomography, and/or magnetic resonance cholangiopancreatography). The diagnostic performance of American Society for Gastrointestinal Endoscopy (ASGE) “high-likelihood” criteria was compared to MLMs. Results We identified 270 patients (mean age 46 years, 62.2% female, 73.7% Hispanic/Latino, 59% with noninvasive imaging positive for choledocholithiasis) with native papilla who underwent ERCP for suspected choledocholithiasis, of whom 230 (85.2%) were found to have ERCP-confirmed choledocholithiasis. Logistic regression identified choledocholithiasis on noninvasive imaging (odds ratio (OR) = 3.045, P = 0.004) and common bile duct (CBD) diameter on noninvasive imaging (OR = 1.157, P = 0.011) as predictors of ERCP-confirmed choledocholithiasis. Among the various MLMs trained, the random forest-based MLM performed best; sensitivity was 61.4% and 77.3% and specificity was 100% and 75.0%, using US-only and using all available imaging, respectively. ASGE high-likelihood criteria demonstrated sensitivity of 90.9% and specificity of 25.0%; using cut-points achieving this specificity, MLMs achieved sensitivity up to 97.7%. Conclusions MLMs using age, sex, race/ethnicity, presence of diabetes, fever, body mass index (BMI), total bilirubin, maximum CBD diameter, and choledocholithiasis on pre-ERCP noninvasive imaging predict ERCP-confirmed choledocholithiasis with good sensitivity and specificity and outperform the ASGE criteria for patients with suspected choledocholithiasis.