Machine learning-based decision tool for selecting patients with idiopathic acute pancreatitis for endosonography to exclude a biliary aetiology

BACKGROUND Biliary microlithiasis/sludge is detected in approximately 30%of patients with idiopathic acute pancreatitis(IAP).As recurrent biliary pancreatitis can be prevented,the underlying aetiology of IAP should be established.AIM To develop a machine learning(ML)based decision tool for the use of endosonography(EUS)in pancreatitis patients to detect sludge and microlithiasis.METHODS We retrospectively used routinely recorded clinical and laboratory parameters of 218 consecutive patients with confirmed AP admitted to our tertiary care hospital between 2015 and 2020.Patients who did not receive EUS as part of the diagnostic work-up and whose pancreatitis episode could be adequately explained by other causes than biliary sludge and microlithiasis were excluded.We trained supervised ML classifiers using H_(2)O.ai automatically selecting the best suitable predictor model to predict microlithiasis/sludge.The predictor model was further validated in two independent retrospective cohorts from two tertiary care centers(117 patients).RESULTS Twenty-eight categorized patients’variables recorded at admission were identified to compute the predictor model with an accuracy of 0.84[95%confidence interval(CI):0.791-0.9185],positive predictive value of 0.84,and negative predictive value of 0.80 in the identification cohort(218 patients).In the validation cohort,the robustness of the prediction model was confirmed with an accuracy of 0.76(95%CI:0.673-0.8347),positive predictive value of 0.76,and negative predictive value of 0.78(117 patients).CONCLUSION We present a robust and validated ML-based predictor model consisting of routinely recorded parameters at admission that can predict biliary sludge and microlithiasis as the cause of AP.

Endoscopic ultrasound diagnostic gain over computed tomography and magnetic resonance cholang-iopancreatography in defining etiology of idiopathic acute pancreatitis

BACKGROUND About 10%-30%of acute pancreatitis remain idiopathic(IAP)even after clinical and imaging tests,including abdominal ultrasound(US),contrast-enhanced computed tomography(CECT)and magnetic resonance cholangiopancreatography(MRCP).This is a relevant issue,as up to 20%of patients with IAP have recurrent episodes and 26%of them develop chronic pancreatitis.Few data are available on the role of EUS in clarifying the etiology of IAP after failure of one or more cross-sectional techniques.AIM To evaluate the diagnostic gain after failure of one or more previous crosssectional exams.METHODS We retrospectively collected data about consecutive patients with AP and at least one negative test between US,CECT and MRCP,who underwent linear EUS between January 2017 and December 2020.We investigated the EUS diagnostic yield and the EUS diagnostic gain over different combinations of these crosssectional imaging techniques for the etiologic diagnosis of AP.Types and frequency of EUS diagnosis were also analyzed,and EUS diagnosis was compared with the clinical parameters.After EUS,patients were followed-up for a median of 31.5 mo to detect cases of pancreatitis recurrence.RESULTS We enrolled 81 patients(63%males,mean age 61±18,23%with previous cholecystectomy,17%with recurrent pancreatitis).Overall EUS diagnostic yield for AP etiological diagnosis was 79%(20%lithiasis,31%acute on chronic pancreatitis,14%pancreatic solid or cystic lesions,5%pancreas divisum,5%autoimmune pancreatitis,5%ductal abnormalities),while 21%remained idiopathic.US,CECT and MRCP,taken alone or in combination,led to AP etiological diagnosis in 16(20%)patients;among the remaining 65 patients,49(75%)obtained a diagnosis at EUS,with an overall EUS diagnostic gain of 61%.Sixty-eight patients had negative US;among them,EUS allowed etiological diagnosis in 59(87%).Sixty-three patients had a negative CECT;among them,47(74%)obtained diagnosis with EUS.Twenty-four had a negative MRCP;among them,20(83%)had EUS diagnosis.Twenty-one had negative CT+MRCP,of which 17(81%)had EUS diagnosis,with a EUS diagnostic gain of 63%.Patients with biliary etiology and without previous cholecystectomy had higher median values of alanine aminotransferase(154 vs 25,P=0.010),aspartate aminotransferase(95 vs 29,P=0.018),direct bilirubin(1.2 vs 0.6,P=0.015),gammaglutamyl transpeptidase(180 vs 48,P=0.006)and alkaline phosphatase(150 vs 72,P=0.015)Chronic pancreatitis diagnosis was more frequent in patients with recurrent pancreatitis at baseline(82%vs 21%,P<0.001).During the follow-up,AP recurred in 3 patients,one of which remained idiopathic.CONCLUSION EUS is a good test to define AP etiology.It showed a 63%diagnostic gain over CECT+MRCP.In suitable patients,EUS should always be performed in cases of IAP.Further prospective studies are needed.

Biliary microlithiasis,sludge,crystals, microcrystallization,and usefulness of assessment of nucleation time

BACKGROUND:The process of microcrystallization,its sequel and the assessment of nucleation time is ignored.This systematic review aimed to highlight the importance of biliary microlithiasis,sludge,and crystals,and their association with gallstones,unexplained biliary pain,idiopathic pancreatitis, and sphincter of Oddi dysfunction.DATA SOURCES:Three reviewers performed a literature search of the PubMed database.Key words used were'biliary microlithiasis','biliary sludge','bile crystals','cholesterol crystallisation','bile microscopy','microcrystal formation of bile','cholesterol monohydrate crystals','nucleation time of cholesterol','gallstone formation','sphincter of Oddi dysfunction'and'idiopathic pancreatitis'.Additional articles were sourced from references within the studies from the PubMed search.RESULTS:We found that biliary microcrystals account for almost all patients with gallstone disease,7%to 79%with idiopathic pancreatitis,83%with unexplained biliary pain, and 25%to 60%with altered biliary and pancreatic sphincter function.Overall,the detection of biliary microcrystals in gallstone disease has a sensitivity ranging from 55%to 87%and a specificity of 100%.In idiopathic pancreatitis,the presence of microcrystals ranges from 47%to 90%.A nucleation time less than 10 days in hepatic bile or ultra-filtered gallbladder bile has a specificity of 100%for cholesterol gallstone disease.CONCLUSIONS:Biliary crystals are associated with gallstone disease,idiopathic pancreatitis,sphincter of Oddi dysfunction, unexplained biliary pain,and post-cholecystectomy biliary pain.Pathways of cholesterol super-saturation,crystallisation, and gallstone formation have been described with scientificsupport.Bile microscopy is a useful method to detect microcrystals and the assessment of nucleation time is a good method of predicting the risk of cholesterol crystallisation.

Steroid-responsive pancreatitides

Autoimmune pancreatitis has received considerable attention,especially due to the marked effect of corticosteroid therapy on its clinical course.Knowledge,especially regarding type 1 autoimmune pancreatitis,has significantly increased over the last decades,and despite significant differences in pathophysiology and outcomes,both type 1 and 2 autoimmune pancreatitis are still considered different types of the same disease.Some have proposed a different nomenclature reflecting these differences.Although the term steroid-responsive pancreatitides may be interpreted as synonymous to type 1 and 2 autoimmune pancreatitis,these are not the only pancreatic conditions that show a response to steroid therapy.Acute pancreatitis caused by vasculitis and connective tissue diseases and acute pancreatitis secondary to checkpoint inhibitors or programmed cell death receptor antibody-mediated blockage cancer therapy may also benefit from steroid treatment.This review presents current concepts on these disorders,aiming to increase awareness,analyze similarities and differences,and propose a new nomenclature that reflects their specific particularities,clustering them under the term"steroid-responsive pancreatitides".