A machine learning-based choledocholithiasis prediction tool to improve ERCP decision making: a proof-of-concept study.

BACKGROUND: Previous studies demonstrated limited accuracy of existing guidelines for predicting choledocholithiasis, leading to overutilization of endoscopic retrograde cholangiopancreatography (ERCP). More accurate stratification may improve patient selection for ERCP and allow use of lower-risk modalities. METHODS: A machine learning model was developed using patient information from two published cohort studies that evaluated performance of guidelines in predicting choledocholithiasis. Prediction models were developed using the gradient boosting model (GBM) machine learning method. GBM performance was evaluated using 10-fold cross-validation and area under the receiver operating characteristic curve (AUC). Important predictors of choledocholithiasis were identified based on relative importance in the GBM. RESULTS: 1378 patients (mean age 43.3 years; 61.2% female) were included in the GBM and 59.4% had choledocholithiasis. Eight variables were identified as predictors of choledocholithiasis. The GBM had accuracy of 71.5% (SD 2.5%) (AUC 0.79 [SD 0.06]) and performed better than the 2019 American Society for Gastrointestinal Endoscopy (ASGE) guidelines (accuracy 62.4% [SD 2.6%]; AUC 0.63 [SD 0.03]) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines (accuracy 62.8% [SD 2.6%]; AUC 0.67 [SD 0.02]). The GBM correctly categorized 22% of patients directed to unnecessary ERCP by ASGE guidelines, and appropriately recommended as the next management step 48% of ERCPs incorrectly rejected by ESGE guidelines. CONCLUSIONS: A machine learning-based tool was created, providing real-time, personalized, objective probability of choledocholithiasis and ERCP recommendations. This more accurately directed ERCP use than existing ASGE and ESGE guidelines, and has the potential to reduce morbidity associated with ERCP or missed choledocholithiasis.

Dynamic liver test patterns do not predict bile duct stones.

BACKGROUND: Numerous models have been developed to predict choledocholithiasis. Recent work has shown that these algorithms perform suboptimally. Identification of clinical predictors with high positive and negative predictive value would minimize adverse events associated with unnecessary diagnostic endoscopic retrograde cholangiopancreatography (ERCP) while limiting the use of expensive tests including magnetic resonance cholangiopancreatography (MRCP) and endoscopic ultrasound (EUS) for indeterminate cases. METHODS: Consecutive unique inpatients who received their first ERCP at Los Angeles County Medical Center between January 2010 and November 2016 for suspected bile duct stones were reviewed. The primary outcome was the proportion of patients with specific combinations of liver enzyme patterns, transabdominal ultrasound, and clinical features who had stones confirmed on ERCP. As a secondary outcome, we assessed the performance of the American Society for Gastrointestinal Endoscopy (ASGE) risk stratification algorithm in our population. RESULTS: Of the 604 included patients, bile duct stones were confirmed in 410 (67.9%). Detailed assessment of liver enzyme patterns alone and in combination with clinical features and imaging findings yielded no highly predictive algorithms. Additionally, the ASGE high-risk criterion had a positive predictive value of only 68% for stones. For the 236 patients for whom MRCP was performed, this imaging modality was shown to have highest predictive value for the presence of stones on ERCP. CONCLUSION: Exhaustive exploration of various threshold values and dynamic patterns of liver enzymes combined with clinical features and basic imaging findings did not reveal an algorithm to accurately predict the presence of stones on ERCP. The ASGE risk stratification criteria were also insensitive in our population. Though desirable, there may be no "perfect" combination of clinical features that correlate with persistent bile duct stones. MRCP or EUS may be considered to avoid unnecessary ERCP and associated complications.

The Pancreatitis Activity Scoring System predicts clinical outcomes in acute pancreatitis: findings from a prospective cohort study.

OBJECTIVES: The Pancreatitis Activity Scoring System (PASS) has been derived by an international group of experts via a modified Delphi process. Our aim was to perform an external validation study to assess for concordance of the PASS score with high face validity clinical outcomes and determine specific meaningful thresholds to assist in application of this scoring system in a large prospectively ascertained cohort. METHODS: We analyzed data from a prospective cohort study of consecutive patients admitted to the Los Angeles County Hospital between March 2015 and March 2017. Patients were identified using an emergency department paging system and electronic alert system. Comprehensive characterization included substance use history, pancreatitis etiology, biochemical profile, and detailed clinical course. We calculated the PASS score at admission, discharge, and at 12 h increments during the hospitalization. We performed several analyses to assess the relationship between the PASS score and outcomes at various points during hospitalization as well as following discharge. Using multivariable logistic regression analysis, we assessed the relationship between admission PASS score and risk of severe pancreatitis. PASS score performance was compared to established systems used to predict severe pancreatitis. Additional inpatient outcomes assessed included local complications, length of stay, development of systemic inflammatory response syndrome (SIRS), and intensive care unit (ICU) admission. We also assessed whether the PASS score at discharge was associated with early readmission (re-hospitalization for pancreatitis symptoms and complications within 30 days of discharge). RESULTS: A total of 439 patients were enrolled, their mean age was 42 (±15) years, and 53% were male. Admission PASS score >140 was associated with moderately severe and severe pancreatitis (OR 3.5 [95% CI 2.0, 6.3]), ICU admission (OR 4.9 [2.5, 9.4]), local complications (3.0 [1.6, 5.7]), and development of SIRS (OR 2.9 [1.8, 4.5]) as well as prolongation of hospitalization by a mean of 1.5 (1.3-1.7) days. For the prediction of moderately severe/severe pancreatitis, the PASS score (AUC = 0.71) was comparable to the more established Ranson's (AUC = 0.63), Glasgow (AUC = 0.72), Panc3 (AUC = 0.57), and HAPS (AUC = 0.54) scoring systems. Discharge PASS score >60 was associated with early readmission (OR 5.0 [2.4, 10.7]). CONCLUSIONS: The PASS score is associated with important clinical outcomes in acute pancreatitis. The ability of the score to forecast important clinical events at different points in the disease course suggests that it is a valid measure of activity in patients with acute pancreatitis.

Native T1 and ECV of Noninfarcted Myocardium and Outcome in Patients With Coronary Artery Disease.

BACKGROUND: Coronary artery disease (CAD) remains the major cause of cardiac morbidity and mortality worldwide, despite the advances in treatment with coronary revascularization and modern antiremodeling therapy. Risk stratification in CAD patients is primarily based on left ventricular volumes, ejection fraction (LVEF), risk scores, and the presence and extent of late gadolinium enhancement (LGE). The prognostic role of T1 mapping in noninfarcted myocardium in CAD patients has not yet been determined. OBJECTIVES: This study sought to examine prognostic significance of native T1 mapping of noninfarcted myocardium in patients with CAD. METHODS: A prospective, observational, multicenter longitudinal study of consecutive patients undergoing routine cardiac magnetic resonance imaging with T1 mapping and LGE. The primary endpoint was all-cause mortality. Major adverse cardiocerebrovascular events (MACCE) (cardiac mortality, nonfatal acute coronary syndrome, stroke, and appropriate device discharge) are also reported. RESULTS: A total of 34 deaths and 71 MACCE (n = 665, males n = 424, median age [interquartile range] 57 [22] years; 64%; median follow-up period of 17 [11] months) were observed. Native T1 and extracellular volume were univariate predictors of outcome. Native T1 and LGE were stronger predictors of survival and MACCE compared with extracellular volume, LVEF, cardiac volumes, and clinical scores (p < 0.001). Native T1 of noninfarcted myocardium was the sole independent predictor of all-cause mortality (chi-square = 21.7; p < 0.001), which was accentuated in the absence of LGE or LVEF ≤35%. For MACCE, native T1 and LGE extent were joint independent predictors (chi-square = 25.6; p < 0.001). CONCLUSIONS: Characterization of noninfarcted myocardium by native T1 is an important predictor of outcome in CAD patients, over and above the traditional risk stratifiers. The current study's results provide a basis for a novel risk stratification model in CAD based on a complementary assessment of noninfarcted myocardium and post-infarction scar, by native T1 mapping and LGE, respectively.

Randomized trial of cholangioscopy-guided laser lithotripsy versus conventional therapy for large bile duct stones (with videos).

BACKGROUND AND AIMS: Bile duct stones >1 cm have a decreased incidence of successful endoscopic extraction and often require lithotripsy. Although previous guidelines suggested mechanical lithotripsy for large common bile duct stones, current guidelines suggest cholangioscopy-guided lithotripsy as an adjunct with or without balloon dilation or mechanical lithotripsy. However, no randomized trials have assessed the usefulness of this practice. METHODS: Patients with bile duct stones >1 cm in diameter were randomized in a 2:1 ratio to cholangioscopy-guided laser lithotripsy versus conventional therapy only. Conventional therapies such as mechanical lithotripsy or balloon dilation were also allowed in the laser lithotripsy group. Randomization was stratified by history of ERCP in the past 3 months. The primary outcome was endoscopic clearance of the bile duct stones. RESULTS: Endoscopic clearance was achieved in 39 (93%) of 42 patients treated with cholangioscopy-guided laser lithotripsy and 12 (67%) of 18 treated with conventional therapy only (P = .009). The 9 patients in whom ERCP was unsuccessful underwent surgical common duct exploration with stone removal. Mean procedure time was 120.7 ± 40.2 minutes for the cholangioscopy-guided laser lithotripsy group compared with 81.2 ± 49.3 minutes for the conventional therapy group (P = .0008). There was no significant difference in fluoroscopy time, number of procedures, or adverse events (cholangitis) (cholangioscopy, 2; conventional, 1) and post-ERCP pancreatitis (cholangioscopy, 2; conventional, 1). CONCLUSION: Cholangioscopy-guided laser lithotripsy increases the incidence of endoscopic clearance of large bile duct stones and decreases the need for surgery compared with conventional therapy alone. However, it is associated with longer procedure times. (Clinical trial registration number: NCT0175997.).

T1-Mapping and Outcome in Nonischemic Cardiomyopathy: All-Cause Mortality and Heart Failure.

OBJECTIVES: The study sought to examine prognostic relevance of T1 mapping parameters (based on a T1 mapping method) in nonischemic dilated cardiomyopathy (NIDCM) and compare them with conventional markers of adverse outcome. BACKGROUND: NIDCM is a recognized cause of poor clinical outcome. NIDCM is characterized by intrinsic myocardial remodeling due to complex pathophysiological processes affecting myocardium diffusely. Lack of accurate and noninvasive characterization of diffuse myocardial disease limits recognition of early cardiomyopathy and effective clinical management in NIDCM. Cardiac magnetic resonance (CMR) supports detection of diffuse myocardial disease by T1 mapping. METHODS: This is a prospective observational multicenter longitudinal study in 637 consecutive patients with dilated NIDCM (mean age 50 years [interquartile range: 37 to 76 years]; 395 males [62%]) undergoing CMR with T1 mapping and late gadolinium enhancement (LGE) at 1.5-T and 3.0-T. The primary endpoint was all-cause mortality. A composite of heart failure (HF) mortality and hospitalization was a secondary endpoint. RESULTS: During a median follow-up period of 22 months (interquartile range: 19 to 25 months), we observed a total of 28 deaths (22 cardiac) and 68 composite HF events. T1 mapping indices (native T1 and extracellular volume fraction), as well as the presence and extent of LGE, were predictive of all-cause mortality and HF endpoint (p < 0.001 for all). In multivariable analyses, native T1 was the sole independent predictor of all-cause and HF composite endpoints (hazard ratio: 1.1; 95% confidence interval: 1.06 to 1.15; hazard ratio: 1.1; 95% confidence interval: 1.05 to 1.1; p < 0.001 for both), followed by the models including the extent of LGE and right ventricular ejection fraction, respectively. CONCLUSIONS: Noninvasive measures of diffuse myocardial disease by T1 mapping are significantly predictive of all-cause mortality and HF events in NIDCM. We provide a basis for a novel algorithm of risk stratification in NIDCM using a complementary assessment of diffuse and regional disease by T1 mapping and LGE, respectively.

T1 Mapping in Discrimination of Hypertrophic Phenotypes: Hypertensive Heart Disease and Hypertrophic Cardiomyopathy: Findings From the International T1 Multicenter Cardiovascular Magnetic Resonance Study.

BACKGROUND: The differential diagnosis of left ventricular (LV) hypertrophy remains challenging in clinical practice, in particular, between hypertrophic cardiomyopathy (HCM) and increased LV wall thickness because of systemic hypertension. Diffuse myocardial disease is a characteristic feature in HCM, and an early manifestation of sarcomere-gene mutations in subexpressed family members (G+P- subjects). This study aimed to investigate whether detecting diffuse myocardial disease by T1 mapping can discriminate between HCM versus hypertensive heart disease as well as to detect genetically driven interstitial changes in the G+P- subjects. METHODS AND RESULTS: Patients with diagnoses of HCM or hypertension (HCM, n=95; hypertension, n=69) and G+P- subjects (n=23) underwent a clinical cardiovascular magnetic resonance protocol (3 tesla) for cardiac volumes, function, and scar imaging. T1 mapping was performed before and >20 minutes after administration of 0.2 mmol/kg of gadobutrol. Native T1 and extracellular volume fraction were significantly higher in HCM compared with patients with hypertension (P<0.0001), including in subgroup comparisons of HCM subjects without evidence of late gadolinium enhancement, as well as of hypertensive patients LV wall thickness of >15 mm (P<0.0001). Compared with controls, native T1 was significantly higher in G+P- subjects (P<0.0001) and 65% of G+P- subjects had a native T1 value >2 SD above the mean of the normal range. Native T1 was an independent discriminator between HCM and hypertension, over and above extracellular volume fraction, LV wall thickness and indexed LV mass. Native T1 was also useful in separating G+P- subjects from controls. CONCLUSIONS: Native T1 may be applied to discriminate between HCM and hypertensive heart disease and detect early changes in G+P- subjects.

Native T1 in discrimination of acute and convalescent stages in patients with clinical diagnosis of myocarditis: a proposed diagnostic algorithm using CMR.

OBJECTIVES: This study investigated whether T1 mapping by cardiac magnetic resonance (CMR) reflects the clinical evolution of disease in myocarditis and supports its diagnosis independently of the disease stages. BACKGROUND: Acute viral myocarditis is characterized by a range of intracellular changes due to viral replication and extracellular spill of debris within days of viral infection. Convalescence may be characterized by a chronic low-grade inflammation leading to ventricular remodelling, but also a complete resolution of myocardial changes. METHODS: Patients with clinical diagnosis of viral myocarditis (N = 165) underwent routine clinical CMR protocol (1.5- and 3.0-T) for assessment of cardiac function and structure, and tissue characterization with T2-weighted imaging and late gadolinium enhancement. T1 mapping was obtained in a mid-ventricular short-axis slice before and >20 min after administration of 0.2 mmol/kg of gadobutrol. RESULTS: Compared with control subjects (n = 40), T1 indexes were increased in patients with myocarditis. Patients with acute symptoms (n = 61) had higher values of T1 indexes compared with patients in clinical convalescence (n = 67). Native T1 is an independent discriminator between health and disease, as well as a discriminator between acute and convalescent stage of the disease. Native T1- was superior to T2-weighted imaging and late gadolinium enhancement with high diagnostic accuracy and positive and negative predictive values. Using pre-defined cutoff values for normal ranges, we demonstrated that acute myocarditis can be independently identified by native T1 of >5 SD above the mean of normal range, whereas convalescence is best defined by either abnormal native T1 (>2 SD) or presence of late gadolinium enhancement. We prospectively tested a new diagnostic algorithm in an independent dataset of patients with clinical diagnosis of myocarditis and achieved similar diagnostic performance. CONCLUSIONS: The new diagnostic algorithm using native T1 can reliably discriminate between health and disease and determine the clinical disease stage in patients with a clinical diagnosis of myocarditis.

Reference values for healthy human myocardium using a T1 mapping methodology: results from the International T1 Multicenter cardiovascular magnetic resonance study.

BACKGROUND: T1 mapping is a robust and highly reproducible application to quantify myocardial relaxation of longitudinal magnetisation. Available T1 mapping methods are presently site and vendor specific, with variable accuracy and precision of T1 values between the systems and sequences. We assessed the transferability of a T1 mapping method and determined the reference values of healthy human myocardium in a multicenter setting. METHODS: Healthy subjects (n=102; mean age 41 years (range 17-83), male, n=53 (52%)), with no previous medical history, and normotensive low risk subjects (n=113) referred for clinical cardiovascular magnetic resonance (CMR) were examined. Further inclusion criteria for all were absence of regular medication and subsequently normal findings of routine CMR. All subjects underwent T1 mapping using a uniform imaging set-up (modified Look- Locker inversion recovery, MOLLI, using scheme 3(3)3(3)5)) on 1.5 Tesla (T) and 3 T Philips scanners. Native T1-maps were acquired in a single midventricular short axis slice and repeated 20 minutes following gadobutrol. Reference values were obtained for native T1 and gadolinium-based partition coefficients, λ and extracellular volume fraction (ECV) in a core lab using standardized postprocessing. RESULTS: In healthy controls, mean native T1 values were 950±21 msec at 1.5 T and 1052±23 at 3 T. λ and ECV values were 0.44±0.06 and 0.25±0.04 at 1.5 T, and 0.44±0.07 and 0.26±0.04 at 3 T, respectively. There were no significant differences between healthy controls and low risk subjects in routine CMR parameters and T1 values. The entire cohort showed no correlation between age, gender and native T1. Cross-center comparisons of mean values showed no significant difference for any of the T1 indices at any field strength. There were considerable regional differences in segmental T1 values. λ and ECV were found to be dose dependent. There was excellent inter- and intraobserver reproducibility for measurement of native septal T1. CONCLUSION: We show transferability for a unifying T1 mapping methodology in a multicenter setting. We provide reference ranges for T1 values in healthy human myocardium, which can be applied across participating sites.

Comparative performance of PET tracers in biochemical recurrence of prostate cancer: a critical analysis of literature.

Positron emission tomography (PET) with a number of tracers targeted to particular biological features of cancer has been explored for the imaging evaluation of patients with biochemical recurrence of prostate cancer after curative primary treatment. However, these reports are often heterogeneous in study design, patient cohorts, standards of reference for the imaging findings, data analysis, and data reporting. The aim of our study was to address these limitations by extracting and re-analyzing the PET detection data only from studies that satisfied pre-defined sets of patient selection criteria and verification standards. Our investigation analyzed the effects of 5 tracers ((18)F-fluorodeoxyglucose (FDG), (11)C-acetate (ACET), (11)C- or (18)F-choline (CHOL), anti-1-amino-3-(18)F-fluorocyclobutane-1-carboxylic acid (FACBC), and radiolabeled ligand targeted to prostate-specific membrane antigen (PSMA)), 2 treatment types (radical prostatectomy and radiation therapy), and whether the detected disease was local or metastatic, including lesion type (bone, lymph node, soft tissue). FDG exhibited the lowest detection rate for any suspected disease. ACET tended to be advantageous over CHOL in detecting local recurrence and lymph node lesions, even though the difference was not statistically significant. FACBC had greater likelihood of detecting local recurrence, when compared to CHOL, though this difference was not statistically significant. PSMA tended to show a higher proportion of patients with suspected disease compared to the other four tracers. Patients treated with radiation therapy had greater odds of displaying local recurrence on PET than those treated with radical prostatectomy. We also provide suggestions for future investigations that facilitate communication and the impact of the findings.

Accuracy and clinical outcomes of computed tomography coronary angiography in the presence of a high coronary calcium score.

BACKGROUND: A high coronary calcium burden may adversely affect image quality of CT coronary angiography (CTCA). The ability to rule out clinically significant disease in this setting is uncertain. METHODS: We examined CTCA findings in patients with a calcium score of >600. Utilising a search of death notices, structured patient interview and medical records, downstream investigations, cardiovascular events, revascularisation and mortality were recorded. RESULTS: Sixty patients with a calcium score >600 had CTCA performed on the same day. Coronary disease findings were: mild 28%, moderate 33%, severe 32% and non-diagnostic 7%. During a median 1.75-year follow-up, 31 (53%) of patients underwent further assessment for coronary disease, eight patients (13%) underwent revascularisation and there were two non-cardiovascular and one cardiovascular deaths. No patient with mild or moderate disease at CTCA had subsequently demonstrated ischaemia, was deemed to require PCI or suffered cardiac mortality. The negative predictive value of CTCA for subsequent PCI and all-cause mortality was 97% (100% for cardiac mortality only). The positive predictive value of CTCA for revascularisation or CV death was 42%. CONCLUSION: In patients with an elevated coronary calcium score, a negative CTCA implies an excellent short-term outcome and appears to exclude clinically significant coronary disease.

Four-dimensional image processing of myocardial CT perfusion for improved image quality and noise reduction.

BACKGROUND: Image noise and multiple sources of artifact may affect the accurate interpretation of myocardial CT perfusion (CTP) studies. Although artifact within the image is often time dependent, tissue characteristics remain unchanged irrespective of cardiac phase. OBJECTIVE: We assessed a new technique of 4-dimensional, spatiotemporal analysis, using redundant time domain information within additional phase acquisitions to reduce CTP image noise. METHODS: Four-dimensional analysis was assessed in a static phantom and in 10 CTP studies with invasive fractional flow reserve (FFR) correlation. For each voxel within the CTP study the distribution of local Hounsfield values was measured in both time and space with the use of a customized program within MATLAB software. These values were filtered to eliminate those likely to represent noise or rapidly changing beam hardening artifact. All CTP images were acquired within a single heartbeat with 320 detector-row CT. Image noise was quantified as the SD of voxel values within myocardial segments. Contrast was measured between normal and abnormal vascular territories as assessed by FFR. RESULTS: The mean image noise within the unprocessed CTP images was 30 HU (range, 23-42 HU). After 4-dimensional filtering the mean image noise was 22 HU (range, 15-29 HU). The mean reduction in image noise was 28% (P < 0.001). The mean contrast between normally perfused and ischemic segments was not significantly changed. The mean increase in contrast-to-noise ratio between ischemic territories and the myocardial average was 52% (P < 0.001). CONCLUSION: Four-dimensional analysis of CTP significantly reduces image noise and may assist in the assessment of myocardial perfusion studies.

Defining the mid-diastolic imaging period for cardiac CT - lessons from tissue Doppler echocardiography.

BACKGROUND: Aggressive dose reduction strategies for cardiac CT require the prospective selection of limited cardiac phases. At lower heart rates, the period of mid-diastole is typically selected for image acquisition. We aimed to identify the effect of heart rate on the optimal CT acquisition phase within the period of mid-diastole. METHODS: We utilized high temporal resolution tissue Doppler to precisely measure coronary motion within diastole. Tissue-Doppler waveforms of the myocardium corresponding to the location of the circumflex artery (100 patients) and mid-right coronary arteries (50 patients) and the duration and timing of coronary motion were measured. Using regression analysis an equation was derived for the timing of the period of minimal coronary motion within the RR interval. In a validation set of 50 clinical cardiac CT examinations, we assessed coronary motion artifact and the effect of using a mid-diastolic imaging target that was adjusted according to heart rate vs a fixed 75% phase target. RESULTS: Tissue Doppler analysis shows the period of minimal cardiac motion suitable for CT imaging decreases almost linearly as the RR interval decreases, becoming extinguished at an average heart rate of 91 bpm for the circumflex (LCX) and 78 bpm for the right coronary artery (RCA). The optimal imaging phase has a strong linear relationship with RR duration (R2 = 0.92 LCX, 0.89 RCA). The optimal phase predicted by regression analysis of the tissue-Doppler waveforms increases from 74% at a heart rate of 55 bpm to 77% at 75 bpm. In the clinical CT validation set, the optimal CT acquisition phase similarly occurred later with increasing heart rate. When the selected cardiac phase was adjusted according to heart rate the result was closer to the optimal phase than using a fixed 75% phase. While this effect was statistically significant (p < 0.01 RCA/LCx), the mean effect of heart-rate adjustment was minor relative to typical beat-to-beat variability and available precision of clinical phase selection. CONCLUSION: High temporal resolution imaging of coronary motion can be used to predict the optimal acquisition phase in cardiac CT. The optimal phase for cardiac CT imaging within mid-diastole increases with increasing heart rate although the magnitude of change is small.