Prediction of Severe Acute Pancreatitis Using a Decision Tree Model Based on the Revised Atlanta Classification of Acute Pancreatitis

A deep learning-powered diagnostic model for acute pancreatitis

BACKGROUND

Acute pancreatitis is one of the most common diseases requiring emergency surgery. Rapid and accurate recognition of acute pancreatitis can help improve clinical outcomes. This study aimed to develop a deep learning-powered diagnostic model for acute pancreatitis.

MATERIALS AND METHODS

In this investigation, we enrolled a cohort of 190 patients with acute pancreatitis who were admitted to Sichuan Provincial People's Hospital between January 2020 and December 2021. Abdominal computed tomography (CT) scans were obtained from both patients with acute pancreatitis and healthy individuals. Our model was constructed using two modules: (1) the acute pancreatitis classifier module; (2) the pancreatitis lesion segmentation module. Each model's performance was assessed based on precision, recall rate, F1-score, Area Under the Curve (AUC), loss rate, frequency-weighted accuracy (fwavacc), and Mean Intersection over Union (MIOU).

RESULTS

Upon admission, significant variations were observed between patients with mild and severe acute pancreatitis in inflammatory indexes, liver, and kidney function indicators, as well as coagulation parameters. The acute pancreatitis classifier module exhibited commendable diagnostic efficacy, showing an impressive AUC of 0.993 (95%CI: 0.978-0.999) in the test set (comprising healthy examination patients vs. those with acute pancreatitis, P < 0.001) and an AUC of 0.850 (95%CI: 0.790-0.898) in the external validation set (healthy examination patients vs. patients with acute pancreatitis, P < 0.001). Furthermore, the acute pancreatitis lesion segmentation module demonstrated exceptional performance in the validation set. For pancreas segmentation, peripancreatic inflammatory exudation, peripancreatic effusion, and peripancreatic abscess necrosis, the MIOU values were 86.02 (84.52, 87.20), 61.81 (56.25, 64.83), 57.73 (49.90, 68.23), and 66.36 (55.08, 72.12), respectively. These findings underscore the robustness and reliability of the developed models in accurately characterizing and assessing acute pancreatitis.

CONCLUSION

The diagnostic model for acute pancreatitis, driven by deep learning, exhibits excellent efficacy in accurately evaluating the severity of the condition.

TRIAL REGISTRATION

This is a retrospective study.

Diagnostic utility of HFLC% and IG% for acute Pancreatitis-A retrospective Case-Control study

Objective

To evaluate the occurrence, development and outcome value of hyperfluorescent lymphocyte percentage (HFLC%) and immature granulocyte percentage (IG%) for acute pancreatitis (AP).

Methods

The laboratory data collected from 1533 patients diagnosed with AP between August 2018 and August 2022 were retrospectively analyzed. The patients were classified into mild acute pancreatitis (MAP) and non-mild acute pancreatitis (Non-MAP) groups; non-MAP groups were additionally subgrouped based on HFLC% at day 7. White blood cells (WBC), HFLC%, and IG% were examined from day 1 (baseline) to day 14 post-admission using Sysmex XN Series Hematology Analyzers. C-reactive protein (CRP), serum amylase (AMY), and lipase (LPS) were detected by Beckman AU5800.

Results

A total of 623 patients were finally included in the study [MAP group (n = 358) and Non-MAP group (n = 265)]. WBC, IG%, and CRP were higher in the Non-MAP group from day 1 to day 12 (all P<0.05). The HFLC% was not statistically significant from day 1 to day 6; yet, it increased on day 6 and 7 in the Non-MAP group. We divided patients in the Non-MAP group with complete data(101 patients) into HFLC% ≥ 2.9 %(31 patients) and HFLC% < 2.9 %(70 patients) according to the threshold of 7th day HFLC%. WBC, HFLC%, IG%, and CRP effectively predicted the progression of MAP to Non-MAP (all P < 0.001). HFLC% was the most obvious value, followed by CRP and IG%. Combined with HFLC%, IG%,CRP and WBC in day7, the ROC analysis showed that the area under ROC curve of the combined indicators was the largest (AUC = 0.912, P < 0.001) and had higher sensitivity and specificity than single-item assessment of AP outcomes(P < 0.05). HFLC% < 2.9 %, IG% > 1.7 %, CRP >28.66 mg/L, and WBC >9.24 × 109/L indicated the possibility of AP disease aggravation. Also, HFLC% <2.9 % was directly associated with infection, SIRS, APPACHII grade, and ICU admission (all P < 0.05). In non-MAP there was a significant negative correlation between HFLC% and APACHE-II score (rs = -0.312, P = 0.023).

Conclusion

HFLC% <2.9 % on 7th day was directly indicated more infection, systemic inflammatory response syndrome(SIRS), higher APPACH II grade and ICU admission. HFLC% may be an independent laboratory marker for prognosis in AP. Combining HFLC% with IG%, CRP, and WBC helps evaluate AP patients' disease development and outcome.

Predictive value of machine learning for the severity of acute pancreatitis: A systematic review and meta-analysis

Background

Predicting the severity of acute pancreatitis (AP) early poses a challenge in clinical practice. While there are well-established clinical scoring tools, their actual predictive performance remains uncertain. Various studies have explored the application of machine-learning methods for early AP prediction. However, a more comprehensive evidence-based assessment is needed to determine their predictive accuracy. Hence, this systematic review and meta-analysis aimed to evaluate the predictive accuracy of machine learning in assessing the severity of AP.

Methods

PubMed, EMBASE, Cochrane Library, and Web of Science were systematically searched until December 5, 2023. The risk of bias in eligible studies was assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Subgroup analyses, based on different machine learning types, were performed. Additionally, the predictive accuracy of mainstream scoring tools was summarized.

Results

This systematic review ultimately included 33 original studies. The pooled c-index in both the training and validation sets was 0.87 (95 % CI: 0.84-0.89) and 0.88 (95 % CI: 0.86-0.90), respectively. The sensitivity in the training set was 0.81 (95 % CI: 0.77-0.84), and in the validation set, it was 0.79 (95 % CI: 0.71-0.85). The specificity in the training set was 0.84 (95 % CI: 0.78-0.89), and in the validation set, it was 0.90 (95 % CI: 0.86-0.93). The primary model incorporated was logistic regression; however, its predictive accuracy was found to be inferior to that of neural networks, random forests, and xgboost. The pooled c-index of the APACHE II, BISAP, and Ranson were 0.74 (95 % CI: 0.68-0.80), 0.77 (95 % CI: 0.70-0.85), and 0.74 (95 % CI: 0.68-0.79), respectively.

Conclusions

Machine learning demonstrates excellent accuracy in predicting the severity of AP, providing a reference for updating or developing a straightforward clinical prediction tool.

Association between serum creatinine to albumin ratio and short- and long-term all-cause mortality in patients with acute pancreatitis admitted to the intensive care unit: a retrospective analysis based on the MIMIC-IV database

Background

Serum creatinine (Cr) and albumin (Alb) are important predictors of mortality in individuals with various diseases, including acute pancreatitis (AP). However, most previous studies have only examined the relationship between single Cr or Alb levels and the prognosis of patients with AP. To our knowledge, the association between short- and long-term all-cause mortality in patients with AP and the blood creatinine to albumin ratio (CAR) has not been investigated. Therefore, this study aimed to evaluate the short- and long-term relationships between CAR and all-cause mortality in patients with AP.

Methods

We conducted a retrospective study utilizing data from the Medical Information Market for Intensive Care (MIMIC-IV) database. The study involved analyzing various mortality variables and obtaining CAR values at the time of admission. The X-tile software was used to determine the optimal threshold for the CAR. Kaplan-Meier (K-M) survival curves and multivariate Cox proportional hazards regression models were used to assess the relationship between CAR and both short- and long-term all-cause mortality. The predictive power, sensitivity, specificity, and area under the curve (AUC) of CAR for short- and long-term mortality in patients with AP after hospital admission were investigated using Receiver Operating Characteristic analysis. Additionally, subgroup analyses were conducted.

Results

A total of 520 participants were included in this study. The CAR ideal threshold, determined by X-tile software, was 0.446. The Cox proportional hazards model revealed an independent association between CAR≥0.446 and all-cause mortality at 7-day (d), 14-d, 21-d, 28-d, 90-d, and 1-year (y) before and after adjustment for confounders. K-M survival curves showed that patients with CAR≥0.446 had lower survival rates at 7-d, 14-d, 21-d, 28-d, 90-d, and 1-y. Additionally, CAR demonstrated superior performance, with higher AUC values than Cr, Alb, serum total calcium, Glasgow Coma Scale, Systemic Inflammatory Response Syndrome score, and Sepsis-related Organ Failure Assessment score at 7-d, 14-d, 21-d, 28-d, 90-d, and 1-y intervals. Subgroup analyses showed that CAR did not interact with a majority of subgroups.

Conclusion

The CAR can serve as an independent predictor for short- and long-term all-cause mortality in patients with AP. This study enhances our understanding of the association between serum-based biomarkers and the prognosis of patients with AP.

Real-time prediction of organ failures in patients with acute pancreatitis using longitudinal irregular data

It is extremely important to identify patients with acute pancreatitis who are at high risk for developing persistent organ failures early in the course of the disease. Due to the irregularity of longitudinal data and the poor interpretability of complex models, many models used to identify acute pancreatitis patients with a high risk of organ failure tended to rely on simple statistical models and limited their application to the early stages of patient admission. With the success of recurrent neural networks in modeling longitudinal medical data and the development of interpretable algorithms, these problems can be well addressed. In this study, we developed a novel model named Multi-task and Time-aware Gated Recurrent Unit RNN (MT-GRU) to directly predict organ failure in patients with acute pancreatitis based on irregular medical EMR data. Our proposed end-to-end multi-task model achieved significantly better performance compared to two-stage models. In addition, our model not only provided an accurate early warning of organ failure for patients throughout their hospital stay, but also demonstrated individual and population-level important variables, allowing physicians to understand the scientific basis of the model for decision-making. By providing early warning of the risk of organ failure, our proposed model is expected to assist physicians in improving outcomes for patients with acute pancreatitis.

Volume, but Not the Location of Necrosis, Is Associated with Worse Outcomes in Acute Pancreatitis: A Prospective Study

Background and Objectives: The course and clinical outcomes of acute pancreatitis (AP) are highly variable. Up to 20% of patients develop pancreatic necrosis. Extent and location of it might affect the clinical course and management. The aim was to determine the clinical relevance of the extent and location of pancreatic necrosis in patients with AP. Materials and Methods: A cohort of patients with necrotizing AP was collected from 2012 to 2018 at the Hospital of Lithuanian University of Health Sciences. Patients were allocated to subgroups according to the location (entire pancreas, left and right sides of pancreas) and extent (<30%, 30−50%, >50%) of pancreatic necrosis. Patients were reviewed for demographic features, number of performed surgical interventions, local and systemic complications, hospital stay and mortality rate. All contrast enhanced computed tomography (CECT) scans were evaluated by at least two experienced abdominal radiologists. All patients were treated according to the standard treatment protocol based on current international guidelines. Results: The study included 83 patients (75.9% males (n = 63)) with a mean age of 53 ± 1.7. The volume of pancreatic necrosis exceeded 50% in half of the patients (n = 42, 51%). Positive blood culture (n = 14 (87.5%)), multiple organ dysfunction syndrome (n = 17 (73.9%)) and incidences of respiratory failure (n = 19 (73.1%)) were significantly more often diagnosed in patients with pancreatic necrosis exceeding 50% (p < 0.05). Patients with >50% of necrosis were significantly (p < 0.05) more often diagnosed with moderately severe (n = 24 (41.4%)) and severe (n = 18 (72%)) AP. The number of surgical interventions (n = 18 (72%)) and ultrasound-guided interventions (n = 26 (65%)) was also significantly higher. In patients with whole-pancreas necrosis, incidence of renal insufficiency (n = 11 (64.7%)) and infected pancreatic necrosis (n = 19 (57.6%)) was significantly higher (p < 0.05). Conclusions: The clinical course and outcome were worse in the case of pancreatic necrosis exceeding 50%, rendering the need for longer and more complex treatment.

Correlation Between Severity of Illness and Levels of Free Triiodothyronine, Interleukin-6, and Interleukin-10 in Patients with Acute Pancreatitis

BACKGROUND Acute pancreatitis (AP) is a common acute abdominal disease. Rapid evaluation of the severity is important for AP prognosis and treatment. Free triiodothyronine (fT3) level is associated with the prognosis of AP patients. This study aimed to investigate the fT3 level in patients with acute pancreatitis; early warning signs of inflammation, including interleukin-6 (IL-6) and interleukin-10 (IL-10); and the correlation of fT3 level with illness severity. MATERIAL AND METHODS Enrolled AP patients (N=312) were divided into an SAP group (N=92) and a non-SAP group (N=220) according to the Revision of Atlanta classification. Blood or tissue samples and baseline clinical characteristics were recorded. The t test and chi-square test were used to evaluate differences between the 2 groups. Multivariate logistic regression analysis and receiver operating characteristic (ROC) curves were used to investigate protective factors. One-way repeated measures analysis of variance was used to evaluate the prognosis of SAP patients. RESULTS In our study, compared with APACHII score (AUC 0.829 [95% CIs 0.769-0.889]) and Ranson score (AUC 0.629 [95% CIs 0.542-0.715]), our predictive model (AUC 0.918 [95% CIs 0.875-0.961]) showed better prognostic performance in predicting poor patient outcomes. In the SAP group, changes in fT3 level were significantly associated with prognosis (P<0.05). CONCLUSIONS The predictive model can improve the diagnostic accuracy and prediction of the severity of disease. FT3 level could be used as an independent risk factor to predict the mortality of SAP patients.

Serum Lactate Dehydrogenase Is a Sensitive Predictor of Systemic Complications of Acute Pancreatitis

BACKGROUND

Acute pancreatitis (AP) is a common and potentially life-threatening inflammatory disease that can cause various complications, including systemic inflammatory response syndrome (SIRS), pleural effusion, ascitic fluid, myocardial infarction, and acute kidney injury (AKI). However, there is still a lack of rapid and effective indicators to assess the disease. The aim of this study was to investigate the associations of high serum lactate dehydrogenase (LDH) levels with AP severity and systemic complications.

METHODS

AP patients treated from July 2014 to December 2020 were retrospectively enrolled. They were divided into elevated (n = 93) and normal (n = 143) LDH groups. Their demographic data, clinical data, hospital duration, and hospital expenses were analyzed. Linear and binary logistic regression analyses were used to determine whether elevated LDH is a risk factor for AP severity and complications after adjusting for confounders.

RESULTS

There were significant differences in AP severity scores (Ranson, MODS, BISAP, APACHE II, and CTSI), hospital duration, hospital expenses, and the incidences of complications (SIRS, pleural effusion, ascitic fluid, myocardial infarction, and AKI) between the elevated and normal LDH groups. After adjusting for confounders, elevated LDH was associated with AP severity scores and hospital duration and expenses (based on linear regression analyses) and was a risk factor for the occurrence of AP complications and interventions, that is, diuretic and vasoactive agent use (based on binary logistic regression analyses).

CONCLUSIONS

Elevated LDH is associated with high AP severity scores and high incidences of complications (SIRS, pleural effusion, ascitic fluid, myocardial infarction, and AKI).

Machine learning predictive models for acute pancreatitis: A systematic review

INTRODUCTION

Acute pancreatitis (AP) is a common clinical pancreatic disease. Patients with different severity levels have different clinical outcomes. With the advantages of algorithms, machine learning (ML) has gradually emerged in the field of disease prediction, assisting doctors in decision-making.

METHODS

A systematic review was conducted using the PubMed, Web of Science, Scopus, and Embase databases, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Publication time was limited from inception to 29 May 2021. Studies that have used ML to establish predictive tools for AP were eligible for inclusion. Quality assessment of the included studies was conducted in accordance with the IJMEDI checklist.

RESULTS

In this systematic review, 24 of 2,913 articles, with a total of 8,327 patients and 47 models, were included. The studies could be divided into five categories: 10 studies (42%) reported severity prediction; 10 studies (42%), complication prediction; 3 studies (13%), mortality prediction; 2 studies (8%), recurrence prediction; and 2 studies (8%), surgery timing prediction. ML showed great accuracy in several prediction tasks. However, most of the included studies were retrospective in nature, conducted at a single centre, based on database data, and lacked external validation. According to the IJMEDI checklist and our scoring criteria, two studies were considered to be of high quality. Most studies had an obvious bias in the quality of data preparation, validation, and deployment dimensions.

CONCLUSION

In the prediction tasks for AP, ML has shown great potential in assisting decision-making. However, the existing studies still have some deficiencies in the process of model construction. Future studies need to optimize the deficiencies and further evaluate the comparability of the ML systems and model performance, so as to consequently develop high-quality ML-based models that can be used in clinical practice.

Predictors of prolonged hospitalization in patients with mild and moderate pancreatitis presenting to the emergency department

BACKGROUND

Patients with acute pancreatitis often present to the emergency department (ED) and usually require hospital admission. The aim of this study was to determine predictors of prolonged hospital stays in patients with mild forms of acute pancreatitis.

METHODS

This retrospective cohort study was conducted in patients diagnosed in the ED with mild and moderate acute pancreatitis according to the revised Atlanta classification. Patients with available data between 2007 and 2017 were included and were divided based on their admission duration. Eight days or more was considered a long hospitalization. A multivariate logistic regression model was constructed to determine the independent predictors of prolonged hospital stays.

RESULTS

Of the 485 evaluated patients, 335 were included in the analysis. Baseline characteristics, determined by vital signs and laboratory parameters, were similar between the short and long hospitalization groups. However, the long hospitalization group received more intravenous crystalloid in the ED, and this group used more diuretics and more angiotensin-converting enzyme inhibitor and angiotensin-receptor blocker (ACEI/ARB) drugs than the other group. Diuretic use was present in 8 patients (4.8%) in the short hospital length of stay group and in 26 patients (15.3%) in the long hospital length of stay group. Age, gender, ACEI/ARB use, diuretic use, total amount of intravenous crystalloid administered in the ED, aPTT, BUN, creatinine, and presence of biliary pathology were included in the multivariate analysis. Regarding the final analysis, diuretic use was an independent predictor of prolonged hospitalization (odds ratio 2.89, 95% confidence interval 1.05-8.00, p = 0.041).

CONCLUSION

Diuretic use is an independent predictor of long hospitalization in patients with mild and moderate pancreatitis. Drugs affecting total volume balance may prolong hospital stays in these patients.

A contrast-enhanced-CT-based classification tree model for classifying malignancy of solid lung tumors in a Chinese clinical population

Background

To develop and validate a contrast-enhanced CT based classification tree model for classifying solid lung tumors in clinical patients into malignant or benign.

Methods

Between January 2015 and October 2017, 827 pathologically confirmed solid lung tumors (487 malignant, 340 benign; median size, 27.0 mm, IQR 18.0–39.0 mm) from 827 patients from a dedicated Chinese cancer hospital were identified. Nodules were divided randomly into two groups, a training group (575 cases) and a testing group (252 cases). CT characteristics were collected by two radiologists, and analyzed using a classification and regression tree (CART) model. For validation, we used the decision analysis threshold to evaluate the classification performance of the CART model and radiologist’s diagnosis (benign; malignant) in the testing group.

Results

Three out of 19 characteristics [margin (smooth; slightly lobulated/lobulated/spiculated), and shape (round/oval; irregular), subjective enhancement (no/uniform enhancement; heterogeneous enhancement)] were automatically generated by the CART model for classifying solid lung tumors. The sensitivity, specificity, PPV, NPV, and diagnostic accuracy of the CART model is 98.5%, 58.1%, 80.6%, 98.6%, 79.8%, and 90.4%, 54.7%, 82.4% 98.5%, 74.2% for the radiologist’s diagnosis by using three-threshold decision analysis.

Conclusions

Tumor margin and shape, and subjective tumor enhancement were the most important CT characteristics in the CART model for classifying solid lung tumors as malignant. The CART model had higher discriminatory power than radiologist’s diagnosis. The CART model could help radiologists making recommendations regarding follow-up or surgery in clinical patients with a solid lung tumor.

THE ACCEPTANCE OF CHANGES IN THE MANAGEMENT OF PATIENTS WITH ACUTE PANCREATITIS AFTER THE REVISED ATLANTA CLASSIFICATION

BACKGROUND

New recommendations for the management of patients with acute pancreatitis were set after the Atlanta Classification was revised in 2012.

OBJECTIVE

The aim of the present systematic review is to assess whether these recommendations have already been accepted and implemented in daily medical practices.

METHODS

A systematic literature review was carried out in studies conducted with humans and published in English and Portuguese language from 10/25/2012 to 11/30/2018. The search was conducted in databases such as PubMed/Medline, Cochrane and SciELO, based on the following descriptors/Boolean operator: "Acute pancreatitis" AND "Atlanta". Only Randomized Clinical Trials comprising some recommendations released after the revised Atlanta Classification in 2012 were included in the study.

RESULTS

Eighty-nine studies were selected and considered valid after inclusion, exclusion and qualitative evaluation criteria application. These studies were stratified as to whether, or not, they applied the recommendations suggested after the Atlanta Classification revision. Based on the results, 68.5% of the studies applied the recommendations, with emphasis on the application of severity classification (mild, moderately severe, severe); 16.4% of them were North-American and 14.7% were Chinese. The remaining 31.5% just focused on comparing or validating the severity classification.

CONCLUSION

Few studies have disclosed any form of acceptance or practice of these recommendations, despite the US and Chinese efforts. The lack of incorporation of these recommendations didn't enable harnessing the benefits of their application in the clinical practice (particularly the improvement of the communication among health professionals and directly association with the worst prognoses); thus, it is necessary mobilizing the international medical community in order to change this scenario.

Machine learning based on clinical characteristics and chest CT quantitative measurements for prediction of adverse clinical outcomes in hospitalized patients with COVID-19

Objectives

To develop and validate a machine learning model for the prediction of adverse outcomes in hospitalized patients with COVID-19.

Methods

We included 424 patients with non-severe COVID-19 on admission from January 17, 2020, to February 17, 2020, in the primary cohort of this retrospective multicenter study. The extent of lung involvement was quantified on chest CT images by a deep learning–based framework. The composite endpoint was the occurrence of severe or critical COVID-19 or death during hospitalization. The optimal machine learning classifier and feature subset were selected for model construction. The performance was further tested in an external validation cohort consisting of 98 patients.

Results

There was no significant difference in the prevalence of adverse outcomes (8.7% vs. 8.2%, p = 0.858) between the primary and validation cohorts. The machine learning method extreme gradient boosting (XGBoost) and optimal feature subset including lactic dehydrogenase (LDH), presence of comorbidity, CT lesion ratio (lesion%), and hypersensitive cardiac troponin I (hs-cTnI) were selected for model construction. The XGBoost classifier based on the optimal feature subset performed well for the prediction of developing adverse outcomes in the primary and validation cohorts, with AUCs of 0.959 (95% confidence interval [CI]: 0.936–0.976) and 0.953 (95% CI: 0.891–0.986), respectively. Furthermore, the XGBoost classifier also showed clinical usefulness.

Conclusions

We presented a machine learning model that could be effectively used as a predictor of adverse outcomes in hospitalized patients with COVID-19, opening up the possibility for patient stratification and treatment allocation.

Key Points

• Developing an individually prognostic model for COVID-19 has the potential to allow efficient allocation of medical resources.

• We proposed a deep learning–based framework for accurate lung involvement quantification on chest CT images.

• Machine learning based on clinical and CT variables can facilitate the prediction of adverse outcomes of COVID-19.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-07957-z.

New markers in predicting the severity of acute pancreatitis in the emergency department: Immature granulocyte count and percentage

Background:

Acute pancreatitis (AP) may vary in severity, from mild, self-limiting pancreatic inflammation to rapidly progressive life-threatening clinical course. If the severity of AP can be predicted early and treated quickly, it may lead to a decrease in morbidity and mortality rates. There?fore, we aimed to investigate the clinical utility of immature granulocyte count (IGC) and IGC percentage (IG%) in showing the severity of AP in this study.

Methods:

Two hundred and twenty-seven patients who were admitted to our emergency department and diagnosed with AP between March 1 and September 30, 2019, were included in the study. The patients were divided into two groups as mild and severe AP (MAP and SAP) according to the severity of the disease. Demographic characteristics of the patients, disease etiology, disease severity, and inflammation markers [white blood cell count (WBC), IGC, IG%, neutrophil–lymphocyte ratio (NLR), and C-reactive protein (CRP)] were recorded. Differences between the groups were statistically analyzed.

Results:

Of the patients included in the study, 183 (80.7%) were in the MAP group and 44 (19.3%) were in the SAP group. The mean WBC, NLR, CRP, IGC, and IG% levels were significantly higher in the SAP group compared to the MAP group. The power of IGC and IG% in predicting SAP was higher than other inflammation markers (WBC, NLR, and CRP) [(AUC for IGC: 0.902; sensitivity: 78.2%; specificity: 92.8%); (AUC for IG%: 0.843; sensitivity: 72.7%; specificity: 84.6%)].

Conclusion:

IGC and IG% show the severity of AP more effectively than WBC, NLR, and CRP, which are traditional inflammation markers.

Prognostic value of the creatinine-albumin ratio in acute pancreatitis debridement

Background

Increases in the levels of serum C-reactive protein (CRP) and creatinine (Cr) and decreases in those of albumin (Alb) are commonly observed in acute pancreatitis (AP). We aimed to evaluate the efficacy of the Cr/Alb and CRP/Alb ratios in the prediction of surgical treatment effect in AP patients.

Methods

This study retrospectively analyzed clinical data obtained from 140 AP patients who underwent debridement from January 2008 to November 2018 in Shanghai Ruijin Hospital. The Cr/Alb and CRP/Alb ratios at admission and before surgery were assessed in the analysis of clinical statistics, prediction of prognoses, and logistic regression analysis.

Results

The admission Cr/Alb had the best predictive value of the four ratios. This value was significantly higher in patients with re-operation and those who died (P < 0.05) and was correlated with the Acute Physiology and Chronic Health Evaluation (APACHE II) score, admission CRP/Alb, preoperative Cr/Alb, and post-operative complications. The admission Cr/Alb could predict the risk of AP-related re-operation and mortality with sensitivities, specificities and areas under the curve of 86.3%, 61.7% and 0.824, and 73.4%, 81.3% and 0.794, respectively. At a cut-off value of 3.43, admission Cr/Alb values were indicative of a worse clinical state, including impaired laboratory test values, APACHE II scores, rates of post-operative complications and re-operation, and mortality (P < 0.05). In the logistic regression analysis, admission Cr/Alb values were independently related to the APACHE II score, post-operative renal failure, and mortality.

Conclusion

Cr/Alb is a novel but promising, easy-to-measure, reproducible, non-invasive prognostic score for the prediction of the effect of debridement in AP patients.

The early predictive value of routine laboratory tests on the severity of acute pancreatitis patients in pregnancy: a retrospective study

Acute pancreatitis in pregnancy (APIP) varies in severity from a self-limiting mild condition to a severe life-threatening condition, and its severity is significantly correlated with higher risks of maternal and foetal death. This study evaluated the early predictive value of routine laboratory tests on the severity of APIP patients. We enrolled 100 patients with APIP in West China Hospital. Initial routine laboratory tests, including the biochemistry and hematologic tests were collected within 48 hours after the onset of APIP. For predicting SAP in AP, LDH had the highest specificity of 0.879. RDW was a suitable predictive marker as it had the sensitivity of 0.882. Lower levels of triglycerides (<4.72 mmol/L) predicted mild AP of APIP, with an area under the curve (AUC) of 0.724, and a negative predictive value of 0.80. Furthermore, a risk score was calculated based on white blood cells, neutrophils, RDW, LMR and LDH, as an independent marker (adjusted odds ratio = 3.013, 95% CI 1.893 to 4.797, P < 0.001), with the highest AUC of 0.906, a sensitivity of 0.875 and a specificity of 0.828. In conclusion, the risk score we recommended was the powerful marker to aid in the early prediction of the severity of APIP patients.

Etiology and mortality in severe acute pancreatitis: A multicenter study in Japan

BACKGROUND/OBJECTIVES

Severe acute pancreatitis (SAP) has a high mortality rate despite ongoing attempts to improve prognosis through a various therapeutic modalities. This study aimed to delineate etiology-based routes that may guide clinical decisions for the treatment of SAP.

METHODS

Using data from a recent retrospective multicenter study in Japan, we analyzed the association between clinical outcomes, mainly in-hospital mortality and pancreatic infection, and various etiologies while considering confounding factors. We performed additional multivariate analyses and built decision tree models.

RESULTS

The 1097 participating patients were classified into the following groups by etiology: alcohol (n = 436, 39.7%); cholelithiasis (n = 230, 21.0%); idiopathic (n = 227, 20.7%); and others (n = 204, 18.6%). Mortality at hospital discharge was 8.4%, 12.2%, 16.7%, and 16.2% in the alcohol, cholelithiasis, idiopathic, and others groups, respectively. According to multivariable analysis, early enteral nutrition (EN) was significantly associated with reduced in-hospital mortality only in the cholelithiasis group. However, there was a consistent association between age and the need for mechanical ventilation and increased mortality, regardless of etiology. Our decision tree models presented different contributing factors depending on the etiology and patient background. Interaction analysis showed that EN and the use of prophylactic antibiotics may influence these results differently according to etiology.

CONCLUSIONS

No study has yet used comprehensive models to investigate etiology-related prognostic factors for SAP; our results can, therefore, be used as a reference for improving clinical decisions.

A Subsolid Nodules Imaging Reporting System (SSN-IRS) for Classifying 3 Subtypes of Pulmonary Adenocarcinoma

OBJECTIVES

To develop an imaging reporting system for the classification of 3 adenocarcinoma subtypes of computed tomography (CT)-detected subsolid pulmonary nodules (SSNs) in clinical patients.

METHODS

Between November 2011 and October 2017, 437 pathologically confirmed SSNs were retrospectively identified. SSNs were randomly divided 2:1 into a training group (291 cases) and a testing group (146 cases). CT-imaging characteristics were analyzed using multinomial univariable and multivariable logistic regression analysis to identify discriminating factors for the 3 adenocarcinoma subtypes (pre-invasive lesions, minimally invasive adenocarcinoma, and invasive adenocarcinoma). These factors were used to develop a classification and regression tree model. Finally, an SSN Imaging Reporting System (SSN-IRS) was constructed based on the optimized classification model. For validation, the classification performance was evaluated in the testing group.

RESULTS

Of the CT-derived characteristics of SSNs, qualitative density (nonsolid or part-solid), core (non-core or core), semantic features (pleural indentation, vacuole sign, vascular invasion), and diameter of solid component (≤6 mm or >6 mm), were the most important factors for the SSN-IRS. The total sensitivity, specificity, and diagnostic accuracy of the SSN-IRS was 89.0% (95% confidence interval [CI], 84.8%-92.4%), 74.6% (95% CI, 70.8%-78.1%), and 79.4% (95% CI, 76.5%-82.0%) in the training group and 84.9% (95% CI, 78.1%-90.3%), 68.5% (95% CI, 62.8%-73.8%), and 74.0% (95% CI, 69.6%-78.0%) in the testing group, respectively.

CONCLUSIONS

The SSN-IRS can classify 3 adenocarcinoma subtypes using CT-based characteristics of subsolid pulmonary nodules. This classification tool can help clinicians to make follow-up recommendations or decisions for surgery in clinical patients with SSNs.

C-reactive protein to albumin ratio is a key indicator in a predictive model for anastomosis leakage after esophagectomy: Application of classification and regression tree analysis

Background

Anastomotic leakage (AL), a serious complication after esophagectomy, might impair patient quality of life, prolong hospital stay, and even lead to surgery‐related death. The aim of this study was to show a novel decision model based on classification and regression tree (CART) analysis for the prediction of postoperative AL among patients who have undergone esophagectomy.

Methods

A total of 450 patients (training set: 356; test set: 94) with perioperative information were included. A decision tree model was established to identify the predictors of AL in the training set, which was validated in the test set. A receiver operating characteristic curve was also created to illustrate the diagnostic ability of the decision model.

Results

A total of 12.2% (55/450) of the 450 patients suffered AL, which was diagnosed at median postoperative day 7 (range: 6–16). The decision tree model, containing surgical duration, postoperative lymphocyte count, and postoperative C‐reactive protein to albumin ratio, was established by CART analysis. Among the three variables, the postoperative C‐reactive protein to albumin ratio was identified as the most important indicator in the CART model with normalized importance of 100%. According to the results validated in the test set, the sensitivity, specificity, positive and negative predictive value, and diagnostic accuracy of the prediction model were 80%, 98.8%, 88.9%, 97.6%, and 96.8%, respectively. Moreover, the area under the receiver operating characteristic curve was 0.95.

Conclusion

The decision model based on CART analysis presented good performance for predicting AL, and might allow the early identification of patients at high risk.

A new prediction model for assessing the clinical outcomes of ICU patients with community-acquired pneumonia: a decision tree analysis

PURPOSE

We aimed to develop a new scoring index based on decision-tree analysis to predict clinical outcomes of patients with community-acquired pneumonia (CAP) admitted to the intensive care unit (ICU).

METHODS

Data of 3519 ICU patients with CAP were obtained from the Medical Information Mart for Intensive Care III (MIMIC III) 2001-2012 database and analysed between 30-d survivors and non-survivors. Accuracy, sensitivity, and specificity of the new decision tree model were compared with those of CURB-65 and SOAR.

RESULTS

The newly developed classification and regression tree (CART) model identified coexisting illnesses as the most important single discriminating factor between survivors and non-survivors. The CART model area under the curve (AUC) 0.661 was superior to that of CURB-65 (0.609) and SOAR (0.589). CART sensitivity was 73.4%, and specificity 49.0%. CURB-65 and SOAR sensitivity for predicting 30-d mortality were 74.5 and 80.7%, and specificity was 42.3 and 33.9%, respectively. After smoothing, the CART model had higher sensitivity and specificity than both CURB-65 and SOAR.

CONCLUSIONS

The new CART prediction model has higher specificity and better receiver operating characteristics (ROC) curves than CURB-65 and SOAR score indices although its accuracy and sensitivity are only moderately better than the other systems. Key messages The new CART prediction model has higher specificity and better ROC curves than CURB-65 and SOAR score indices. However, accuracy and sensitivity of the new CART prediction model are only moderately better than the other systems in predicting 30-day mortality in CAP patients.

Leptin alleviates intestinal mucosal barrier injury and inflammation in obese mice with acute pancreatitis

BACKGROUND/OBJECTIVES

Obesity is an independent risk factor for severe acute pancreatitis (AP). Leptin plays an important role in energy homeostasis. It has been reported that leptin might also participate in the regulation of the intestinal mucosal barrier and inflammatory response. This study aimed to evaluate the effects of leptin on the intestinal mucosal barrier and inflammatory injury in obese mice with AP.

SUBJECTS/METHODS

AP was induced in leptin-deficient (ob/ob) or wild type (WT) mice by peritoneal injection of caerulein. The animals were divided into 4 groups: WT mice with or without exogenous leptin injection and ob/ob mice with or without leptin treatment. The inflammatory scoring of the pancreas and intestine were evaluated. Intestinal permeability, ileal interleukin (IL)-6 and IL-1β, proliferation, apoptosis and intestinal expression levels of claudin-1 and occludin were measured.

RESULTS

Pancreatic pathologic scores (8.50 ± 0.96 vs. 3.78 ± 1.35, p < 0.001), pancreatic levels of IL-6 (8.34 ± 3.21 ng/mg vs. 4.99 ± 0.53 ng/mg, p = 0.022), intestinal oedema scores (2.25 ± 0.46 vs. 1.14 ± 0.69, p = 0.001) and intestinal permeability to FD4 (0.78 ± 0.06 μg/ml vs. 0.53 ± 0.11 μg/ml, p < 0.001) were significantly higher in ob/ob mice than those in WT mice. Leptin replacement in ob/ob mice greatly improved the intestinal permeability (FD4 0.66 ± 0.03 μg/ml, vs. 0.78 ± 0.06 μg/ml, p = 0.012), increased the ileal expression of claudin-1(1.07 ± 0.08 vs. 0.83 ± 0.07 relative densitometry, p = 0.001) and reduced intestinal IL-6 and IL-1β to levels comparable to those in WT mice. The pancreatic level of IL-6 in ob/ob mice treated with leptin was also significantly decreased relative to that of untreated ob/ob mice (4.45 ± 1.71 ng/mg vs. 8.34 ± 3.21 ng/mg, p = 0.010).

CONCLUSIONS

Obesity may aggravate intestinal inflammation and increase intestinal permeability under the condition of acute pancreatitis. Exogenous leptin supplementation was in favour of anti-inflammation and improvement of intestinal mucosal barrier.

The Reliability of Classification of Terminal Nodes in GUIDE Decision Tree to Predict the Nonalcoholic Fatty Liver Disease

Tree structured modeling is a data mining technique used to recursively partition a dataset into relatively homogeneous subgroups in order to make more accurate predictions on generated classes. One of the classification tree induction algorithms, GUIDE, is a nonparametric method with suitable accuracy and low bias selection, which is used for predicting binary classes based on many predictors. In this tree, evaluating the accuracy of predicted classes (terminal nodes) is clinically of special importance. For this purpose, we used GUIDE classification tree in two statuses of equal and unequal misclassification cost in order to predict nonalcoholic fatty liver disease (NAFLD), considering 30 predictors. Then, to evaluate the accuracy of predicted classes by using bootstrap method, first the classification reliability in which individuals are assigned to a unique class and next the prediction probability reliability as support for that are considered.