Predicting Acute Pancreatitis Severity: Comparison of Prognostic Scores

Impact of obstructive jaundice on outcomes in acute biliary pancreatitis: a retrospective study

OBJECTIVE

To examine the effects of obstructive jaundice on the outcomes of patients with acute biliary pancreatitis.

METHODS

A retrospective chart review was conducted on 332 cases of acute biliary pancreatitis admitted to Rambam Health Care Campus, Israel, from January 1st, 2018, to December 31st, 2021. Patients were categorized based on the presence or absence of obstructive jaundice. Various clinical, laboratory, and radiological parameters were analyzed, including severity prediction scores, length of stay, interventions, and complications.

RESULTS

Obstructive jaundice was observed in 136 patients, while 196 patients had no jaundice. Initial predictive scores (Ranson and Glasgow-Imrie) indicated higher severity in the jaundiced group, but this difference did not translate into significant variations in the final outcomes. Endoscopic procedures and sonography were more frequently performed in jaundiced patients, affecting the diagnosis and management. Cholecystectomy was performed sooner in the jaundiced group, leading to fewer recurrent admissions.

CONCLUSION

The outcomes of jaundiced and non-jaundiced patients with acute biliary pancreatitis were found to be similar, despite initial predictions of worse outcomes in the jaundiced population. A lower threshold for initiation of rigorous treatment, including more frequent endoscopic procedures, administration of antibiotics and early surgical intervention may facilitate these results. Further studies with a larger sample size and long-term follow-up are warranted to validate these findings.

Improved outcome prediction in acute pancreatitis with generated data and advanced machine learning algorithms

OBJECTIVES

Traditional scoring systems have been widely used to predict acute pancreatitis (AP) severity but have limitations in predictive accuracy. This study investigates the use of machine learning (ML) algorithms to improve predictive accuracy in AP.

METHODS

A retrospective study was conducted using data from 101 AP patients in a tertiary hospital in Türkiye. Data were preprocessed, and synthetic data were generated with Gaussian noise addition and balanced with the ADASYN algorithm, resulting in 250 cases. Supervised ML models, including random forest (RF) and XGBoost (XGB), were trained, tested, and validated against traditional clinical scores (Ranson's, modified Glasgow, and BISAP) using area under the curve (AUC), F1 score, and recall.

RESULTS

RF outperformed XGB with an AUC of 0.89, F1 score of 0.82, and recall of 0.82. BISAP showed balanced performance (AUC = 0.70, F1 = 0.44, and recall = 0.85), whereas the Glasgow criteria had the highest recall but lower precision (AUC = 0.70, F1 = 0.38, and recall = 0.95). Ranson's admission criteria were the least effective (AUC = 0.53, F1 = 0.42, and recall = 0.39), probable because it lacked the 48th h features.

CONCLUSION

ML models, especially RF, significantly outperform traditional clinical scores in predicting adverse outcomes in AP, suggesting that integrating ML into clinical practice could improve prognostic assessments.

Diagnostic value of CRP for predicting the severity of acute pancreatitis: a systematic review and meta-analysis

BACKGROUND

C-reactive protein (CRP) is a pentameric protein commonly used as a biomarker of inflammation or stress response which can be obtained during routine blood tests. Therefore, we conducted a systematic review and meta-analysis to explore its ability to predict the severity of acute pancreatitis (AP). This meta-analysis was registered in the PROSPERO system (registration number: CRD42022353769).

METHODS

41 studies with 6156 cases of acute pancreatitis, retrieved from PubMed, Cochrane Library, Springer, and Embase databases, were incorporated. We calculated the pooled estimates for predicting the severity of acute pancreatitis based on CRP levels. We also calculated the combined negative likelihood ratio (NLR), combined positive likelihood ratio (PLR) and combined diagnostic odds ratio (DOR) using a bivariate mixed model. Sensitivity analysis was used to examine the robustness of the results. Factors associated with heterogeneity were identified by meta-regression analysis. A summary operating characteristic (SROC) curve was generated to assess the diagnostic value of CRP in predicting severe acute pancreatitis. Fagan's test was used to calculate likelihood ratios and post-test probabilities, and publication bias was gauged by asymmetry tests.

RESULTS

SROC analysis yielded an AUC of 0.85 (95%CI: 0.81-0.88) with a sensitivity of 0.76 (95%CI: 0.69-0.83) and specificity of 0.79 (95%CI: 0.74-0.83). The combined NLR, PLR and DOR were 0.30 (0.23-0.40), 3.66 (2.94-4.55) and 12.19 (8.05-18.44) respectively. Sensitivity analysis demonstrated the stability of our results after omitting any study. Finally, meta-regression analysis indicated that the description of the reference test, prospective design, blinding method and spectrum of the disease could account for heterogeneity in this meta-analysis.

CONCLUSION

CRP has significant value as a biomarker for assessing AP severity. Besides, other parameters such as patient history, physical signs, and imaging should be considered to determine disease severity.

A new logistic regression model for early prediction of severity of acute pancreatitis using magnetic resonance imaging and Acute Physiology and Chronic Health Evaluation II scoring systems

BACKGROUND

The aim of this study was to develop a new model constructed by logistic regression for the early prediction of the severity of acute pancreatitis (AP) using magnetic resonance imaging (MRI) and the Acute Physiology and Chronic Health Evaluation II (APACHE II) scoring system.

METHODS

This retrospective study included 363 patients with AP. The severity of AP was evaluated by MRI and the APACHE II scoring system, and some subgroups of AP severity were constructed based on a combination of these two scoring systems. The length of stay and occurrence of organ dysfunction were used as clinical outcome indicators and were compared across the different subgroups. We combined the MRI and APACHE II scoring system to construct the regression equations and evaluated the diagnostic efficacy of these models.

RESULTS

In the 363 patients, 144 (39.67%) had systemic inflammatory response syndrome (SIRS), 58 (15.98%) had organ failure, and 17 (4.68%) had severe AP. The AP subgroup with a high MRI score and a simultaneously high APACHE II score was more likely to develop SIRS and had a longer hospitalization. The model, which predicted the severity AP by combining extrapancreatic inflammation on magnetic resonance (EPIM) and APACHE II, was successful, with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.912, which was higher than that of any single parameter. Other models that predicted SIRS complications by combining MRI parameters and APACHE II scores were also successful (all P<0.05), and these models based on EPIM and APACHE II scores were superior to other models in predicting outcome.

CONCLUSIONS

The combination of MRI and clinical scoring systems to assess the severity of AP is feasible, and these models may help to develop personalized treatment and management.

Comparison of modified Glasgow-Imrie, Ranson, and Apache II scoring systems in predicting the severity of acute pancreatitis

&lt;b&gt;Aim:&lt;/b&gt; The course of acute pancreatitis is variable with patients at risk of poor outcomes. The purpose of this study was to compare Modified Glasgow-Imrie, Ranson, and APACHE II scoring systems in predicting the severity of acute pancreatitis. &lt;/br&gt;&lt;/br&gt; &lt;b&gt; Material and Methods: &lt;/b&gt; After a brief history, clinical examination and qualifying inclusion criteria, 70 patients (41 women, 29 men) diagnosed with acute pancreatitis were included in the study. The three scores were calculated for each patient and evaluated for their role in the assessment of specific outcomes. &lt;/br&gt;&lt;/br&gt; &lt;b&gt;Results:&lt;/b&gt; 34.3% patients were diagnosed with severe acute pancreatitis, while 65.7% patients had mild acute pancreatitis. A strong positive correlation was found between all the prognostic scores and the severity of disease. In the prediction of the severity of disease according to AUC, it was found that Glasgow-Imrie score had an AUC of 0.864 (0.7560.973), followed very closely by APACHE II score with an AUC of 0.863 (0.7580.968). APACHE II had the highest sensitivity (79.17%) in predicting severity while Glasgow-Imrie score was the most specific (97.83%) of all the scores. Patients with a Glasgow-Imrie score above the cut-off value of 3 had more complications and a longer hospital stay. &lt;/br&gt;&lt;/br&gt; &lt;b&gt;Conclusion:&lt;/b&gt; The Glasgow-Imrie score was comparable to APACHE II score and better than Ranson score statistically in predicting the severity of acute pancreatitis. Its administration in predicting the severity of acute pancreatitis is recommended.

Sequential organ failure assessment score is superior to other prognostic indices in acute pancreatitis

BACKGROUND

Acute pancreatitis (AP) is a common surgical condition, with severe AP (SAP) potentially lethal. Many prognostic indices, including; acute physiology and chronic health evaluation II score (APACHE II), bedside index of severity in acute pancreatitis (BISAP), Glasgow score, harmless acute pancreatitis score (HAPS), Ranson's score, and sequential organ failure assessment (SOFA) evaluate AP severity and predict mortality.

AIM

To evaluate these indices' utility in predicting severity, intensive care unit (ICU) admission, and mortality.

METHODS

A retrospective analysis of 653 patients with AP from July 2009 to September 2016 was performed. The demographic, clinical profile, and patient outcomes were collected. SAP was defined as per the revised Atlanta classification. Values for APACHE II score, BISAP, HAPS, and SOFA within 24 h of admission were retrospectively obtained based on laboratory results and patient evaluation recorded on a secure hospital-based online electronic platform. Data with < 10% missing data was imputed via mean substitution. Other patient information such as demographics, disease etiology, and patient outcomes were also derived from electronic medical records.

RESULTS

The mean age was 58.7 ± 17.5 years, with 58.7% males. Gallstones (n = 404, 61.9%), alcohol (n = 38, 5.8%), and hypertriglyceridemia (n = 19, 2.9%) were more common aetiologies. 81 (12.4%) patients developed SAP, 20 (3.1%) required ICU admission, and 12 (1.8%) deaths were attributed to SAP. Ranson's score and APACHE-II demonstrated the highest sensitivity in predicting SAP (92.6%, 80.2% respectively), ICU admission (100%), and mortality (100%). While SOFA and BISAP demonstrated lowest sensitivity in predicting SAP (13.6%, 24.7% respectively), ICU admission (40.0%, 25.0% respectively) and mortality (50.0%, 25.5% respectively). However, SOFA demonstrated the highest specificity in predicting SAP (99.7%), ICU admission (99.2%), and mortality (98.9%). SOFA demonstrated the highest positive predictive value, positive likelihood ratio, diagnostic odds ratio, and overall accuracy in predicting SAP, ICU admission, and mortality. SOFA and Ranson's score demonstrated the highest area under receiver-operator curves at 48 h in predicting SAP (0.966, 0.857 respectively), ICU admission (0.943, 0.946 respectively), and mortality (0.968, 0.917 respectively).

CONCLUSION

The SOFA and 48-h Ranson's scores accurately predict severity, ICU admission, and mortality in AP, with more favorable statistics for the SOFA score.

Acute Pancreatitis During COVID-19 Pandemic: An Overview of Patient Demographics, Disease Severity, Management and Outcomes in an Acute District Hospital in Northern Ireland

Background Acute pancreatitis (AP) is a common disease requiring admissions under surgical and critical care units. The two most common causes are alcohol and gallstones. Coronavirus disease 2019 (COVID-19) pandemic had a significant impact on service delivery and patient management throughout all surgical specialties. In this study, the primary aim was to ascertain the incidence of COVID-19 in acute pancreatitis patients. Secondary objectives were to study aetiology, demographics, severity, 30-day mortality, outcomes and management of acute pancreatitis patients from 1st March, 2020 till 31st August, 2020. Methods A retrospective observational review of all patients admitted under the General Surgical team was performed. Information regarding demographics, severity of AP (using Glasgow score, Atlanta classification and CT severity index score), ICU admission and organ support, treatment modalities and follow-up data for outcomes was collected based on data collection tool used by COVID-PAN study and results were compared to outcomes results of COVID-PAN study. Results Forty-three (43) patients were admitted with AP. Only one patient (2.3%) was diagnosed with COVID-19 at the time of pancreatitis. Gallstones were noted to be the most common cause of AP in our population. Mortality was 7% (3 patients). Five patients (11%) needed ITU admission due to organ dysfunction. Three patients (7%) developed ARDS. Conclusion The overall incidence of COVID-19 in pancreatitis in our population of the study was low. The incidence of COVID-19 during the first wave in Derry/Londonderry area was low and this may explain why the incidence was low in our study as well. Patients with AP in our target population were mostly elderly, one in five had moderate to severe or severe pancreatitis and in 16.3% the aetiology could not be identified. As observed in other centres globally, urgent cholecystectomy for gallstone pancreatitis faced significant delays with no patients being offered index cholecystectomy and only 4 out of 19 patients having undergone interval cholecystectomy within six months of index admission for gallstone pancreatitis in our centre.

Comparison of Different Scoring Systems in Predicting the Severity of Acute Pancreatitis: A Prospective Observational Study

Background

Acute pancreatitis (AP) is an inflammatory condition of the pancreas mostly due to alcohol or gallstones. Various scoring systems were involved in identifying the severity of the disease. The standard single score to identifying the severity remains uncertain.

Methodology

This prospective observational study was carried out for two years in a tertiary care center from South India. The diagnosis of AP was made based on Atlanta criteria, and a total of 164 patients were included. All patients were assessed by acute physiology and chronic health evaluation ll (APACHE II), bedside index for severity in AP (BISAP), modified Glasgow score (MGS), and Ranson score on admission and 48 hours after admission scores. Procalcitonin was done in all patients with AP. Contrast-enhanced computed tomography (CT) of the abdomen was done in 69 patients who had features of severe acute pancreatitis (SAP). Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy were calculated for each score, and procalcitonin for CT documented severe patients and organ failure patients together.

Results

A total of 164 patients were included in this study. CT abdomen showed a modified CT severity index (MCSI) ≥8 in all 69 (100%) patients. APACHE II score could predict SAP based on CT findings in 44 patients (63.76%), BISAP score in 22 patients (31.88%), MGS in 55 patients (79.71%), Ranson score at admission in 31 patients (44.92%), Ranson score 48 hours after admission in 44 patients (63.76%), and procalcitonin on admission in 69 patients (100%) when cut-off used as per the literature. APACHE II score could predict SAP in cases of AP (n=164) in 52 patients (50%), BISAP score in 27 patients (26%), MGS in 79 patients (76%), Ranson score at admission in 34 patients (33%), and Ranson score 48 hours after admission in 61 (59%) patients when cut-off was used as per the literature. This study demonstrated that Ranson score on admission had a good area under the curve (AUC). AUC (0.8483), APACHE II (AUC 0.7708), and Ranson score 48 hours after admission (AUC 0.8167) had a fair accuracy. BISAP (AUC 0.6399) and MGS (AUC 0.6486) had poor accuracy for the prediction of severity in AP based on receiver operator characteristic (ROC) curves.

Conclusion

Among the scoring system compared, MGS had the highest sensitivity for predicting the severity of AP. However, Ranson score on admission had better diagnostic accuracy for predicting severity, organ failure, and mortality based on ROC curves. Procalcitonin had the best sensitivity, specificity, PPV, NPV, and diagnostic accuracy for association with severity in AP.

Can C-reactive Protein Increase the Efficiency of the Bedside Index of Severity in Acute Pancreatitis Scoring System?

Background

Early diagnosis and accurate assessment of the severity of the disease are critical factors in the management of acute pancreatitis (AP). In this study, we investigated the success rates of combinations of Bedside Index of Severity in Acute Pancreatitis (BISAP) scores with C-reactive protein (CRP) values in predicting severe AP.

Methods

The medical records of all patients with AP admitted to our hospitals from September 2015 to September 2018 were reviewed retrospectively. To evaluate the severity of AP, the revised Atlanta criteria were used, and patients who developed organ failure lasting more than 48 hours were considered to have severe AP. We analyzed patient CRP values at the 24-hour mark via receiver operating characteristic (ROC) curve analysis. Four groups were then formed to separate mild AP from moderate to severe AP. The first group had BISAP scores ≥ 3, the second group had CRP values ≥ 90.7 mg/L, the third group had BISAP scores ≥ 3 and CRP values ≥ 90.7 mg/L, and the fourth group had BISAP scores ≥ 3 or measured CRP values ≥ 90.7 mg/L. Predictive accuracy, sensitivity, specificity, and positive and negative predictive values of groups in the prediction of severe AP were calculated.

Results

Our study population consisted of 207 patients, and according to the revised Atlanta scoring, 165 patients (79.7%) had mild AP, 30 (14.4%) had moderate, and 12 (5.8%) had severe AP. Comparing the mild, moderate, severe AP groups, we noted a significant difference between the mean hospital stay time, BISAP scores, and CRP values (p<0.001). Group 1, 2, 3, and 4 values of mild AP and all severe AP (moderate and severe) were significant (p<0.001). The highest specificity values were found in Group 3 (97.6%), while the highest sensitivity values were observed in Group 4 (88.1%).

Conclusion 

CRP may increase the success of BISAP scoring in predicting the severity of AP.

Are traditional scoring systems for severity stratification of acute pancreatitis sufficient?

BACKGROUNDS/AIMS

Ranson's score (RS) and Glasgow score (GS) have been utilized to stratify the severity of acute pancreatitis (AP). The aim of this study was to validate RS and GS for stratifying the severity of acute pancreatitis and audit our experience of managing AP.

METHODS

We conducted a retrospective review of patients treated for AP from July 2009 to September 2016. Final severity was determined using the revised Atlanta classification. Mortality and complications were analyzed.

RESULTS

From July 2009 to September 2016, a total of 675 patients with a diagnosis of AP were admitted at the hospital. Of them, 669 patients who had sufficient data were analyzed. Their average age±SD was 58.7±17.4 years (range, 21-98 years). There was a male preponderance (n=393, 53.8%). A total of 82 (12.3%) patients had eventual severe pancreatitis. RS demonstrated a sensitivity of 92.7% and a specificity of 52.8% with a positive predictive value (PPV) of 21.5% and a negative predictive value (NPV) of 98.1%. GS demonstrated a sensitivity of 76.8% and a specificity of 69.2% with a PPV of 25.8% and a NPV of 95.5%. For severity prediction, areas under the curve (AUCs) for RS and GS were 0.848 (95% CI: 0.819-0.875) and 0.784 (95% CI: 0.750-0.814), respectively (p=0.003). Twelve (1.6%) patients died in the hospital.

CONCLUSIONS

RS has higher sensitivity, NPV and AUC for predicting severity of AP than GS.

Current concepts in the management of acute pancreatitis

Guidelines for the management of acute pancreatitis (AP) are based on the Western experience, which may be difficult to extrapolate in India due to socioeconomic constraints. Hence, modifications based on the available resources and referral patterns should be introduced so as to ensure appropriate care. We reviewed the current literature on the management of AP available in English on Medline and proposed guidelines locally applicable. Patients of AP presenting with systemic inflammatory response syndrome are at risk of moderate-severe pancreatitis and hence, should be referred to a tertiary center early. The vast majority of patients with AP have mild disease and can be managed at smaller centers. Early aggressive fluid resuscitation with controlled fluid expansion, early enteral nutrition, and culture-directed antibiotics improve outcomes in AP. Infected pancreatic necrosis should be managed in a tertiary care hospital within a multidisciplinary setup. The "step up" approach involving antibiotics, percutaneous drainage, and minimally invasive necrosectomy instituted sequentially based on clinical response has improved the outcomes in this subgroup of patients.