F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.133278.1 Research Article Articles A comparative evaluation of different scores in predicting severity and outcome in acute pancreatitis

[version 1; peer review: 1 not approved]

 Mahajan Ojas Data Curation https://orcid.org/0000-0001-6215-1726 1 Mahajan Satish Conceptualization a 1 Acharya Sourya Supervision 1 Wanjari Anil Supervision 1 Kumar Sunil Supervision 1 Bawankule Shilpa Supervision 1 Giri Anamika Writing – Original Draft Preparation 1 Khurana Kashish Writing – Original Draft Preparation https://orcid.org/0000-0002-6672-3784 1 Department of Medicine, Jawaharlal Nehru Medical College, Datta Meghe Institute of Higher Education and Research, Wardha, Maharashtra, 442001, India satish.mahajan@dmiher.edu.in

No competing interests were disclosed.

 13 7 2023 2023 12 824 12 4 2023 Copyright: © 2023 Mahajan O et al. 2023 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Acute pancreatitis (AP) is an inflammatory condition usually caused by alcohol or gallstones. Our goal was to prospectively compare the diagnostic efficacy of the Acute Physiology & Chronic Health Evaluation (APACHE) II, the Bedside Index of Severity in Acute Pancreatitis (BISAP), the Ranson's score & the Modified Glasgow Score (MGS) in determining the severity & outcome of Acute pancreatitis in a tertiary care facility in central India.

Methods: Between December 2020 & December 2022, this prospective observational study was done in rural area of Wardha district. 110 subjects were included, and the diagnosis of acute pancreatitis was done using Atlanta criteria. APACHE II, MGS, Ranson score on admission, Ranson score 48 hours after admission & BISAP were used to evaluate each subject. The reciever operating curve was used to measure the specificity, sensitivity, NPV, PPV, diagnostic accuracy, area under the curve (AUC) & these scoring methods were then prospectively compared.

Results: When a cut-off based on the literature was used, the APACHE II score could accurately diagnose severe cases of AP (n=110) in 69 patients, BISAP in 68 patients, MGS in 49, Ranson score on admission in 48 patients & after 48 hours in 48 patients. This study showed that Ranson score 48 hours after admission had a AUC (0.991), Ranson score at admission (AUC 0.989) & Modified Glasgow Scale (AUC 0.6486) had fair accuracy as compared to APACHE II (AUC 0.974) & BISAP (AUC 0.896) for determining the level of severity among AP patients based on ROC curves.

Conclusion: To predict the severity of AP, the Ranson score after 48 hours showed the highest NPV, PPV, sensitivity, specificity, and diagnostic accuracy of all the scoring methods tested. The BISAP score had the highest specificity, sensitivity, PPV& NPV for determining the outcome of AP.

 Acute Pancreatitis APACHE II BISAP Ranson’s score MGS The author(s) declared that no grants were involved in supporting this work. Introduction

The pancreas is a unique endocrine & exocrine organ. The exocrine glands discharges enzymes into the digestive system, while the endocrine glands secrete hormones into the circulatory system. 1 According to Claude Bernard's theory from 1856, acute pancreatitis was brought on by bile reflux into the common pancreatic duct. However, the controversy did not begin to be settled until 1901, when Eugene Opie proposed that gallstone immigration into the common bile duct was the major cause of acute pancreatitis. 2 When intracellular defences that stop trypsinogen activation or lessen trypsin activity are overpowered, AP results. Pancreatic enzyme activation causes the gland to digest itself & causes localised inflammation. 3 Necrotizing pancreatitis & interstitial oedematous pancreatitis are two types of acute pancreatitis. Inflammation & edema are present in the peripancreatic tissues and pancreatic parenchyma in interstitial edematous pancreatitis. Necrotizing pancreatitis appears when this condition deteriorates to the point of peripancreatic or pancreatic tissue death. 4 Most patients have localised inflammation, but about one-fifth of them go on to develop multiple organ dysfunction syndrome (MODS), which is linked to a significant increase in mortality. A diagnosis of acute pancreatitis requires abdominal pain, high pancreatic amylase and/or lipase values that are at least three times above normal, & imaging studies that show anomalies unique to acute pancreatitis. The Atlanta Classification has been used to gauge the severity of acute pancreatitis ever since its establishment in 1992. 5 The original classification's boundaries given by Atlanta, notably the definition of severity is unclear. In 2012, the Atlanta classification was revised to include chronic organ failure. 6 Early diagnosis of acute pancreatitis is crucial for lowering morbidity & mortality caused by the condition. To determine the severity of AP, many biochemical, radiological & clinical scores had been devised in the past. Ranson's score, the CT-severity index (CTSI), the bedside index of severity in acute pancreatitis (BISAP), the Modified Glasgow Score, & APACHE-II are some of them. 7 Acute pancreatitis severity can now be determined with the use of these forecasting methods. Though complex & difficult to use in clinical settings, it has been shown that these multi-factorial scoring systems work with a high negative predictive value but a meagre overall sensitivity. 8 In this study, we aim to determine which scoring method best predicted acute pancreatitis severity. Its secondary aim was to determine which scoring method best predicted the outcome of AP.

 Methods

This observational and prospective research work was conducted for two years at a tertiary care centre in the rural area of Wardha district. Patients with significant abdominal pain, elevated serum amylase and serum lipase levels more than three times the upper limit of the normal range and abdominal ultrasonography findings suggestive of acute pancreatitis were included in the study. 110 subjects were studied. The Atlanta Criteria was used for the diagnosis of acute pancreatitis. The study was approved by Institutional Ethical Committee, Datta Meghe Institute of Medical Science (Deemed to be University) with Ref.No. DMIMS (DU)/IEC/2020-21/9284.

The enrolment of the subjects was started after approval from the ethics committee. The study included subjects over the age of 18 who provided written consent. The study excluded all subjects with chronic pancreatitis and those who received outside treatment prior to coming to the emergency room.

All subjects admitted for acute pancreatitis had their prospectively gathered demographic, clinical, biochemical, & radiological data. One can determine if they have acute pancreatitis if they meet two out of the following three criteria: (i) AP-specific abdominal pain, (ii) Serum amylase and serum lipase levels that are at least three times above normal (iii) Radiological abnormalities on abdominal ultrasonography and/or computerized tomography (CT) scan characteristic of acute pancreatitis. Patients with features of chronic pancreatitis, such as dilated pancreatic ducts, pancreatic calcifications, pseudocysts & areas of atrophy, discovered during radiological examinations performed while they were hospitalised or who had chronic pancreatitis based on their prior hospital records were excluded from the research.

After a thorough history & physical examination, several clinical & biochemical factors were assessed. All subjects underwent abdominal ultrasonography at the time of admission. Subjects with mild AP did not experience any local consequences or organ failure, however those with severity did. The local sequelae included pseudocyst, pancreatic necrosis, walled off necrosis, & acute fluid collections. MGS, BISAP, APACHE II score, Ranson score on admission & 48 hrs after were the grading systems used for assessment of all the subjects. Subjects were followed until they were discharged or died.

Statistical analysis

IBM's SPSS version 23.0 in Chicago, United States, was used to do the analysis. We used the ROC curve to establish a cut-off for the APACHE II score, the BISAP score, the Modified Glasgow score, the Ranson score at admission & after 48 hours, all of which predict outcome and severity. The diagnostic accuracy, sensitivity, and specificity of the four scores were evaluated for their ability to foretell the likelihood of severe outcomes or death.

 Results

According to the classification given by Atlanta, 110 subjects displayed symptoms of acute pancreatitis. Out of them, 60 (54.54%) had mild and 50 (45.46%) had severe acute pancreatitis. nine (0.08%) of them died while being treated in the hospital ( Tables 1 and 2; Figures 1 and 2; Table 3).

Baseline characteristics in the study group.
Characteristics Category N (%)
110 (100%)
Age 18-30 years 24 (21.8%)
31-40 years 38 (34.5%)
41-50 years 24 (21.8%)
51-60 years 16 (14.5%)
61-70 years 7 (6.4%)
71-80 years 1 (0.9%)
Sex Male 100 (90.9%)
Female 10 (9.1%)
Severity of Acute Pancreatitis Mild 60 (54.54%)
Severe 50 (45.46%)
APACHE II <8 41 (37.3%)
≥8 69 (62.7%)
BISAP <2 42 (38.2%)
≥2 68 (61.8%)
Ranson Score at admission <3 62 (56.4%)
≥3 48 (43.6%)
Ranson Score 48 hours after admission <5 62 (56.4%)
≥5 48 (43.6%)
Modified Glasgow Score <3 61 (55.5%)
≥3 49 (44.5%)
Outcome Discharged 101 (99.92%)
Death 9 (0.08%)
Comparison of the diagnostic performance of various scoring methods in predicting severity in AP.
Predictor AUROC 95% CI P Sn Sp PPV NPV DA
APACHE II Score 0.812 0.726-0.899 <0.001 75% 72% 68% 78% 73%
BISAP Score 0.751 0.657-0.846 <0.001 73% 67% 71% 83% 70%
Ranson Score At Admission 0.885 0.823-0.947 <0.001 85% 75% 73% 86% 80%
Ranson Score After 48 Hours 0.900 0.841-0.960 <0.001 87% 86% 84% 89% 87%
Modified Glasgow Score 0.812 0.729-0.895 <0.001 77% 72% 70% 80% 74%

AUROC: Area under ROC curve; CI: Confidence interval; P: P value; Sn: Sensitivity; Sp: Specificity; PPV: Positive predictive value; NPV: Negative predictive value; DA: Diagnostic Accuracy.

 ROC curve representing diagnostic performance of various scoring systems in predicting severity in AP. ROC curve analysis showing diagnostic performance of scoring methods in predicting outcome in AP. Analyzing the diagnostic accuracy of scoring methods for outcome prediction in AP.
Predictor AUROC 95% CI P Sn Sp PPV NPV DA
APACHE II Score 0.812 0.696-0.927 <0.001 75% 60% 34% 89% 63%
BISAP Score 0.841 0.724-0.958 <0.001 83% 94% 80% 95% 91%
Ranson Score At Admission 0.824 0.748-0.900 <0.001 70% 73% 42% 90% 72%
Ranson Score After 48 Hours 0.736 0.613-0.860 0.001 70% 80% 50% 90% 78%
Modified Glasgow Score 0.819 0.738-0.899 <0.001 66% 75% 43% 89% 73%

AUROC: Area under ROC curve; CI: Confidence interval; P: P value; Sn: Sensitivity; Sp: Specificity; PPV: Positive predictive value; NPV: Negative predictive value; DA: Diagnostic Accuracy.

 Discussion

Acute pancreatitis is pancreatic inflammation which can range from mild to severe. The majority of patients suffer with mild condition with low morbidity, while the others have severe acute pancreatitis, which has a mortality rate of 10% to 20%. 9 In present study mean age of the subjects was 40.49±12.12 years and maximum i.e. 38 (34.5%) of the participants were in the age group of 31-40 years. 72 individuals, evaluated by Ajay K. Khanna et al., had mean age of 40.5 years , similar to our study. 9 Similarly, in a study by Anubhav Kumar et al., 50 patients with acute pancreatitis had the mean age of 48.42 years. 10 In our study there were 100 (90.9%) males and 10 (9.1%) of the subjects were females. Similar findings were reported by Lankisch PG et al., as they found, out of 274 patients 172 were males and 102 were females. 11 We found that the mean value of Serum Amylase was 399.82±104.82 U/L and 680.06±157.54 U/L in Mild and Severe acute pancreatitis groups, respectively. Kiat et al, also found in their study that the mean serum amylase was 1151.1±753.5 U/L and 1484.6±736.5 U/L in Mild and Severe acute pancreatitis groups, respectively. 12 In the present study, out of 110 subjects, 41 (37.3%) had APACHE II Score of <8 and 69 (62.7%) had APACHE II Score of ≥8. In their analysis of 161 Acute Pancreatitis patients, Cho J H et al., also observed that 52 (32.2%) of Severe Acute Pancreatitis patients had APACHE II score of <8 and 109 (67.8%) of Severe Acute Pancreatitis had APACHE II score of ≥8. 13 In our study, 42 (38.2%) subjects had BISAP Score of <2 and 68 (61.8%) subjects had BISAP Score of ≥2. Cho et al., studied 161 subjects and found that 109 (67.70%) participants had BISAP Score of <2 and 52 (32.30%) participants had BISAP Score ≥2, which was in contrast to our study. 13 In the present study, 62 (56.4%) subjects had Ranson score of <3 at admission and 48 (43.6%) subjects had Ranson score of ≥3 at admission. Similarly, Khanna et al. in their study of 72 participants showed that 37 (51.4%) participants had Ranson score of <3 and 35 (48.6%) participants had Ranson of ≥ 3. 9 In our study, 61 (55.5%) subjects had Modified Glasgow Score of <3 and 49 (44.5%) subjects had Modified Glasgow Score of ≥3. Similar results were reported by by Kiat TT et al. in their study of 669 participants which showed that 425 (63.52%) participants had Modified Glasgow Score of <3 and 244 (36.48%) participants had Modified Glasgow Score ≥3. 12

Comparison of different scores in predicting the severity of acute pancreatitis patients

In present study, we found that Ranson score 48 hours after admission (AUROC=0.900) was better than Ranson score at admission, Modified Glasgow Score, APACHE II score and BISAP score in predicting severity of acute pancreatitis. According to Kiat TT et al., Ranson score had higher sentivity, negative predictive value and area under ROC curve (AUROC=0.848) for predicting severity in acute pancreatitis than Modified glasgow score, similar to our study. 12 According to Arif A et al., Ranson score predicted Severe Acute Pancreatitis more correctly than BISAP score similar to our study. 14 According to Wei Gao et al., Severe Acute Pancreatitis was predicted more correctly by Ranson's score than by BISAP score with area under ROC curve (AUROC) 0.83, sensitivity 66% and specificity 78%. This was consistent with our study. 15 Similarly, Zhang J et al., revealed that the AUC for predicting severity by BISAP was 0.793, APACHE II was 0.836 and by Ranson score was 0.903. 16

Comparison of different scores in predicting the outcome of acute pancreatitis patients

In our study, we found that BISAP score was best in predicting outcome of acute pancreatitis followed by APACHE II score then Modified Glasgow Score then Ranson score at admission and Ranson score 48 hours after admission. Wei Gao et al., reported that Ranson score had the area under ROC curve (AUROC) 0.92, sensitivity and specificity were 93% and 69% respectively. BISAP score had the area under ROC curve (AUROC) 0.82, sensitivity and specificity were 81% and 70% respectively. Compared with the Ranson criteria, BISAP score outperformed in specificity, but having a suboptimal sensitivity for mortality in Acute Pancreatitis. This was consistent with our study. 15 Eachempati SR et al., found that Ranson score was a good predictor of outcomes in patients with Severe Acute Pancreatitis. Ranson score after 48 hours predicted outcomes in patients more accurately than Ranson score at admission with Severe Acute Pancreatitis. 17 In 2013, Chen L. et al., reported that to accurately predict death in Severe Acute Pancreatitis patients, the BISAP score performed comparably to other scoring systems with area under ROC curve (AUROC) of 0.808, sensitivity of 83.3% and specificity of 67.4%. 18 Zhang J et al., revealed that the AUC for mortality predicted by BISAP was 0.791, APACHE II was 0.812 and by Ranson score was 0.904. BISAP score was found as valuable source for prognostic prediction in Chinese patients with Acute Pancreatitis. 16

Conclusion

We conclude that the best scoring system for predicting the severity of Acute Pancreatitis is Ranson Score after 48 Hours with area under the curve, 0.900 (0.841 – 0.960), at 95% CI with p value <0.001 which was found to have the highest sensitivity and specificity of 87% and 86%, respectively. That was followed by Ranson score at admission with area under the curve 0.885, sensitivity of 85%, specificity of 75%. Then Modified glasgow scale with area under the curve 0.812, sensitivity of 77%, specificity of 72% then APACHE II score with area under the curve 0.812, sensitivity of 75%, specificity of 72% and lastly BISAP score with area under the curve 0.751, sensitivity of 73%, specificity of 67%.

We also conclude that BISAP Score was best with area under the curve, 0.841 (0.724 – 0.958) at 95% CI, with p<0.001 was found to have the highest sensitivity and specificity of 83% and 94%, respectively for predicting the outcome of Acute Pancreatitis as compared to Ranson score at admission with area under the curve 0.824, sensitivity of 70%, specificity of 73%, Modified glasgow scale with area under the curve 0.819, sensitivity of 66%, specificity of 75%, APACHE II with area under the curve 0.812, sensitivity of 75%, specificity of 60% and Ranson score after 48 hours with area under the curve 0.736, sensitivity of 70%, specificity of 80%.

 Data availability

Zenodo. A Comparative Evaluation of Different Scores in Predicting Severity and Outcome in Acute Pancreatitis. DOI: https://doi.org/10.5281/zenodo.7777706. 19

This project contains the following data:

This dataset is regarding my study which I conducted on 110 subjects

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Reporting guidelines

Zenodo. STROBE checklist. DOI: https://doi.org/10.5281/zenodo.7777876

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

 Acknowledgements

We thank all the participants who have contributed to this study.

 References Frantz E Souza-Mello V Mandarim-de-Lacerda C : Pancreas: Anatomy, diseases and health implications. Pancreas. 2012;41:1–9. 10.1097/MPA.0b013e3182236320 Frossard JL Steer ML Pastor CM : Acute pancreatitis. Lancet. 2008;371:143–152. 10.1016/S0140-6736(08)60107-5 Wang GJ Gao CF Wei D : Acute pancreatitis: etiology and common pathogenesis. World J. Gastroenterol: WJG. 2009 Mar 3;15(12):1427–1430. 19322914 10.3748/wjg.15.1427 PMC2665136 Foster BR Jensen KK Bakis G : Revised Atlanta classification for acute pancreatitis: a pictorial essay. Radiographics. 2016 May;36(3):675–687. 27163588 10.1148/rg.2016150097 Bradley EL : A clinically based classification system for acute pancreatitis: summary of the International Symposium on Acute Pancreatitis, Atlanta, Ga, September 11 through 13, 1992. Arch. Surg. 1993 May 1;128(5):586–590. 10.1001/archsurg.1993.01420170122019 Banks PA Bollen TL Dervenis C : Classification of acute pancreatitis—2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013 Jan 1;62(1):102–111. 23100216 10.1136/gutjnl-2012-302779 Wu BU Johannes RS Sun X : The early prediction of mortality in acute pancreatitis: a large population-based study. Gut. 2008 Dec 1;57(12):1698–1703. 18519429 10.1136/gut.2008.152702 Papachristou GI Muddana V Yadav D : Comparison of BISAP, Ranson's, APACHE-II, and CTSI scores in predicting organ failure, complications, and mortality in acute pancreatitis. Am. J. Gastroenterol. 2010 Feb 1;105(2):435–441. 19861954 10.1038/ajg.2009.622 Khanna AK Meher S Prakash S : Comparison of Ranson, Glasgow, MOSS, SIRS, BISAP, APACHE-II, CTSI Scores, IL-6, CRP, and procalcitonin in predicting severity, organ failure, pancreatic necrosis, and mortality in acute pancreatitis. HPB Surg. 2013 Sep 24;2013:1–10. 24204087 10.1155/2013/367581 PMC3800571 Harshit Kumar A Singh GM : A comparison of APACHE II, BISAP, Ranson’s score and modified CTSI in predicting the severity of acute pancreatitis based on the 2012 revised Atlanta Classification. Gastroenterol. Rep. 2018 May;6(2):127–131. 29780601 10.1093/gastro/gox029 PMC5952961 Lankisch PG Assmus C Lehnick D : Acute pancreatitis: does gender matter? Dig. Dis. Sci. 2001 Nov;46:2470–2474. 10.1023/A:1012332121574 Kiat TT Gunasekaran SK Junnarkar SP : Are traditional scoring systems for severity stratification of acute pancreatitis sufficient? Ann. Hepatobiliary Pancreat. Surg. 2018 May 1;22(2):105–115. 29896571 10.14701/ahbps.2018.22.2.105 PMC5981140 Bollen TL Singh VK Maurer R : A comparative evaluation of radiologic and clinical scoring systems in the early prediction of severity in acute pancreatitis. Am. J. Gastroenterol. 2012 Apr 1;107(4):612–619. 22186977 10.1038/ajg.2011.438 Arif A Jaleel F Rashid K : Accuracy of BISAP score in prediction of severe acute pancreatitis. Pak. J. Med. Sci. 2019 Jul;35(4):1008–1012. 31372133 10.12669/pjms.35.4.1286 Gao W Yang HX Ma CE : The value of BISAP score for predicting mortality and severity in acute pancreatitis: a systematic review and meta-analysis. PLoS One. 2015 Jun 19;10(6):e0130412. 26091293 10.1371/journal.pone.0130412 PMC4474919 Zhang J Shahbaz M Fang R : Comparison of the BISAP scores for predicting the severity of acute pancreatitis in Chinese patients according to the latest Atlanta classification. J. Hepatobiliary Pancreat. Sci. 2014 Sep;21(9):689–694. 24850587 10.1002/jhbp.118 Eachempati SR Hydo LJ Barie PS : Severity scoring for prognostication in patients with severe acute pancreatitis: comparative analysis of the Ranson score and the APACHE III score. Arch. Surg. 2002 Jun 1;137(6):730–736. 12049546 Chen L Lu G Zhou Q : Evaluation of the BISAP score in predicting severity and prognoses of acute pancreatitis in Chinese patients. Int. Surg. 2013;98(1):6–12. 23438270 10.9738/0020-8868-98.1.6 PMC3723156 Mahajan O : A Comparative Evaluation of Different Scores in Predicting Severity and Outcome in Acute Pancreatitis.[Data set]. Zenodo. 2023. 10.5281/zenodo.7777706 10.5256/f1000research.146257.r209095 Reviewer response for version 1 Mao Wenjian 1 Referee https://orcid.org/0000-0002-0033-8939 Jinling Hospital, Medical School of Nanjing University, Nanjing, Jiangsu, China

Competing interests: No competing interests were disclosed.

 20 10 2023 Copyright: © 2023 Mao W 2023 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation reject

In the Abstract and the results section, cut-off value of each scoring system should be given.

At present, the revised Atlanta Criteria was widely used to define or classify acute pancreatitis (AP) over the world. Out of the 110 AP patients, 60 (54.54%) had mild and 50 (45.46%) had severe acute pancreatitis. It is unbelievable that there are no patients diagnosed with moderate severe acute pancreatitis (MSAP) according to the revised Atlanta Criteria. Please check and make some revision.

What's the outcome of AP. It should be given definitely in the methods section. All definition should not be ambiguous.

AP patients' severities are affected by various factors. To compare the predictive value of each scoring system, multivariate logistic regression models are recommended to evaluate them.

Some attention should be paid to the writing style and grammar.

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Not applicable

Are all the source data underlying the results available to ensure full reproducibility?

No

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

No

Are sufficient details of methods and analysis provided to allow replication by others?

No

Reviewer Expertise:

Acute pancreatitis

I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.

References : Classification of acute pancreatitis--2012: revision of the Atlanta classification and definitions by international consensus. Gut .2013;62(1) : 10.1136/gutjnl-2012-302779 102-11 23100216 10.1136/gutjnl-2012-302779