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

Abstract

Aim: 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.
Material and Methods: 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.
Results: 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 (0.756–0.973), followed very
closely by APACHE II score with an AUC of 0.863 (0.758–0.968). 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.
Conclusion: 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.

Full text

DOI: 10.5604/01.3001.0015.8384 POL PRZEGL CHIR 2022: 94: 1-7 AHEAD OF PRINT
original article
1
Comparison of modified Glasgow-Imrie, Ranson,
and Apache II scoring systems in predicting the severity
of acute pancreatitis
Rohit Chauhan
ABCE
, Neeraj Saxena
AD
, Neeti Kapur
F
, Dinesh Kumar Kardam
D
Department of General & Minimally Invasive Surgery, Atal Bihari Vajpayee Institute of Medical Sciences & Dr. Ram Manohar Lohia
Hospital, New Delhi, India
Article history: Received: 12.08.2021 Accepted: 28.04.2022 Published: 02.05.2022
ABSTRACT: Aim: 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.
Material and Methods: Ater 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.
Results: 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 (0.756–0.973), followed very
closely by APACHE II score with an AUC of 0.863 (0.758–0.968). 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-of value of 3 had more complications and a longer hospital stay.
Conclusion: 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.
KEYWORDS: APACHE II, Glasgow-Imrie, prognostic scores, Ranson, severe acute pancreatitis
ABBREVIATIONS
ANC – Acute necrotic collection
AP – Acute Pancreatitis
APACHE II – Acute Physiology and Chronic Health Evaluation II
APFC – Acute peripancreatic fluid collection
AUC – Area Under the Curve
CECT – Contrast-Enhanced Computed Tomography
HDU – High Dependency Unit
ICU – Intensive Care Unit
NPV – Negative Predictive Value
PPV – Positive Predictive Value
ROC – Receiver Operating Characteristic
USG – Ultrasonography
INTRODUCTION
Acute Pancreatitis (AP) is an inflammation of the pancreatic and
peripancreatic tissue with the clinical course ranging from amild,
self-limiting disease in most patients to severe, multiple organ
dysfunction in very few [1–3]. AP occurs due to an abnormal ac-
tivation of pancreatic enzymes resulting in the autodigestion of
the pancreatic parenchyma [4]. is leads to alocal as well as sys-
temic inflammatory response. ere is arelease of pro-inflam-
matory cytokines and anti-inflammatory mediators. All these
mediators cause increased permeability and damage to the mi-
crocirculation of the pancreas [5]. e cascade of inflammation
is self-limiting in approximately 80–90% of all patients. However,
in the remaining few, there is amassive release of inflammatory
mediators into the systemic circulation leading to amultiple organ
dysfunction syndrome and rarely, death of the patient [1].
Confirmation of AP is done by history, clinical findings, and raised
levels of pancreatic enzymes in the plasma. Rise of amylase or li-
pase of more than 3 times its normal levels is confirmatory of the
diagnosis of AP [6]. To evaluate the pancreas, contrast-enhanced
computed tomography (CECT) is the best modality for imaging,
especially for the assessment of complications such as sterile or
infected peripancreatic fluid collections, pancreatic necrosis, pan-
creatic pseudocyst, pancreatic-pleural fistulas, and vascular com-
plications [7–9]. Surgical intervention may sometimes be needed
for these complications. Image-guided aspiration or necrosec-
tomy may be performed in infected pancreatic necrosis. Surgi-
cal debridement and drainage may also be needed in pancreatic
abscesses if it fails to respond to percutaneous catheter drainage
and antibiotics. Pseudocysts may rarely require drainage by lapa-
roscopic and endoscopic means [10].
e variable course of the disease ranging from mild AP to severe
AP with ahigh rate of mortality in the severe form, necessitates
early and accurate prediction of severity to strategize its manage-
ment [1]. Athorough assessment of the severity of disease is also
important to predict prolonged hospitalization, complications, and
to prevent mortality.
e severity of AP was divided into 3 categories by the Revised Atlan-
ta classification (2012) [11]. When organ failure or local or systemic
Authors’ Contribution:
A – Study Design
B – Data Collection
C – Statistical Analysis
D – Data Interpretation
E – Manuscript Preparation
F – Literature Search

WWW.PPCH.PL
2
original article
Invasive Surgery at Atal Bihari Vajpayee Institute of Medical Scienc-
es & Dr. Ram Manohar Lohia Hospital, New Delhi, India between
the 1st November 2018 and 31st March 2020.
Inclusion criteria
All patients aged > 18 years who presented to our centre and di-
agnosed as AP were included in the study. e diagnostic criteria
were the presence of any 2 out of the following three criteria – (a)
history of pain in the abdomen radiating to the back and relieved
on bending forward associated with tenderness/guarding in the
upper abdomen, (b) elevation over 3 times the upper normal limit
of serum amylase (normal range – 30 to 110 U/L)/serum lipase
(normal range – 23 to 300 U/L), and (c) radiographic evidence
– USG (Ultrasonography) or CECT findings suggestive of AP
such as pancreatic oedema, pancreatic necrosis, peripancreatic
fluid collections.
Exclusion criteria
e exclusion criteria were – (a) recurrent AP, (b) chronic pan-
creatitis (± calcific), and (c) patients not willing to participate in
the study.
Atotal of 70 patients diagnosed with AP were included in the study
after taking an informed and written consent from the patient.
Adetailed history was taken and clinical examination was carried
out as per written proforma. All the biochemical parameters were
complications were absent, it was classified as mild AP. Transient
organ failure that resolved within 48 h was classified as moderately
severe AP. Persistent organ failure for ≥ 48 h was termed severe AP.
Several prognostic markers have been developed for severity grad-
ing in AP. Multifactorial scoring systems that take into account
clinical and biochemical criteria for severity stratification have been
used for the past many years. ese include the criteria described
by Ranson in the 1970s [12, 13], the Acute Physiology and Chronic
Health Evaluation (APACHE II) score in 1981 [1, 14], and the modi-
fied Glasgow-Imrie score in 1984 [15]. Various studies have been
conducted in the past to identify the best predictor of severity in
AP but with conflicting results and thus no ideal single method for
assessing the severity in AP.
is study was conducted to find out ascoring system that provides
an early as well as accurate prediction of the severity of the disease.
Aprognostic score that is easy to calculate and convenient to use for
practical purposes, is desired. e scores were evaluated for their
assessment of severity, complication rates, mortality and length of
hospital stay in case of AP.
MATERIAL AND METHODS
With the approval of the Ethics Committee, aprospective obser-
vational study was carried out on patients who were clinically sus-
pected to have AP, in the Department of General and Minimally
Fig. 1. Demographic features and incidence of severe disease, mortality and complications.
0
5
10
15
20
25
20-30 30-40 40-50 50-60 60-70 >70
NUMBER OF PATIENTS
AGE GROUP (IN YEARS)
Male Female Incidence of severe disease Complications Mortality

3
POL PRZEGL CHIR 2022: 94: 1-7 AHEAD OF PRINT
original article
noted at the time of admission and after 48 hours of admission.
Severity of the disease was classified as per Atlanta Classification at
48 hours after admission. e modified Marshall score was used to
assess organ failure [16, 17]. CECT of the abdomen was done after
72 hours of admission in all patients with moderate or severe AP.
e modified Glasgow-Imrie score, Ranson score and APACHE II
score were calculated. e outcome measures were severity, mor-
tality, complications and length of hospital stay. All patients were
followed up in the outpatient department at 3 months.
Statistics
e data was entered in MS EXCEL spreadsheet and analysis was
done using Statistical Package for Social Sciences (SPSS) version
21.0. Categorical variables were presented in numbers and percent-
ages (%) and continuous variables were presented as mean ± SD and
median. e scores were correlated with the outcomes, that is, se-
verity, mortality, complications and length of hospital stay statisti-
cally using two-by-two contingency tables, odds ratio, Chi-square
test, Fisher’s Exact test & Mann-Whitney-Utest. AP-value of < 0.05
was considered statistically significant. Predictive accuracy of each
score was determined by AUC (Area Under the Curve) in the ROC
(Receiver Operating Characteristic) analysis.
RESULTS
Epidemiology
e mean and median age of the patients included was 41.6 years
and 40 years respectively. e oldest patient was 72 years old and
the youngest person was 23 years old. ere were 62.8% (n = 44) of
patients in the range of 30–50 years. Afemale predominance with
a1.41:1 ratio i.e. 58.6% females (n = 41) and 41.4% males (n = 29)
was observed. e mean age was 40.7 years in males and 42.2 years
in females. e most common aetiology was gall stones (77.14%),
followed by alcohol intake (17.14%). e cause for the rest 5.71%
of patients was idiopathic.
The incidence of severe disease in the study group was 34.3%
(n = 24) as calculated by the Revised Atlanta criteria [11]. e remain-
ing, 65.71% (n = 46) of the patients had amild form of AP. e compli-
cations that were encountered were acute peripancreatic fluid collec-
tion (APFC) and acute necrotic collection (ANC) with an incidence of
28.6% (n = 20). Of the 15 patients with APFC, 80% (n = 12) had aspon-
taneous resolution of the collection at the 3-month follow-up. Only
3 patients had apersistent fluid collection in the form of pseudocyst. All
3 underwent surgical intervention in the form of internal drainage. Four
out of the 5 patients with ANC died. Only 1 patient with awalled-off
necrosis underwent surgical intervention after his 3-month follow-up.
Surgical intervention was done for persistent symptoms in only 5.7%
(n = 4) of patients, all of whom had severe AP.
Patients were also stratified by the type of organ failure according
to the modified Marshall scoring system [16, 17]. Out of the 24 pa-
tients who had severe AP, 17 patients had only asingle organ sys-
tem failure and 7 had multiple organ failure persistent beyond 48
hours. Majority of patients with persistent single organ failure had
an acute kidney injury reflected by their raised serum creatinine.
e mean and median length of hospital stay was 6.2 days (+ 2.9
days) and 5 days respectively. e mean length of hospital stay was
SCORE PATIENTS WITH AP % (N) SEVERE AP % (N) MORTALITY % (N) COMPLICATIONS % (N) MEAN LENGTH OF HOSPITAL
STAY ( DAYS )
Ranson Score
<3 72.9 (51) 10 (7) - 11.4 (8) 5.46
≥3 27.1 (19) 24.3 (17) 7.1 (5) 17.1 (12) 8.10
Total 100 (70) 34.3 (24) 7.1 (5) 28.6 (20) 6.21
Glasgow-Imrie Score
<3 72.9 (51) 8.6 (6) - 12.9 (9) 5.77
≥3 27.1 (19) 25.7 (18) 7.1 (5)15.7 (11)7.59
Total 100 (70) 34.3 (24)7.1 (5) 28.6 (20) 6.21
APACHE II Score
<8 68.6 (48) 7.1 (5) 1.4 (1) 10 (7) 5.13
≥8 31.4 (22) 27.2 (19) 5.7 (4) 18.6 (13) 8.43
Total 100 (70) 34.3 (24) 7.1 (5) 28.6 (20) 6.21
SENSITIVITY SPECIFICITY PPV NPV ACCURACY
Ranson Score 70.83 95.65 89.47 86.27 87.14
Glasgow-Imrie Score 75 97.83 94.74 88.24 90.00
APACHE II 79.17 93.48 86.36 89.58 88.57
Tab. I. Incidence of severe AP, mortality, complications, and length of hospital stay stratified among diferent scoring systems.
Tab. II. Sensitivity, Specificity, PPV, NPV, and Accuracy of diferent scoring systems for severe AP.
PPV – Positive Predictive Value; NPV – Negative Predictive Value. All values are expressed in percentage.

WWW.PPCH.PL
4
original article
e sensitivity, specificity, PPV (Positive Predictive Value), NPV
(Negative Predictive Value) and accuracy for predicting the
incidence of severe disease is depicted in Tab. II. It was observed
that APACHE II score had the highest sensitivity, of 79.17%, in
terms of the incidence of severe disease. Ranson and Glasgow-
-Imrie scores had asensitivity of 70.83% and 75% respectively.
Among the three scores, Glasgow-Imrie had the highest specific-
ity, of 97.83%. Ranson and APACHE II had aspecificity of 95.65%
and 93.62% respectively.
In Tab. III., the degree of association between the outcomes and the
categories of the scoring systems has been shown. e Chi-square
test showed that the incidence of severe disease was higher with all
the three scoring systems when avalue higher than the cut-off was
attained. Glasgow-Imrie score had the strongest correlation in pre-
dicting the severity of AP with an odds ratio of 135. Similar observa-
tions were made in terms of mortality and complications. As there
was no mortality when the score was less than the cut-off value, the
odds ratio for Ranson and Glasgow-Imrie scores could not be calcu-
lated. However, by Fisher’s exact test there was significant evidence
(P < 0.05) of higher mortality in the group with ascore above the
cut-off values. APACHE II score had an odds ratio of 10.44 and P
= 0.031 (significant) by Fisher’s exact test in predicting mortality.
In predicting complications, acomparable odds ratio of 9.21, 6.42,
and 8.46 was found for the Ranson, Glasgow-Imrie, and APACHE
II scores respectively, indicating that the three scores were almost
equivalent in predicting complications.
From the results of the Mann-Whitney Utest, it could be conclud-
ed that the mean length of hospital stay was significantly higher for
patients with ascore above the cut-off value.
longer in patients with asevere disease (8.1 days + 3 days) as com-
pared to those with amild form of the disease (5.2 days + 2 days).
ere was amortality of 7.14% (n = 5) and in all cases where mor-
tality was confirmed, the disease was of severe type. e mean age
of mortality was 57.6 (+ 10.7) years. Fig. 1. shows the distribution of
the three outcomes – severity, mortality and complications – strati-
fied by age groups and gender.
Prognostic scores
As shown in Tab. I., it was observed that out of all patients, 34.3%
(n = 24) had asevere form of the disease. In this subset, 24.3%
(n = 17) had aRanson score above its cut-off value, that is ≥3.
Similarly, 25.7% (n = 18) had aGlasgow Imrie score ≥3 and 27.2%
(n = 19) of the patients had an APACHE II score ≥ 8, i.e. above their
respective cut-off values.
Mortality as an outcome was observed in 7.1% (n = 5) of the cases.
All 5 cases had aRanson and Glasgow-Imrie score of more than their
cut-off values of 3. Out of these 5 cases, only 4 had an APACHE II
score above its cut-off value of 8.
Complications were observed in 28.6% (n = 20) of the cases where
17.1% (n = 12) had aRanson score ≥3, 15.7% (n = 11) had aGlasgow-
-Imrie score ≥3, and 18.6% (n = 13) had an APACHE II score ≥8.
e mean length of hospital stay was found to be 6.2 days (3.3 to 9.1
days) with alonger length of stay observed in patients with ascore
above the cut-off value for each scoring system.
Tab. I. depicts the cross-tabulations among the scoring systems and
the outcomes.
SCORES SEVERITY MORTALITY COMPLICATIONS LENGTH OF STAY
Ranson 53.43a**** 9.21a6 versus 8 days
(P < 0.0001) b(P = 0.001) c(P = 0.0009) b(P < 0.001) d
Glasgow- 135a**** 6.42a6 versus 8 days
-Imrie Score (P < 0.0001) b(P = 0.001) c(P = 0.0009) b(P = 0.020) d
APACHE II 54.47a10.44a8.46a5 versus 8 days
(P < 0.0001) b(P = 0.031) c (P = 0.0001) b(P < 0.001) d
AUC (% CI) SEVERITY MORTALITY COMPLICATIONS
Ranson Score 0.832a (0.716–0.949)b0.892 (0.809–0.975) 0.730 (0.588–0.872)
(0.059)c(0.044) (0.073)
Glasgow-Imrie Score 0.864 (0.756–0.973) 0.892 (0.809–0.975) 0.695 (0.549–0.841)
(0.055) (0.108) (0.075)
APACHE II 0.863 (0.758–0.968) 0.762 (0.547–0.976) 0.735 (0.596–0.874)
(0.054) (0.110) (0.071)
Tab. III. Extent of correlation of scores with the outcomes.
Tab. IV. AUC of diferent scores in predicting severity, mortality, and complications.
a Odds ratio; b Chi-square test; c Fisher's exact test; d Mann-Whitney test; significance was tested at 5%.**** Odds ratio not calculated.
a Area under the curve; b Confidence Interval; c Standard Error.

5
POL PRZEGL CHIR 2022: 94: 1-7 AHEAD OF PRINT
original article
In the prediction of the severity of the disease according to AUC
(with 95% Confidence Interval) in the ROC curve, it was found
that Glasgow-Imrie score had an AUC of 0.864 (0.756– 0.973), fol-
lowed very closely by APACHE II score with an AUC of 0.863 (0.758
–0.968). Ranson score had an AUC of 0.832 (0.716–0.949). Similarly,
the ROC curves were plotted for mortality and complications, and
their AUC was calculated as depicted in Table 4.
e area under the ROC curve (AUC) is an indicator of the prob-
ability of correct or accurate prediction by the test of severity, mor-
tality, complications. An AUC of 1 represents aperfect test whereas
an AUC of 0.5 represents aworthless test.
Fig. 2. depicts different ROC curves. Since the incidence of
mortality above the cut-off values for Ranson and Glasgow-
Imrie score are equal, the ROC curves overlap each other,
and thus have an equal AUC of 0.892 (0.809–0.975). APACHE
II had an AUC of 0.762 (0.547–0.976) which is less than the
other 2 scoring systems.
DISCUSSION
e wide variability in the clinical course of AP ranging from amild
form to multiple organ failure and death led to the development
of various prognostic scores to assess the disease severity. In this
study, an analysis of the scoring systems against the outcomes was
done to tag the most appropriate scoring system for predicting the
outcomes studied.
APACHE II score had the highest sensitivity – of 79.17% – in pre-
dicting the severity of AP followed by Glasgow-Imrie and Ranson
scoring systems with asensitivity of 75% and 70.83% respectively
(Tab. II.). e high sensitivity of APACHE II could be attributed to
the greater number of physiological parameters needed to calculate
it, as compared to the other two scores. APACHE II score also had
avery high accuracy as well as NPV making it suitable to rule out
asevere form of AP rather than predicting it. Serial calculation of
APACHE II at regular intervals would probably make it even more
sensitive and specific to predict the severity in AP. Marco Simoes
et al. reported asimilar high sensitivity (79.4%), NPV (91.2%) and
AUC (0.861) for APACHE II score [18].
Glasgow-Imrie score was the most specific of all the scores. e
difference in the sensitivity of Glasgow-Imrie score and APACHE
II was not high. e odds of having severe AP were high when the
score was higher than the cut-off value, as depicted by asignifi-
cantly high odds ratio. Although the AUC for Glasgow-Imrie score
was the highest of all three scores, it was comparable to the AUC
of APACHE II score. Both scores were equally good predictors of
severity of pancreatitis but the difference in the ability of the two
scores i.e., Glasgow-Imrie and APACHE II, to predict the severity
of disease was negligible.
e two scores were thus comparable to each other in predicting
the severity of disease. However, the ease of calculation of Glasgow-
Imrie score makes it amore favourite choice. Ranson score was the
least useful in predicting the severity of AP in our study. e find-
ings were consistent with the study by Savio G. Barreto in which
the author concluded that APACHE II and Glasgow-Imrie scores
were comparable to each other in predicting the severity of AP [19].
A
B
C
Fig. 2. ROC curves for predicting (A) incidence of severe AP, (B) mortality, and (C)
complications.
ROC Curve
1 - Specicity
1 - Specicity
1 - Specicity
Source of the Curve
RANSON SCORE
GLASGOW-IMRIE SCORE
APACHE II SCORE
Reference Line
SensitivitySensitivitySensitivity
Source of the Curve
RANSON SCORE
GLASGOW-IMRIE SCORE
APACHE II SCORE
Reference Line
Source of the Curve
RANSON SCORE
GLASGOW-IMRIE SCORE
APACHE II SCORE
Reference Line
ROC Curve
ROC Curve

WWW.PPCH.PL
6
original article
is result was congruous with the studies of Marco Simoes et al.
[18] and Ajay Khanna [20].
ere has been asignificant advancement in imaging techniques
such as CECT and endoscopic ultrasound which can help in the
assessment of disease accurately. However, the availability of such
methods of imaging is limited to tertiary care centres in metro-
politan areas only. e scoring systems can prove to be helpful in
planning the management of acute pancreatitis in acountry like
ours where the majority of the population resides in rural areas
with limited access to affordable healthcare. e data from our
study exhibit that the Glasgow-Imrie and APACHE II scoring sys-
tems are comparable to each other in predicting the severity of AP.
Ranson score lags behind the other two scores in predicting the
severity of the disease. However, regarding mortality, Glasgow-
Imrie and Ranson scores were equally capable of, and better than
APACHE II score in predicting the outcome. All the three scores
were similar to each other in predicting complications in patients
with AP. Ascore above the cut-off values in each scoring system
was predictive of asignificantly prolonged length of hospital stay.
CONCLUSION
AP can prove to be acritical and life-threatening disease which re-
quires careful consideration in its management. For the prediction
of severe disease, mortality, and complications in patients, various
scores such as the Ranson, Glasgow-Imrie, and APACHE II were
used. All the scores had apositive correlation with the outcomes
in our study. Both Glasgow-Imrie and APACHE II scores were
comparable to each other in predicting the incidence of severe
disease. Ranson score was less sensitive and accurate compared to
the other two scores. e calculation of APACHE II score is based
on alarge number of parameters and is in itself cumbersome to
calculate. Also, APACHE II score was initially designed to prog-
nosticate any patient admitted to the intensive care unit and not
for AP specifically, in contrast to Glasgow-Imrie score which was
specifically fashioned for AP.
Glasgow-Imrie score is, therefore, recommended for predicting
severe AP. e score is sensitive and specific enough to use it as an
indicator of asevere form of the disease. Asimple and accurate pre-
diction of severity will help in proper management of patients with
acute pancreatitis, prevent adverse events such as mortality and
make satisfactory use of hospital resources such as HDU and ICU.
REFERENCES
1. Mederos M.A., Reber H.A., Girgis M.D.: Acute Pancreatitis: A Review. JAMA,
2021; 325(4): 382–390. doi: 10.1001/jama.2020.20317.
2. Portelli M., Jones C.D.: Severe acute pancreatitis: pathogenesis, diagnosis and
surgical management. Hepatobiliary Pancreat Dis Int., 2017; 16(2): 155–159. doi:
10.1016/s1499-3872(16)60163-7. PMID: 28381378.
3. Lee P.J., Papachristou G.I.: New insights into acute pancreatitis. Nat Rev Gastro-
enterol Hepatol., 2019; 16(8): 479–496. doi: 10.1038/s41575-019-0158-2. PMID:
31138897.
4. Norman J.: Role of cytokines in the pathogenesis of acute pancreatitis. Am J Surg ,
1998; 1(175): 76–83.
5. Yadav D., Lowenfels A.B.: e epidemiology of pancreatitis and pancreatic cancer.
Gastroenterology, 2013; (6)144: 1252–1261.
6. Alter D., Koch D.D.: A Review of Acute Pancreatitis. JAMA, 2021; 325(23): 2402.
doi: 10.1001/jama.2021.6012.
7. Ortiz Morales C.M., Girela Baena E.L., Olalla Muñoz J.R., Parlorio de Andrés E.,
López Corbalán J.A.: Radiology of acute pancreatitis today: the Atlanta classifi-
cation and the current role of imaging in its diagnosis and treatment. Radiologia
(Engl Ed), 2019; 61(6): 453–466. English, Spanish. doi: 10.1016/j.rx.2019.04.001.
Epub 2019 May 29. PMID: 31153603.
8. Aghdassi A.A., S eidensticker M.: Bildgebende Diagnostik bei akuter Pankreatitis
[Imaging diagnostics in acute pancreatitis]. Internist (Berl), 2021; 62(10): 1044–1054.
German. doi: 10.1007/s00108-021-01153-3. Epub 2021 Sep 15. PMID: 34524469.
9. Brizi M.G., Perillo F., Cannone F., Tuzza L., Manfredi R.: e role of imaging in
acute pancreatitis. Radiol Med, 2021; 126(8): 1017–1029. doi: 10.1007/s11547-021-
01359-3. Epub 2021 May 12. PMID: 33982269; PMCID: PMC8292294.
10. Tyberg A., Karia K., Gabr M. et al.: Management of pancreatic fluid collections:
A comprehensive review of the literature. World J Gastroenterol, 2016; 22(7):
2256–2270. doi: 10.3748/wjg.v22.i7.2256. PMID: 26900288; PMCID: PMC4735000.
In predicting mortality, APACHE II score had asignificantly high
odds ratio of 10.44 predicting ahigher odd of mortality in patients
with ascore above the cut-off value. e odds ratio could not be
calculated for Ranson and Glasgow-Imrie scores as there was no
mortality in patients with ascore less than the cut-off value. Ran-
son and Glasgow-Imrie score had an equal AUC which was more
than the AUC of APACHE II. An equal AUC for both the scores
suggested that both scores were equally capable of predicting mor-
tality in AP. Ajay Khanna reported asimilar AUC for Glasgow-
Imrie and Ranson score in predicting mortality, however the au-
thor found that APACHE II had agreater AUC than the other 2
scores [20]. e modest difference in this finding could be due to
the small sample size in both studies.
Complications such as APFC and ANC were seen in 28% of pa-
tients. e odds ratio for predicting complications was 9.21, 6.42,
and 8.46 for the Ranson, Glasgow-Imrie, and APACHE II scoring
systems respectively. All values were significant with aP < 0.05 de-
noting ahigher chance of having complications in patients with
ascore above the cut-off value. e difference in AUC of the three
scores was also not substantial. is finding related to predicting
complications concurs with the finding of Ajay Khanna [20]. e
AUC for the three scores regarding complications in the same or-
der was found to be 0.70, 0.64, 0.68 by Ajay Khanna in his study.
us, each score was at par with each other in predicting compli-
cations. In the study by Marco Simoes et al., the author found no
significant correlation between the prognostic scores and inci-
dence of complications [18]. e difference in the etiological fac-
tors leading to the difference in the occurrence of complications
could be the cause for this digression.
e mean length of hospital stay in patients with AP was found
to be shorter as compared to studies by Marco Simoes et al. [18]
and Ajay Khanna [20]. Both authors found amean length of stay
of 10 days. In patients with severe AP, the values for the length of
hospital stay were falsely low in our study. e data was skewed
towards the lower value as all the deaths in severe AP took place
within 4 to 6 days of admission. Also, most patients in our study
with severe AP had asingle organ failure in the form of acute kid-
ney injury which responded well within aweek to adequate re-
suscitation. However, this outcome could vary in different stud-
ies depending upon the level of care, monitoring, as well as the
management conducted by the health-care staff of the hospi-
tal. e length of hospital stay was significantly prolonged for all
the three systems when the score was above their cut-off values.

7
POL PRZEGL CHIR 2022: 94: 1-7 AHEAD OF PRINT
original article
11. Banks P.A., B ollen T.L ., Dervenis C. et al.: Acute Pancreatitis Classification Wor-
king Group. Classification of acute pancreatitis 2012: revision of the Atlanta clas-
sification and definitions by international consensus. Gut, 2013; 62(1): 102–111.
doi: 10.1136/gutjnl-2012-302779. Epub 2012 Oct 25. PMID: 23100216.
12. Ranson J.H., Rifkind K.M., Roses D.F., Fink S.D., Eng K., Spencer F.C.: Prognostic
signs and the role of operative management in acute pancreatitis. Surg Gynecol
Obstet, 1974; 139(1): 69–81.
13. Ranson J.H.: e timing of Índices de Gravidade. Ann Surg, 1979; 189(5): 654–663.
14. Knaus W.A ., Draper E.A., Wagner D.P., Zimmerman J.E .: APACHE II: a severity of
disease classification system. Crit Care Med, 1985; 13(10): 818–829.
15. Blamey S.L., Imrie C.W., O’Neill J., Gilmour W.H., Carter D.C.: Prognostic factors
in acute pancreatitis. Gut, 1984; 25(12): 1340–1346.
16. Marshall J.C., Cook D.J., Christou N.V. et al.: Multiple organ dysfunction score:
a reliable descriptor of a complex clinical outcome. Crit Care Med, 1995; 23 (10):
1638–1652.
17. Siregar G.A., Siregar G.P.: Management of Severe Acute Pancreatitis. Open Access
Maced J Med Sci., 2019; 7(19): 3319–3323. doi: 10.3889/oamjms.2019.720. PMID:
31949538; PMCID: PMC6953950.
18. Simoes M., Alves P., Esp erto H. et al.: Predicting Acute Pancreatitis Severity: Com-
parison of Prognostic Scores. Gastroenterology Re search, 2011; 4(5): 216–222.
19. Barreto S.G., Rodrigues J.: Comparison of APACHE II and Imrie Scoring Systems
in predicting the severity of Acute Pancreatitis. World J Emerg Surg, 2007; 2: 33.
20. Khanna A.K., Meher S., Prakash S. et al.: 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; 2013: 367581.
https://ppch.pl/resources/html/articlesList?issueId=0 Page of count: 7 Tables: 4 Figures: 2 References: 20
Some right reserved: Fundacja Polski Przegląd Chirurgiczny. Published by Index Copernicus Sp. z o. o.
The authors declare that they have no competing interests.
The content of the journal „Polish Journal of Surgery” is circulated on the basis of the Open Access which means free
and limitless access to scientic data.
This material is available under the Creative Commons – Attribution-NonCommercial 4.0 International (CC BY-NC 4.0).
The full terms of this license are available on: https://creativecommons.org/licenses/by-nc/4.0/legalcode
Dr. Rohit Chauhan (ORCID: 0000-0002-7998-0827); Department of General & Minimally Invasive Surgery, Atal Bihari
Vajpayee Institute of Medical Sciences & Dr. Ram Manohar Lohia Hospital, New Delhi, India; Phone: +91 9650065206;
E-mail: rohitchauhan93@yahoo.com
Chauhan R., Saxena N., Kapur N., Kardam D.K.: Comparison of modied Glasgow-Imrie, Ranson, and Apache II scoring systems
in predicting the severity of acute pancreatitis; Pol Przegl Chir 2022; 94: (1–7); DOI: 10.5604/01.3001.0015.8384
Table of content:
Copyright:
Competing interests:
Corresponding author:
Cite this article as: