Assessment of the Severity of Acute Pancreatitis

Paraskevas S Brestas
1
, Urania G Dafni
2
1
Department of Radiology, General Hospital of Athens.
2
Division of Public Health, Laboratory of
Biostatistics, Department of Nursing, University of Athens. Athens, Greece
Dear Sir:
We read the original article regarding
assessment of the predictive value of the CT
severity index (CTSI) for the severity of acute
pancreatitis by Gürleyic et al. with great
interest [1]. The purpose of this study was to
make a comparative assessment of the
accuracy of the CTSI proposed by Baltazar et
al. [2], the APACHE II score and serum CRP
concentrations in predicting the severity of
acute pancreatitis. The cut-off values of 3 for
the CTSI and 7 for the APACHE II were used
in the present study, based on the results of
studies which did not have exactly the same
discrimination endpoint [2, 3]. The
discrimination endpoint in this study is
defined as the ability to separate those
patients who had mild pancreatitis from those
who had severe pancreatitis, according to the
classification criteria of the 1992 Atlanta
International Symposium [4].
Receiver operating characteristic (ROC)
curves could have been used to determine the
most appropriate cut-off point for the selected
discrimination endpoint which corresponds to
the best possible trade-off between sensitivity
and specificity which were estimated in the
present sample. Moreover, the area under the
ROC curve is a reliable measure of overall
predictive discrimination and a previously
described method for comparison of the areas
under the ROC curves, derived from the same
cases, could also have been used [5, 6].
The additional information provided by ROC
curves in studies of prognostic indices of
acute pancreatitis severity derives from the
complete illustration of the relationship
between sensitivity and specificity for a
certain discrimination endpoint (severe vs.
mild pancreatitis in this case). This might
have been useful because: a) the clinical
impact of the two types of misclassification
(failure to correctly identify a case of severe
pancreatitis or failure to correctly identify a
case of mild pancreatitis) is not the same, and
b) it is necessary to realize new prospective
comparative studies for assessing the clinical
impact of promising imaging techniques, such
as MRI or contrast-enhanced US in the near
future [7, 8].
Moreover, it should be noted that even though
the Atlanta classification system provides a
reliable basis for experimental studies for the
clinical management of acute pancreatitis, it is
not considered to be a perfect system since
intermediate forms of the disease do occur
[9]. If an imperfect gold standard is used, the
estimated accuracy of the tests may suffer
(“imperfect gold standard bias”). Another
type of bias affecting ROC analysis is
“verification bias” and takes place if some of
the patients with test results do not have
verified disease status or if the decision to

Page 2
JOP. J Pancreas (Online) 2006; 7(2):245-248.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 7, No. 2 - March 2006. [ISSN 1590-8577]
246
verify a patient is influenced by the test
results. Calculation of the accuracy of a
diagnostic test using standard definitions
unavoidably includes the risk of some kind of
bias under certain circumstances. ROC
analysis offers the possibility of bias-
correction methods [10] and methods of non-
parametric estimation of ROC curves have
been also suggested recently in the case in
which the gold standard is not binary or in the
absence of a gold standard, [11, 12]. Thus,
ROC analysis should be the preferred method
for the assessment of the predictive value of
imaging techniques.
Received December 14
th
, 2005
Keywords
Inflammation;
Pancreas;
Pancreatitis, Acute Necrotizing; ROC Curve
Abbreviations CTSI: CT severity index
Correspondence
Paraskevas S Brestas
Ag. Marinis 21, Melissia
15127 Athens
Greece
Phone: +30-210.804.6453/4120, +30-
697.272.9154
Fax: +30-210.804.4120
E-mail: pbrestas@yahoo.gr
References
1. Gurleyik G, Emir S, Kilicoglu G, Arman A,
Saglam A. Computed tomography severity index,
APACHE II score, and serum CRP concentration for
predicting the severity of acute pancreatitis. JOP. J
Pancreas (Online) 2005; 6:562-7. [PMID 16286706]
2. Balthazar EJ, Robinson DL, Megibow AJ, Ranson
JH. Acute pancreatitis: value of CT in establishing
prognosis. Radiology 1990; 174:331-6. [PMID
2296641]
3. Larvin M, McMahon MJ. APACHE-II score for
assessment and monitoring of acute pancreatitis.
Lancet 1989; 2:201-5. [PMID 2568529]
4. Bradley EL 3rd. 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; 128:586-90. [PMID 8489394]
5. Hanley JA, McNeil BJ. The meaning and use of
the area under a receiver operating characteristic
(ROC) curve. Radiology 1982; 143:29-36. [PMID
7063747]
6. Hanley JA, McNeil BJ. A method of comparing
the areas under receiver operating characteristic curves
derived from the same cases. Radiology 1983;
148:839-43. [PMID 6878708]
7. Pezzilli R, Fantini L. The imaging assessment of
the severity of acute pancreatitis may change in the
near future. JOP. J Pancreas (Online) 2005; 6:467-9.
[PMID 16186671]
8. Brocchi E, Piscaglia F, Bonora M, Celli N, Venturi
A, Fantini L, et al. Echo-enhanced ultrasonography: is
it the future gold standard of imaging in acute
pancreatitis? JOP. J Pancreas (Online) 2005; 6:464-6.
[PMID 16186670]
9. Balthazar EJ. Acute pancreatitis: assessment of
severity with clinical and CT evaluation. Radiology
2002; 223:603-13. [PMID 12034923]
10. Zhou XH. Correcting for verification bias in
studies of a diagnostic test's accuracy. Stat Methods
Med Res. 1998; 7:337-53. [PMID 9871951]
11. Obuchowski NA. Estimating and comparing
diagnostic tests' accuracy when the gold standard is not
binary. Acad Radiol 2005; 12:1198-204. [PMID
16099683]
12. Zhou XH, Castelluccio P, Zhou C. Nonparametric
estimation of ROC curves in the absence of a gold
standard. Biometrics 2005; 61:600-9. [PMID
16011710]
REPLY
Dear Sir:
We would like to thank Brestas et al. for their
interest in our article published in JOP. J
Pancreas (Online) [1]. In this study, The
results of an imaging method (computed
tomography), a clinical scoring system
(APACHE II) and a biochemical measure-
ment (serum CRP concentration) were
evaluated in a group of patients with acute
pancreatitis. The accuracy of these variables
for predicting the clinical course of the
disease was calculated. This had initially been

Page 3
JOP. J Pancreas (Online) 2006; 7(2):245-248.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 7, No. 2 - March 2006. [ISSN 1590-8577]
247
classified using the Atlanta criteria. Brestas et
al. presented their opinions and suggestions
mainly on two subjects:
- the cut-off values of the computed
tomography severity index (CTSI) and
APACHE II;
- evaluation of the results using the receiver
operating characteristics (ROC) curve.
We chose these cut-off values because of our
personal experience [2] on acute pancreatitis
cases (length of hospital stay, morbidity and
mortality, etc.) and the results of previous
studies. Larwin and McMahon [3] found that
patients with an APACHE II score greater
than 7 were likely to have a severe clinical
course while Connor et al. [4] reported
increasing mortality in patients with
APACHE II scores greater than 7. Balthazar
et al. [5] reported that acute pancreatitis cases
with a CTSI of 0 to 2 points had 2%
morbidity and no mortality while morbidity
and mortality rates for patients with a CTSI
ranging from 0 to 3 were 8% and 3%, those
with a CTSI of 4-6 were 35% and 6% and
those with a CTSI of 7-10 were 92% and
17%, respectively. Therefore, aiming to have
a lower mortality rate for our patients, we
preferred to use a CTSI score of 3 and an
APACHE II value of 7 as our cut-off points.
The variables of our study were compared by
valid statistical methods. The sensitivity,
specificity, positive and negative predictive
values and accuracy were also calculated for
each variable. We agree with Brestas et al.
that the ROC curve could have been used,
especially in assessment of the imaging
technique. Thus, we applied the test to our
results.
The area under the ROC curve (AUC±SE
evaluated by SPSS 10.0) for CTSI was
0.969±0.024, clearly supportive of the high
accuracy of this index in predicting the
severity of acute pancreatitis (Figure 1). A
value of the CTSI equal to 3 was identified as
the best cut-off using the procedure proposed
by Pezzilli et al. [6] (the maximum likelihood
ratio, LR, was 10.3). This value corresponds
to previously published values of sensitivity
and specificity [1]. The APACHE II gave an
AUC value of 0.812±0.074 and a best cut-off
value of 7 (the maximum LR was 3.6).
Sensitivity, specificity, and frequency of cases
correctly identified by applying this cut-off
are 61.5% (8 out of 13 severe acute
pancreatitis patients), 95.2% (40 out of 42
mild pancreatitis patients), and 87.3% (48 out
of 55 overall acute pancreatitis patients),
respectively.
Finally, it should be noted that the AUC of
the CTSI was significantly higher when
compared with the APACHE II score
(P=0.044, z-test).
In conclusion, the use of ROC curve analysis
confirmed our previous results [1] showing
that values of the CTSI greater than 3 are
highly indicative of severe acute pancreatitis.
On the other hand, the ROC curve applied to
our data indicates values of an APACHE II
score greater than 7 are necessary in order to
identify severe acute pancreatitis patients
instead of values equal to or greater than 7 as
we used in our paper [1]. Moreover, both the
CTSI and the APACHE II scores are highly
accurate in predicting the natural outcome of
Figure 1. Receiver operating characteristics (ROC)
curves of the computed tomography severity index
(CTSI) and APACHE II in distinguishing between mild
and severe acute pancreatitis. Data from Gurleyik et al.
[1]. Red bullets show the best cut-off values. Green
bullets shows the point of the ROC curve
corresponding to the values of previously published
APACHE II score [1].

Page 4
JOP. J Pancreas (Online) 2006; 7(2):245-248.
JOP. Journal of the Pancreas - http://www.joplink.net - Vol. 7, No. 2 - March 2006. [ISSN 1590-8577]
248
specific patients with acute pancreatitis, with
the indicated cut-off values clearly specific
for the purpose. Finally, the CTSI is
significantly more accurate than the APACHE
II score.
Gunay Gurleyik
1
Gamze Kilicoglu
2
1
Department of Surgery and
2
Department of
Radiology, Haydarpasa Numune Education
and Research Hospital. Istanbul, Turkey
Received January 24
th
, 2006
Keywords APACHE; Pancreatitis, Acute
Necrotizing; ROC Curve; Tomography, X-
Ray Computed
Correspondence
Günay Gürleyik
Eski Bagdat cad. 29/9
Altintepe 34840 Istanbul
Turkey
Phone: +90-216.489.8325
Fax: +90-216.373.1096
E-mail: ggurleyik@yahoo.com
References
1. Gurleyik G, Emir S, Kilicoglu G, Arman A,
Saglam A. Computed tomography severity index,
APACHE II score, and serum CRP concentration for
predicting the severity of acute pancreatitis. JOP. J
Pancreas (Online) 2005; 6:562-7. [PMID 16286706]
2. Gurleyik G, Cirpici OZ, Aktekin A, Saglam A.
The value of Ranson and APACHE II scoring systems,
and serum levels of interleukin-6 and C-reactive
protein in the early diagnosis of the severity of acute
pancreatitis. Ulus Travma Derg 2004; 10:83-8. [PMID
15103565]
3. Larvin M, Mc Mahon MJ. APACHE -II score for
assessment and monitoring of acute pancreatitis.
Lancet 1989; 2:201-5. [PMID 2568529]
4. Connor S, Ghaneh P, Raraty M, Rosso E, Hartley
MN, Garvey C, et al. Increasing age and APACHE II
scores are the main determinants of outcome from
pancreatic necrosectomy. Br J Surg 2003; 90:1542-7.
[PMID 14648734]
5. Balthazar EJ, Robinson DL, Megibow AJ, Ranson
JH. Acute pancreatitis: value of CT in establishing
prognosis. Radiology 1990; 174:331-6. [PMID
2296641]
6. Pezzilli R, Billi P, Miniero R, Fiocchi M,
Cappelletti O, Morselli-Labate AM, et al. Serum
interleukin-6, interleukin-8, and beta 2-microglobulin
in early assessment of severity of acute pancreatitis.
Comparison with serum C-reactive protein. Dig Dis Sci
1995; 40:2341-8. [PMID 7587812

There are no products listed under this category.