+ MODEL

Kaohsiung Journal of Medical Sciences (2017) xx, 1e5

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http://www.kjms-online.com

Short Communication

Simplified scoring system for prediction of

mortality in acute suppurative cholangitis
Barıs‚ Do
gu Yıldız, Sabri Özden, Barıs‚ Saylam, Fahri Martlı, Mesut Tez*

Ankara Numune Education and Research Hospital, Ankara, Turkey

Received 10 April 2017; accepted 27 December 2017

KEYWORDS Abstract Our objective in this study was to identify the factors contributing to mortality in
Acute cholangitis; acute suppurative cholangitis which could be tested easily in every emergency clinic. This is a
Suppurative; retrospective study enrolling 104 patients with acute suppurative cholangitis. Demographic
Mortality; and laboratory data were collected for analysis. In univariant analysis red cell distribution
Scoring system; width, total bilirubin level, intensive care unit admission was identified as statistically signif-
Red cell distribution icant (p < 0.05) to predict mortality. Three variables were statistically significant in multivar-
width iate analysis: total bilirubin level equal to or more than 6.9 mg/dl, red cell distribution width
equal to or more than 14.45%, and admission to intensive care unit. We found a new scoring
system for prediction of mortality in acute suppurative cholangitis utilizing only three vari-
ables. This would serve as a simplified, rapid way to direct patients for advanced
interventions instead of wasting time with more complicated and time consuming multi-
variable scoring systems.
Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).

Introduction endoscopic retrograde choangiopancreatography (ERCP) or

Acute cholangitis is an acute inflammation and infection of
the biliary tract. The conventional approach for acute
cholangitis is conservative management involving cessation
of oral feeding, intravenous antibiotics and fluids and
Conflicts of interest: All authors declare no conflicts of interest.
* Corresponding author. Ankara Numune Education and Research
Hospital, Department of General Surgey, Talatpas‚a bulvarı no: 5,
06100 Samanpazarı, Ankara, Turkey.
E-mail address: mesuttez@yahoo.com (M. Tez).
surgical intervention when necessary [1].
Acute cholangitis is graded in severity from grade I to III
(mild, moderate, severe). Severe acute cholangitis is
associated with at least one of cardiovascular, neurological,
respiratory, renal, hepatic and/or hematological dysfunc-
tion [2]. Acute suppurative cholangitis (ASC) is a severe
form of acute cholangitis which is life-threatening without
appropriate timely management [3].
Mortality in cholangitis changes between 2.7% and 10%
[4]. It is not clear which patients will progress to sepsis and
death in cholangitis despite medical management and
which need decompression [5].

https://doi.org/10.1016/j.kjms.2017.12.016
1607-551X/Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Yıldız BD, et al., Simplified scoring system for prediction of mortality in acute suppurative cholangitis,
Kaohsiung Journal of Medical Sciences (2017), https://doi.org/10.1016/j.kjms.2017.12.016
 + MODEL
2 B.D. Yıldız et al.

Our aim in this study was to identify the factors
contributing to mortality in ASC and create a reliable
scoring system which could be tested easily in every
emergency clinic.
Patients and methods
This is a retrospective study enrolling 104 patients with ASC
diagnosed with presence of purulent material inside bile
ducts by ERCP or surgical exploration treated at our hos-
pital between January 2010 and July 2015. We extracted
patients with ASC out of cholangitis patients using relevant
International Statistical Classification of Diseases and
Related Health Problems codes. The data about patient
demographics, laboratory values at the time of admission,
radiological study results and clinical progression were
collected retrospectively from patient charts. The data
collected (Table 1) included age, gender, white blood cell
count, hemoglobin, red cell distribution width (RDW)
blood urea, creatinine, liver function tests, albumin, sys-
tolic blood pressure, diastolic blood pressure, mean arterial
pressure, pulse, medical illnesses, admission to intensive
care unit (ICU). Admission to ICU was decided using Sepsis-
related Organ Failure Assessment (qSOFA) criteria (greater
than or equal to 2). qSOFA was recommended to evaluate
sepsis in accordance with its new definition. qSOFA criteria
for sepsis include a Glasgow Coma Scale score of less than
or equal to 13, systolic blood pressure less than or equal to
100 mm Hg, and respiratory rate greater than or equal to 22
per minute (1 point each to yield a score value between
0 and 3) [5], Ultrasonography (USG) findings, computed
tomography findings, ERCP findings, type of surgery also
recorded. Patients were surgically decompressed if ERCP
was not available, unsuccessful or if there was a suspicion
of a malignancy. Diagnosis of suppurative cholangitis
was made when hyperbilirubinemia and either fever or
abdominal pain was present and ERCP or surgical explora-
tion revealed presence of purulent material inside the
common bile duct. Patients without this particular ERCP or
bile duct exploration finding were excluded from the study.
Statistical analysis
The statistical analysis of the study was performed by a
biomedical statistician. Factors associated with mortality
and morbidity were analyzed using univariate and multi-
variate analysis. Primary endpoint of our study was
mortality. Deaths that occurred within 30 days after surgi-
cal treatment or death at the same admission was defined
as hospital mortality.
ShapiroeWilk test was used for assessing normality.
Continuous data were presented as mean
standard devia-
tion while differences between groups were analyzed by
means of Students t test. Categorical variables were analyzed
with c2 test. Logistic regression was used to identify vari-
ables associated with mortality. Variables with p  0.2 in the
univariate analyses were included in multivariate analyses.
Results of the multivariate analysis were shown as odds ratio
(OR) and corresponding 95% confidence interval (CI). The
analysis of the receivereoperator characteristic (ROC) curve
was used to define the optimal cut-off value for continuous
variables in mortality. A clinical score based on the final lo-
gistic regression model was constructed; 1 point was given to
indicate presence of each predictive factor.

Table 1 Data collected from patient charts.

Survivors Non-Survivors p
Gender (F/M) 38/55 6/5 0.521
Age (years) (mean
SD) 65.55
14.69 69.18
19.5 0.440
Blood urea nitrogen (mg/dL) (mean
SD) 34 (23.5e56) 33 (22e61) 0.983
Creatinine (mg/dL) (mean
SD) 1.26
0.94 1.23
0.7 0.876
Albumin (g/dL) (mean
SD) 2.67
1.1 2.9
1.2 0.675
Alanine aminotransferase (IU/L) (mean
SD) 106 (56e188) 170 (91e227) 0.272
Aspartate aminotransferase (IU/L) (mean
SD) 88 (49e158) 144 (73e194) 0.183
Gamma glutamyl transferase (U/L) 328 (189e523) 364 (242e452) 0.658
Alkaline Phosphatase 241 (159e395) 338.5 (242e567) 0.059
Total bilirubin (mg/dL) (mean
SD) 5.95 (2.90e9.7) 9.1 (4.80e13.80) 0.048
Direct bilirubin (mg/dL) (mean
SD) 4.34 (1.77e6.80) 5.85 (2.95e8.3) 0.172
Red cell distribution width % (mean
SD) 14.55 (13.4e15.85) 16.1 (14.7e16.9) 0.044
White cell count (/mm3) (mean
SD) 11.2 (7.60e16.80) 9.5 (7.4e15.7) 0.583
Co-morbidities 0.608
None 35 3
Diabetes Mellitus 14 1
Hypertension 11 2
Cerebrovascular accident 3 0
Combination of co-morbidities 25 3
Other co-morbidities 5 2
Endoscopic retrograde cholangiopancreatography findings NA
Purulent 76 8
Purulent þ Debris 17 3
NA: Not applicable.

Please cite this article in press as: Yıldız BD, et al., Simplified scoring system for prediction of mortality in acute suppurative cholangitis,
Kaohsiung Journal of Medical Sciences (2017), https://doi.org/10.1016/j.kjms.2017.12.016
 + MODEL
Scoring system for cholangitis 3

Model discrimination was measured by the area under

Table 3 Treatment modalities used for cholangitis.
the ROC curve (AUC). The discrimination of a prognostic
model was considered perfect, good, moderate and poor Type of Surgery n
for AUC values of 1; >0,8; 0,6e0,8 and < 0,6; respectively. Choledocotomy and T tube drainage 9
Choledocoduodenostomy 4
Cholecystectomy and T tube drainage 7
Results Cholecystectomy 6
Whipple’s procedure 2
There were 44 female and 60 male patients. Mean age of Medical 76
patients was 65.9
15.1 years; 67.3% patients were older ERCP 80
than 60 years of age. None of the patients had a previous
history of abdominal surgery. The etiology was periampul-
lary tumor in two patients and biliary tract stones in the
rest. Ninety four patients had USG examination and twenty
Table 4 Independent predictors of mortality identified by
five patients had computed tomography (CT) confirmation
multivariate logistic regression analysis.
of the biliary stones and related changes (Table 2). Fifteen
patients (14.4%) had diabetes mellitus, 13 (12.5%) had Predictors p value SE* Odds (95% Cl)
hypertension, three (2.9%) had cerebrovascular accident, ratio
28 (26.9%) had combinations of these co-morbidities and 38 Total bilirubin 0.754 0.817 0.774 0.156e3.838
(36.5%) did not have a co-morbid disease. Sixty-nine (66.3%) (Cut off:6.9 mg/dl)
patients had ASC diagnosed during ERCP. RDW 0.121 0.887 0.252 0.044e1.436
The overall mortality rate was 10.6% (n Z 11). Twenty (Cut off: 14.45%)
eight patients (26.9%) had their attacks treated only Admission to ICU 0.002 0.840 0.078 0.015e0.403
surgically and five (17.8%) of these died. Of note thirteen of
*Standard Error, CI: Confidence Interval, ICU: Intensive Care
these patients had an ERCP done prior to the surgical Unit.
intervention. The rest of the patients were managed solely
with ERCP and 6 (7.89%) of them died. Treatment modal-
ities used are listed on Table 3. approximately 1.9% of the cohort. The third group,
In univariant analysis red cell distribution width (RDW), comprising approximately 2.88% of the patients, included
total bilirubin level, admission to ICU were identified as those with a score of 2 whose risk of mortality was 27.3%
statistically significant (p < 0.05) to predict mortality. and last group with a new score of 3 who had a 54.5% risk of
Three variables were statistically significant in multivariate mortality, this group comprised about 42.8% of the cohort
analysis: total bilirubin level more than or equal to (Table 5). The area under the ROC curve (AUC) was 0.807
6.9 mg/dl (OR Z 0.0445), RDW equal to or more than (95% CI, 0678e0.936) (Fig. 1).
14.45% (OR Z 1.1258), and admission to ICU (OR Z 1.0353)
(Table 4). A probability score was calculated by adding the
points given to these variables. Despite the differences in Discussion
regression coefficients, 1 point was given for each of these
risk factors to simplify the procedure. There are a number of studies which tried to delineate
Three groups of patients were defined based on this new factors predicting mortality in cholangitis. In one study
score. In the first group, with a score 0, there was no serum albumin, creatinine and urea concentrations were
mortality. The second group included patients with score 1, found to be elevated in patients who died from cholangitis
who had a 18.2% risk of mortality; this group comprised [6]. A study published in 1989 showed that acute renal
failure, liver abscess, malignant stricture, liver cirrhosis,
percutaneous transhepatic cholangiography, female sex and
Table 2 Ultrasonography and computed tomography age older than 50 were identified as prognostic factors for
findings. mortality in cholangitis [7]. Our study is the first in medical
Ultrasonography (n) Computed literature identifying RDW in univariate analysis as a factor
tomography (n) predicting mortality in cholangitis. Apart from this, in
multivariant analysis we found higher level of bilirubin than
Normal 4 5
similar studies [8] as predictor of mortality in cholangitis.
Cholelitiasis 17 3
This could be secondary to suppurative nature of the
Dilated IHBD 8 2
Dilated CBD 11 5
Acute cholecystitis 7 4 Table 5 Risk of mortality according to the new score in
Dilated IHBD and CBD 30 3 patients with acute suppurative cholangitis.
Cholelitiasis and 7 2
Score No Mortality n (%) Mortality n (%)
dilated CBD
Cholelitiasis and 9 0 0 23 (27.1) 0 (0)
dilated IHBD 1 29 (34.1) 2 (18.2)
2 25 (29.4) 3 (27.3)
IHBD:intrahepatic bile duct.
3 8 (9.4) 6 (54.5)
CBD:common bile duct.

Please cite this article in press as: Yıldız BD, et al., Simplified scoring system for prediction of mortality in acute suppurative cholangitis,
Kaohsiung Journal of Medical Sciences (2017), https://doi.org/10.1016/j.kjms.2017.12.016
 + MODEL
4 B.D. Yıldız et al.

Figure 1. ROC curve derived for the novel scoring system proposed by our research.

cholangitis in our series. It is not surprising that admission
to ICU would contribute to mortality as it necessitates
organ system failure. Of note, there was not a statistically
significant (p Z 0.05) relationship between surgery and
admission to ICU, thus excluding the possibility of a bias.
RDW is a routine complete blood count parameter
that reflects the measurement of erythrocyte cell size.
Classically, RDW is elevated in ineffective erythropoiesis
or increased erythrocyte destruction. Recent researches
highlighted usefulness of RDW as a novel prognostic marker
in different circumstances such as cardiovascular, throm-
boembolic and neurological diseases, sepsis, trauma, acute
and chronic inflammatory disorders [9,10]. Studies also
demonstrated that critically ill patients requiring intensive
care are likely to display high mortality rates associated
with elevated admission RDW levels [11]. Based on these
relationships between inflammation and RDW we could say
that statistically significant presence of high RDW in our
scoring system is actually a marker and end result of
suppuration in ASC.
The limitations of this study are retrospective design and
single center involvement. The strength of the scoring
system we propose can be increased with involvement of
multiple centers.
In this study, we are proposing a simple novel scoring
system to identify patients at risk for mortality in ASC
using basic laboratory and clinical criteria available in
every primary care facility even in developing countries.
This would serve as a simplified, rapid way to direct
patients for advanced interventions instead of wasting
time with more complicated and time consuming multi-
variable scoring systems.
References
[1] Park TY, Choi JS, Song TJ, Do JH, Choi S-H, Oh H-C. Early oral
antibiotic switch compared with conventional intravenous
antibiotic therapy for acute cholangitis with bacteremia. Dig
Dis Sci 2014;59:2790e6.
[2] Mayumi T, Someya K, Ootubo H, Takama T, Kido T, Kamezaki F,
et al. Progression of Tokyo Guidelines and Japanese Guide-
lines for management of acute cholangitis and cholecystitis. J
UOEH 2013;35:249e57.
[3] Tsujino T, Sugita R, Yoshida H, Yagioka H, Kogure H, Sasaki T,
et al. Risk factors for acute suppurative cholangitis caused by
bile duct stones. Eur J Gastroenterol Hepatol 2007;19:585e8.
[4] Navaneethan U, Jayanthi V, Mohan P. Pathogenesis of chol-
angitis in obstructive jaundice-revisited. Minerva Gastro-
enterol Dietol 2011;57:97e104.
[5] Kimura Y, Takada T, Strasberg SM, Pitt HA, Gouma DJ,
Garden OJ, et al. TG13 current terminology, etiology, and
epidemiology of acute cholangitis and cholecystitis. J Hep-
atobiliary Pancreat Sci 2013;20:8e23.
[6] Leese T, Neoptolemos J, Baker A, Carr-Locke D. Management
of acute cholangitis and the impact of endoscopic sphincter-
otomy. Br J Surg 1986;73:988e92.
[7] Gigot JF, Leese T, Dereme T, Coutinho J, Castaing D,
Bismuth H. Acute cholangitis. Multivariate analysis of risk
factors. Ann Surg 1989;209:435.

Please cite this article in press as: Yıldız BD, et al., Simplified scoring system for prediction of mortality in acute suppurative cholangitis,
Kaohsiung Journal of Medical Sciences (2017), https://doi.org/10.1016/j.kjms.2017.12.016
 + MODEL
Scoring system for cholangitis 5

[8] Dixon J, Armstrong C, Duffy S, Davies G. Factors affecting
morbidity and mortality after surgery for obstructive jaun-
dice: a review of 373 patients. Gut 1983;24:845e52.
[9] Meynaar I, Knook A, Coolen S, Le H, Bos M, Van Der Dijs F,
et al. Red cell distribution width as predictor for mortality in
critically ill patients. Neth J Med 2013;71:488e93.
[10] S‚enol K, Saylam B, Kocaay F, Tez M. Red cell distribution width
as a predictor of mortality in acute pancreatitis. Am J Emerg
Med 2013;31:687e9.
[11] Zhang Z, Xu X, Ni H, Deng H. Red cell distribution width is
associated with hospital mortality in unselected critically ill
patients. J Thorac Dis 2013;5:730e6.

Please cite this article in press as: Yıldız BD, et al., Simplified scoring system for prediction of mortality in acute suppurative cholangitis,
Kaohsiung Journal of Medical Sciences (2017), https://doi.org/10.1016/j.kjms.2017.12.016