Cao et al.

BMC Gastroenterology (2018) 18:128

https://doi.org/10.1186/s12876-018-0854-3

RESEARCH ARTICLE Open Access

Risk factors for post-ERCP cholecystitis: a

single-center retrospective study
Jun Cao1, Chunyan Peng1, Xiwei Ding1, Yonghua Shen1, Han Wu1, Ruhua Zheng1, Lei Wang1,2*
and Xiaoping Zou1,2*

Abstract
Background: The risk factors for post-ERCP cholecystitis (PEC) have not been characterized. Hence, this study
aimed to identify the potential risk factors for PEC.
Methods: The medical records of 4238 patients undergoing the first ERCP in a single center from January 2012 to
December 2016 were analyzed in this study. A multivariate analysis was used to identify the risk factors.
Results: This study included 2672 patients who met the enrollment criteria. Of these, 36 patients (incidence rate of 1.35%)
developed PEC within 2 weeks of the procedure. Univariate and multivariate analyses identified the following factors
associated with PEC: history of acute pancreatitis [odds ratio (OR) = 2.60; 95% confidence interval (CI): 1.29–5.23], history of
chronic cholecystitis (OR = 8.47; 95% CI: 2.54–28.24), gallbladder opacification (OR = 2.79; 95% CI: 1.37–5.70), biliary duct
metallic stent placement (OR = 3.66; 95% CI: 1.78–7.54), and high leukocyte count before ERCP (OR = 1.10; 95% CI:
1.04–1.17). The prediction model incorporating these factors demonstrated an area under the receiver operating
characteristic curve of 0.85 (95% CI, 0.80–0.91). A prognostic nomogram was developed using the aforementioned
variables to estimate the probability of PEC.
Conclusions: The risk factors, including the history of acute pancreatitis, history of chronic cholecystitis, gallbladder
opacification, biliary duct metallic stent placement, and high leucocyte counts before ERCP, increased the occurrence
of PEC and were positive predictors for PEC. The constructed nomogram was used to estimate the risk of PEC, guiding
the implementation of prophylactic measures to prevent PEC in clinical practice.
Keywords: ERCP, Cholecystitis, Nomogram, Risk factors, Success prediction

Background of PEP include suspected sphincter of Oddi dysfunction,

Endoscopic retrograde cholangiopancreatography (ERCP) major papilla pancreatogram, needle-knife precut, and fe-
is an endoscopic procedure performed under visual and male gender [2, 3].
fluoroscopic guidance. It is widely used in diagnosing and In contrast, post-ERCP cholecystitis (PEC) gained much
treating of biliary and pancreatic diseases. ERCP is a tech- less attention. Freemen et al. reported cholecystitis in 0.5%
nically challenging endoscopic procedure that can cause (11/2347) of patients 16 days after biliary sphincterotomy
serious adverse events and occasionally even death. Pos- [4]. In this study, no predictors of cholecystitis were identi-
sible ERCP-related adverse events include acute pancrea- fied other than the presence of stones in the gallbladder.
titis, hemorrhage, perforation, cholangitis, and acute Most studies reporting the adverse events of ERCP did not
cholecystitis. Of these, post-ERCP pancreatitis (PEP) is the investigate the risk factors and predictors of PEC alone,
most common one with 9.7% incidence and 0.7% mortal- which might be due to its relatively low incidence. How-
ity rate [1]. Due to its high incidence, numerous studies ever, most PECs require emergency cholecystectomy and
have investigated the risk factors of PEP. The risk factors extended hospitalization time. In addition, some PECs are
severe and potentially fatal. Identifying the risk factors for
PEC may help prevent this adverse event. The aim of this
* Correspondence: [email protected]; [email protected]
1
Department of Gastroenterology, Nanjing Drum Tower Hospital, The
study was to assess the risk factors for PEC in patients with
Affiliated Hospital of Nanjing University Medical School, Nanjing, China
Full list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Cao et al. BMC Gastroenterology (2018) 18:128 Page 2 of 7

gallbladder in situ within 2 weeks of procedure in a single using retrieval baskets and/or balloon-tipped catheters.
large-volume center. An endoscopic mechanical lithotripsy or laser lithotripsy
was attempted to crush down the stones if the stones were
Methods too big to remove. Obstructive jaundice resulting from
The study was approved by the Ethical Committee at malignant bile duct stenosis was treated by placing naso-
Nanjing Drum Tower Hospital Affiliated to Nanjing biliary drainage (ENBD), plastic stents, or self-expandable
University Medical School (study number 2017–167-01). biliary metal stents. The benign biliary stricture was
All subjects were anonymized; hence, informed consent treated by dilation or placement of plastic stents or fully
was not required. This study conformed to Strengthen- covered self-expandable biliary metal stents. Pancreatico-
ing the Reporting of Observational Studies in Epidemi- biliary maljunction or pancreas divisum was treated by
ology guidelines. placing ENBD or plastic stents.

Patients Diagnostic criteria of acute cholecystitis

The medical records of patients with gallbladder in situ Acute cholecystitis was diagnosed according to the 2018
who underwent ERCP for the first time in the hospital Tokyo guidelines of acute cholecystitis [5]. The diagnos-
from January 2012 to December 2016 were reviewed and tic criteria were based on the following three aspects:
analyzed retrospectively. Patients who had concomitant (A) local signs of inflammation, including (1) Murphy’s
acute cholecystitis at the time of ERCP or had a previous sign and (2) right upper abdominal quadrant mass/pain/
ERCP history were excluded from the study. The med- tenderness; (B) systemic signs of inflammation, including
ical records of eligible patients were reviewed retrospect- (1) fever, (2) elevated CRP, and (3) elevated WBC count;
ively to identify any occurrence of acute cholecystitis and (C) imaging finding characteristic of acute chole-
within 2 weeks after ERCP. cystitis. A definite diagnosis was as follows: one item in
A + one item in B + C.
Risk factors
The following predefined parameters were analyzed for Statistical analysis
PEC within 2 weeks. The demographic information in- Mean ± standard deviation was used to describe devi-
cluded the following: age and sex. The past history ation of the data with the normal distribution of the var-
included the following: acute pancreatitis, chronic chole- iables. Median (quartile spacing) was used to describe
cystitis, acute cholangitis, hypertension, hyperlipidemia, the data that did not meet the normal distribution of the
and diabetes mellitus. The laboratory examination in- variables. Frequency (percentage) was used to describe
dexes before ERCP were as follows: alanine aminotrans- the classification of variables. The differences between
ferase, aspartate aminotransferase, alkaline phosphatase, groups were compared using t-test, chi-square test, or
gamma-glutamyltranspeptidase, total bilirubin, direct rank-sum test [6]. Logistic regression was used to
bilirubin, leukocyte count, hemoglobin, and platelet analyze the findings of a multivariate analysis of acute
count. The indexes during ERCP were as follows: gall- cholecystitis after ERCP [7]. The nomogram was used to
bladder opacification, biliary duct stent, and common visualize the logistic regression model [8]. The Bonfer-
bile duct (CBD) diameter. Other factors before ERCP in- roni method was used to calibrate the adjusted test level
cluded temperature and antibiotics. During the bile duct for pairwise comparison of the findings of the chi-square
opacification, we recorded gallbladder opacification if test. Binned Scatterplot was used to describe the rela-
contrast medium entered into gallbladder and the out- tionship between preoperative leukocytes and the risk of
line of gallbladder could be seen. No additional efforts acute cholecystitis within 2 weeks after ERCP. SPSS 13.0
were made to get the entire gallbladder outlined by con- was used for statistical analysis. pROC and rms package
trast medium if this had not been accomplished simul- in R 3.3.3 software were used to construct receiver oper-
taneously with bile duct opacification. ating characteristic (ROC) curve and nomogram. A
two-tailed value of P <0.05 was established as the thresh-
Endoscopy protocol old of statistical significance.
Duodenal side-viewing endoscopes (JF-260, TJF-240, or
TJF-260; Olympus, Tokyo, Japan) were used to perform Results
the ERCP procedure. The patients were under midazolam Patient population
sedation. Sphincterotomy was performed using a standard A total of 4238 patients who underwent the first ERCP pro-
sphincterotome and/or a needle knife. Balloon sphinctero- cedure between January 1, 2012, and December 31, 2016,
plasty was performed using a Boston Scientific controlled in the hospital were included. Of these, 1352 patients were
radial expansion balloon with a diameter range of 12– excluded from the study due to concomitant acute chole-
15 mm, 15–18 mm or 18–20 mm). Stones were extracted cystitis (n = 182) or a history of cholecystectomy (n = 1170)
Cao et al. BMC Gastroenterology (2018) 18:128 Page 3 of 7

before ERCP. Further, 214 patients with more than 15% of after ERCP are shown in Table 1. The following parame-
missing data were also excluded. Finally, 2672 patients with ters were found to be closely correlated with PEC in the
intact gallbladder were included in the retrospective ana- univariate analysis: history of acute pancreatitis (χ2 =
lysis to analyze the incidence of acute cholecystitis within 17.754, P < 0.001), chronic cholecystitis (χ2 = 20.815, P <
2 weeks after the initial ERCP. The mean age of the patients 0.001), gallbladder opacification (χ2 = 11.816, P = 0.001),
was 62.4 ± 16.2 years (range, 1–106 years); 1166 patients bile duct stents (χ2 = 15.805, P = 0.001), leukocyte
(43.6%) were female (Table 1). Also, 36 patients (incidence count before ERCP (Z = − 3.610, P < 0.001). The mul-
rate of 1.35%) finally developed acute cholecystitis within tiple logistic regression analysis identified the following
2 weeks after the first ERCP (Fig. 1). variables significantly correlated with post-ERCP acute
cholecystitis (Table 2): history of acute pancreatitis (OR =
Risk factors for acute cholecystitis within 2 weeks after 2.60; 95% CI: 1.29–5.23; P = 0.007); history of chronic
the first ERCP cholecystitis (OR = 8.47; 95% CI: 2.54–28.24; P = 0.001),
The results of univariate analysis of potential risk factors gallbladder opacification (OR = 2.79; 95% CI: 1.37–5.70; P
for the development of acute cholecystitis within 2 weeks = 0.005), biliary duct metallic stent placement (OR = 3.66;

Table 1 Univariate analysis of potential risk factors for the development of acute cholecystitis after ERCP
Variable Acute Cholecystitis Statistic P
No (n = 2636) Yes (n = 36)
Age (y), Mean ± SD 62.4 ± 16.28 62.2 ± 12.90 t = 0.097 0.923
2
Female 1145 (43.4%) 21 (58.3%) x = 2.627 0.105
Past history
Hypertension 976 (37.0%) 13 (36.1%) x2 = 0.013 0.910
Hyperlipemia 70 (2.7%) 1 (2.9%) x2 = 0.005 0.942
2
Diabetes mellitus 473 (18.0%) 7 (20.0%) x = 0.098 0.755
Acute pancreatitis 459 (17.4%) 16 (44.4%) x2 = 17.754 < 0.001
2
Acute cholangitis 532 (20.2%) 6 (16.7%) x = 0.273 0.601
Chronic cholecystitis 1427 (54.1%) 33 (91.7%) x2 = 20.185 < 0.001
2
Antibiotics before ERCP 948 (36.0%) 11 (30.6%) x = 0.451 0.601
Gallbladder opacification 1016 (38.5%) 24 (66.7%) x2 = 11.816 0.001
Diameter of CBD(cm) 1.2 ± 0.48 1.1 ± 0.42 t = 1.237 0.216
Temperature before ERCP (°C) 36.6 ± 0.58 36.8 ± 0.84 t = −1.891 0.059
Bile duct stents x2 = 15.805 0.001
No stent 1174 (66.2%) 19 (52.8%)
Metallic stent 414 (15.7%) 14 (38.9%)
Plastic stent 436 (16.5%) 2 (5.6%)
Metallic+plastic stent 42 (1.6%) 1 (2.8%)
Laboratory index before ERCP (Median,P25,P75)
ALT 83.0 (37.3, 193.3) 58.7 (37.5, 191.0) Z = −0.027 0.979
AST 54.9 (27.9, 119.0) 53.5 (28.7, 83.8) Z = −0.143 0.886
AKP 198.1 (114.9, 359.8) 196.3 (94.2, 321.2) Z = −0.414 0.679
GGT 292.5 (125.7, 550.2) 246.9 (91.9, 543.8) Z = −0.181 0.856
TB 34.9 (15.5, 126.8) 23.7 (13.7, 83.3) Z = −0.845 0.391
DB 21.2 (6.7, 98.8) 12.8 (7.1, 73.4) Z = −0.641 0.522
WBC 6.1 (4.8, 8.2) 7.9 (5.9, 11.0) Z = −3.610 < 0.001
Hemoglobin 125.0 (111.0, 136.0) 131.0 (121.0, 138.0) Z = −1.413 0.158
Platelet count 192.0 (148.0, 245.0) 192.0 (146.0, 276.0) Z = −0.210 0.833
CBD common bile duct, ALT alanine aminotransferase, AST aspartate anminotransferase, AKP alkaline phosphatase, GGT gamma-glutamyltranspeptidase, TB total
bilirubin, DB direct bilirubin, WBC white blood cell
Cao et al. BMC Gastroenterology (2018) 18:128 Page 4 of 7

Fig. 1 Flowchart of study results

95% CI: 1.78–7.54; P<0.001) and leukocyte count before or percutaneous transhepatic gallbladder drainage.
ERCP (OR = 1.10; 95% CI: 1.04–1.17; P = 0.001). In the 36 Therefore, PEC should be recognized early. The present
patients who developed PEC, 29 had gallstones. study, included 2666 patients with intact gallbladder
The multivariate models were built to predict the inci- who underwent the first ERCP, and the incidence of
dence of acute cholecystitis after ERCP within 2 weeks. Ac- acute cholecystitis was 1.35% (36/2672) within 2 weeks
cording to the ROC of the multivariate model, the area after ERCP. The univariate and multivariate analyses in-
under the curve (AUC) was 0.852; the sensitivity and speci- dicated that the history of chronic cholecystitis, previous
ficity were 82.3% and 73.3%, respectively (Fig. 2). The result acute pancreatitis, gallbladder opacification, biliary stent
revealed a good concordance and a good predictive ability. placement, and high leukocyte count before ERCP were
Finally, the correlation between white blood cell risk factors for the occurrence of PEC within 2 weeks of
counts before ERCP and PEC was estimated using a the procedure. Of note, biliary metallic stent placement
binned scatterplot diagram (Fig. 3). The results indicated significantly increased the occurrence of PEC.
a curvilinear relationship; also, the risk of PEC increased As a risk factor for PEC, chronic cholecystitis may in-
with the increase in preoperative WBC. crease PEC perhaps owing to gallbladder contamination
by nonsterile contrast or intestinal reflux. The diameter
Discussion of the biliary duct metallic stent was greater than that of
Although PEC is not as common as PEP, it can lead to the plastic stent. Therefore, metallic stent placement
purulent cholecystitis and result in emergency operation during ERCP greatly increased duodenal biliary reflux,

Table 2 Multivariate logistic regression analysis of potential risk factors for subsequent post-ERCP cholecystitis
Variable B S.E P OR (95% CI)
WBC before ERCP 0.099 0.029 0.001 1.10 (1.04, 1.17)
History of acute pancreatitis 0.955 0.357 0.007 2.60 (1.29, 5.23)
History of chronic cholecystitis 2.137 0.614 0.001 8.47 (2.54, 28.24)
Gallbladder opacification 1.026 0.364 0.005 2.79 (1.37, 5.70)
Stent types – – 0.001 –
No reference reference reference reference
Metallic stent 1.298 0.369 < 0.001 3.66 (1.78, 7.54)
Plastic stent −0.578 0.759 0.446 0.56 (0.13, 2.48)
Metallic +plastic stent 1.735 1.077 0.107 5.67 (0.69, 46.78)
Constant reference reference reference reference
Cao et al. BMC Gastroenterology (2018) 18:128 Page 5 of 7

Fig. 2 ROC curve for logistic regression model predicting post-ERCP cholecystitis. It included a history of chronic cholecystitis, history of pancreatitis,
gallbladder opacification, leukocyte count, and biliary metallic duct stent. AUC = 0.85; 95% CI: 0.80–0.91

further increasing the possibility of PEC. Obstructions of estimated using a binned scatterplot diagram. The results
the cystic duct by the stent may also contribute to the de- indicated that the risk of acute cholecystitis had a positive
velopment of PEC. An interesting finding in the study was correlation with preoperative WBC. However, the associ-
that the biliary duct plastic stent did not increase the risk ation was not strong because the OR was low (OR = 1.1).
of PEC. The patients with high leukocyte count before The present study, analyzed whether serum total bilirubin
ERCP were predisposed to PEC. The correlations between level and CBD diameter were risk factors for PEC. The re-
white blood cell counts before ERCP and PEC were sult suggested no correlation between them. The result was

Fig. 3 Binned scatterplot diagram of the relationship between leukocyte count before ERCP and the risk of post-ERCP cholecystitis
Cao et al. BMC Gastroenterology (2018) 18:128 Page 6 of 7

Fig. 4 The Nomogram to predict the risk of post-ERCP cholecystitis. The behavioral variables are presented in rows 2–6, and points for each
variable correspond to the scale in row 1. The points of five variables are added to the total points presented on the scale in row 7, which
corresponds to the risk predictor of post-ERCP acute cholecystitis within 2 weeks in rows 8

different from those of previous studies. Lee et al. assessed Endoscopic gallbladder drainage, as a safe and effica-
the risk factors for acute cholecystitis after endoscopic CBD cious internal drainage, improved patient pain and de-
stone removal during a mean 18-month follow-up [9]. They creased the likelihood of the drain being dislodged [10].
reported that a serum total bilirubin level < 1.3 mg/dL and Therefore, endoscopic gallbladder drainage has been
a CBD diameter < 11 mm at the time of endoscopic CBD used for elderly patients with multiple comorbidities at
stone removal were the risk factors for the development of high risk for cholecystectomy to decompress the gall-
PEC; the incidence of PEC was 17%. However, the bladder as a temporary measure prior to surgery or as
follow-up time of their study on the occurrence of PEC was the definitive treatment [10–13]. Briefly, patients at high
much longer than that in the present study. risk of PEC may undergo drainage with an endoscopic-
The past history of acute pancreatitis was a risk factor of ally placed nasocholecystic tube or plastic stents.
PEC in this study. The causes for previous acute pancrea- The present study had several limitations. First, it was
titis in medical records for most patients were unclear and a single-center retrospective study with a possibility of
indefinite. It was difficult to explain why previous acute accumulation of inappropriate data. Moreover, chole-
pancreatitis was associated with PEC. However, according cystitis was not classified as acalculous or calculous be-
to published endoscopic ultrasonography studies, a num- cause of the small number of patients with PEC.
ber of patients with past “idiopathic” acute pancreatitis Prospective studies should be performed to further es-
might have suffered from acute pancreatitis due to micro- tablish the risk factors for PEC.
lithiasis and sludge.
In the present study, a practical nomogram was estab- Conclusions
lished to predict PEC with a good sensitivity and specifi- A history of acute pancreatitis, history of chronic cholecyst-
city (Fig. 4). According to the ROC of multivariate itis, gallbladder opacification, biliary metal stent placement,
model, the AUC was 0.85 (95% CI 0.80–0.91), and the and high leukocyte counts before ERCP were established as
sensitivity and specificity were 82.3% and 73.3%, respect- potential risk factors for the occurrence of PEC within
ively. The nomogram revealed a good concordance and 2 weeks by univariate and multivariate analyses. When pa-
a good predictive ability for PEC. The present study re- tients with these risk factors undergo ERCP, prophylactic
ported the first nomogram for predicting PEC. External measures should be taken to prevent PEC.
validation of this nomogram is needed in further studies.
Identifying the risk factors related to PEC is important Abbreviations
for taking precautions to reduce the occurrence of PEC. AUC: Area under the curve; CBD: Common bile duct; CI: Confidence interval;
ENBD: Nasobiliary drainage; ERCP: Endoscopic retrograde
When patients with these risk factors undergo ERCP, cholangiopancreatography; PEC: Post-ERCP cholecystitis; PEP: Post-ERCP
prophylactic measures should be taken to prevent PEC. pancreatitis; ROC: Receiver operating characteristic curve
Cao et al. BMC Gastroenterology (2018) 18:128 Page 7 of 7

Funding 10. Itoi T, Kawakami H, Katanuma A, Irisawa A, Sofuni A, Itokawa F, Tsuchiya T,

This study was supported by Nanjing Medical Science and Technique Tanaka R, Umeda J, Ryozawa S, et al. Endoscopic nasogallbladder tube or
Development Foundation (QRX17037). stent placement in acute cholecystitis: a preliminary prospective randomized
trial in Japan (with videos). Gastrointest Endosc. 2015;81(1):111–8.
Availability of data and materials 11. Kjaer DW, Kruse A, Funch-Jensen P. Endoscopic gallbladder drainage of
The data set analyzed in the current study cannot be opened to public patients with acute cholecystitis. Endoscopy. 2007;39(4):304–8.
because patients’ privacy must be protected and IRB does not permit to do 12. Mutignani M, Iacopini F, Perri V, Familiari P, Tringali A, Spada C, Ingrosso M,
so. However, data are available from the author upon reasonable request. Costamagna G. Endoscopic gallbladder drainage for acute cholecystitis:
technical and clinical results. Endoscopy. 2009;41(6):539–46.
Authors’ contributions 13. Widmer J, Alvarez P, Sharaiha RZ, Gossain S, Kedia P, Sarkaria S, Sethi A,
XZ and LW contributed to conception and design of the study and have Turner BG, Millman J, Lieberman M, et al. Endoscopic gallbladder drainage
been involved in revising the manuscript critically. CP and YS contributed to for acute Cholecystitis. Clin Endosc. 2015;48(5):411–20.
analysis and interpretation of data. JC contributed to the study design,
analysis of data and drafting the manuscript. RZ and HW contributed to the
acquisition of data. XD contributed to critically revising the manuscript and
interpretation of data. All authors read and approved the final version of the
manuscript.

Ethics approval and consent to participate

The study was approved by the Ethical Committee at Nanjing Drum Tower
Hospital Affiliated to Nanjing University Medical School (study number 2017–
167-01). The Ethical Committee at Nanjing Drum Tower Hospital Affiliated to
Nanjing University Medical School approved the waiver of consent.

Consent for publication

Not applicable.

Competing interests
The authors declare that they have no competing interests.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

Author details
1
Department of Gastroenterology, Nanjing Drum Tower Hospital, The
Affiliated Hospital of Nanjing University Medical School, Nanjing, China.
2
Zhongshan Road 321, Department of Gastroenterology, Nanjing Drum
Tower Hospital, The Affiliated Hospital of Nanjing University Medical School,
Nanjing 210008, Jiang Su Province, China.

Received: 7 April 2018 Accepted: 31 July 2018

References
1. Kochar B, Akshintala VS, Afghani E, Elmunzer BJ, Kim KJ, Lennon AM,
Khashab MA, Kalloo AN, Singh VK. Incidence, severity, and mortality of post-
ERCP pancreatitis: a systematic review by using randomized, controlled
trials. Gastrointest Endosc. 2015;81(1):143–9.
2. Cotton PB, Garrow DA, Gallagher J, Romagnuolo J. Risk factors for
complications after ERCP: a multivariate analysis of 11,497 procedures over
12 years. Gastrointest Endosc. 2009;70(1):80–8.
3. Wang P, Li ZS, Liu F, Ren X, Lu NH, Fan ZN, Huang Q, Zhang X, He LP, Sun
WS, et al. Risk factors for ERCP-related complications: a prospective
multicenter study. Am J Gastroenterol. 2009;104(1):31–40.
4. Freeman ML, Nelson DB, Sherman S, Haber GB, Herman ME, Dorsher PJ,
Moore JP, Fennerty MB, Ryan ME, Shaw MJ, et al. Complications of
endoscopic biliary sphincterotomy. N Engl J Med. 1996;335(13):909–18.
5. Yokoe M, Hata J, Takada T, Strasberg SM, Asbun HJ, Wakabayashi G, Kozaka
K, Endo I, Deziel DJ, Miura F, et al. Tokyo guidelines 2018: diagnostic criteria
and severity grading of acute cholecystitis (with videos). J Hepatobiliary
Pancreat Sci. 2018;25(1):41–54.
6. Zhang Z. Univariate description and bivariate statistical inference: the first
step delving into data. Ann Transl Med. 2016;4(5):91.
7. Zhang Z. Variable selection with stepwise and best subset approaches. Ann
Transl Med. 2016;4(7):136.
8. Zhang Z, Kattan MW. Drawing Nomograms with R: applications to
categorical outcome and survival data. Ann Transl Med. 2017;5(10):211.
9. Lee JK, Ryu JK, Park JK, Yoon WJ, Lee SH, Lee KH, Kim YT, Yoon YB. Risk
factors of acute cholecystitis after endoscopic common bile duct stone
removal. World J Gastroenterol. 2006;12(6):956–60.