Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

380

Patarapong Kamalaporn, M.D.*, Supphamat Chirnaksorn, M.D.*, Sasivimol Rattanasiri, Ph.D.**, Taya Kitiyakara,

MBBS*

* Department of Medicine, Faculty of Medicine Ramathibodi Hospital, **Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine

Ramathibodi Hospital, Mahidol University, Bangkok 10400, ailand.

Comparison of Radial Echoendoscopy and

Predictive Factors in the Evaluation of Patients with

Suspected Choledocholithiasis

ABSTRACT

Objective: e aim of this study was to compare predictive factors and endoscopic ultrasound (EUS) in the diagnosis

of choledocholithiasis.

Materials and Methods: Patients with suspected choledocholithiasis were recruited from April 2011 to January

2018. All patient characteristics, EUS ndings and ERCP ndings were recorded and analyzed.

Results: Eighty patients were enrolled in this study. Clinical symptoms, blood chemistry and liver function tests

were similar in patients with and without choledocholithiasis. Using the ndings of ERCP as the gold standard,

radial EUS had a sensitivity and specicity for the detection of choledocholithiasis of 90.2% and 97.4%, and for

choledocholithiasis and/or common bile duct sludge 92.7% and 100%, respectively. For patients with intermediate

likelihood and high likelihood of having choledocholithiasis, as calculated from their predictive factors (33 and

45), radial EUS was positive for choledocholithiasis in 51.5% (17/33) and 46.7% (21/45), and ERCP was positive

for choledocholithiasis in 54.5% (18/33) and 48.9% (22/45), respectively.

Conclusion: Predictive factors, for both the intermediate and high likelihood groups, were not accurate to diagnose

these patients. EUS is a good diagnostic tool and should be performed in both groups of patients to avoid unnecessary

ERCP.

Keywords: Echoendoscopy; choledocholithiasis (Siriraj Med J 2021; 73: 380-385)

Corresponding author: Patarapong Kamalaporn

E-mail: Patarakamla@yahoo.com

Received 10 October 2020 Revised 4 January 2021 Accepted 7 January 2021

ORCID ID: http://orcid.org/0000-0003-0397-1892

http://dx.doi.org/10.33192/Smj.2021.50

INTRODUCTION

Cholelithiasis is a common problem and occurs 6-9%

in the population.

Most patients are asymptomatic but

some can develop biliary colic and other complications.

Choledocholithiasis is one of the common complications

and can occur in about 20% of these patients.

Once

they have choledocholithiasis, cholangitis and acute

biliary pancreatitis can occur. e diagnosis of patients

with suspected choledocholithiasis can be made by

clinical symptoms, physical examination, changes of

liver function test and transabdominal ultrasonography.

From previous studies, certain factors were found to

improve the diagnostic accuracy of choledocholithiasis

by up to 70%. Such factors included clinical ascending

cholangitis, the common bile duct being larger than

6 mm with the gall bladder in situ on transabdominal

Kamalaporn et al.

Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

381

Original Article

SMJ

ultrasonography, and a bilirubin level higher than 1.8

mg/dL.

Some imaging methods also have a role in

the diagnosis of this condition, such as Endoscopic

Ultrasound (EUS), Computerized Tomography (CT),

Magnetic Resonance Imaging and Magnetic Resonance

Cholangiopancreatography (MRI and MRCP), but the

choice of which method to use depends on the local

accessibility and the accuracy of each method.

An Endoscopic Ultrasound is an endoscope

with an ultrasound probe at the tip which allows the

endosonographer to examine the gastrointestinal tract,

hepatobiliary tract and pancreas closely by ultrasound.

is method is safe and has a low rate of complications,

such as bowel perforation and bleeding (0.12%).

e

accuracy of hepatobiliary and pancreatic examinations

is reported to be as good as MRI and MRCP, at about

91-93% from a previous study.

For CT, MRI and MRCP,

patients receive radiation or magnetic resonance as well

as intravenous contrast media which may deteriorate

renal function. Patients with renal insuciency may

have some limitation for both studies. e gold standard

for the detection of choledocholithiasis is Endoscopic

Retrograde Cholangiopancreatography (ERCP). ERCP

can be both diagnostic and therapeutic at the same time,

but as it is more invasive, complications can occur at a

higher rate than for EUS, at about 4%, and include acute

pancreatitis, perforation, bleeding and infection.

EUS and ERCP are both highly accurate for detecting

choledocholithiasis. EUS has a sensitivity of 89-94% and

a specicity of 94-95% when ERCP ndings are used

as the gold standard. But EUS has lower complications

(Relative risk 0.35, 95% CI 0.2-0.62; p<0.001) and a

lower incidence of pancreatitis (Relative risk 0.21, 95%

CI 0.06-0.83; p 0.03).

e American Society of Gastrointestinal Endoscopy

(ASGE) guideline

categorize patients with suspected

choledocholithiasis into low, intermediate and high

likelihood using predictive factors. Additional tests (MRCP

or EUS) are recommended to conrm choledocholithiasis

in the intermediate likelihood group before doing ERCP.

is study was designed to compare the sensitivity

and specicity of the patients’ predictive factors with the

EUS ndings in the detection of choledocholithiasis, in

patients who underwent ERCP.

MATERIALS AND METHODS

is prospective, descriptive study was approved

by Ramathibodi Ethical Committee. The study was

performed in Ramathibodi Endoscopic unit from April

2011 to January 2018. We recruited patients 18 to 80 years

old who were suspected of having choledocholithiasis

and transabdominal ultrasonography was negative for

choledocholithiasis. Patients had at least one of the following:

clinical symptoms of cholangitis (fever, abdominal pain

and jaundice), alkaline phosphatase > 300 unit/L, direct

bilirubin > 1.8 mg/dL, clinical symptoms of gall stone

pancreatitis, common bile duct (CBD) dilatation to at

least 6 mm in patients with the gall bladder in-situ or

common bile duct dilatation to at least 10 mm in patients

aer cholecystectomy.

3,8

We excluded patients who were

unable to undergo esophagogastroduodenoscopy such

as those with esophageal stricture, pyloric stricture and

patients who refused to participate in the study. All

patients were informed about risks and benets of both

procedures, radial EUS and ERCP. Patients signed the

consent forms before starting all procedures.

EUS was performed in these patients as soon as

possible by two endosonographers and an Olympus GF-

UE160-AL5 Radial Array Ultrasound Gastrovideoscope

was used. With the Radial EUS, the CBD was carefully

evaluated for choledocholithiasis or common bile duct

sludge. We dened choledocholithiasis as a hyperechoic

lesion with posterior acoustic shadow or a hypoechoic

lesion which was movable in the CBD and common bile

duct sludge as hyperechoic foci or content in common

bile duct which included microlithiasis and viscous bile

uid. Microlithiasis and viscous bile uid are known

to cause intermittent common bile duct obstruction

and pancreatitis.

9-11

ERCP was done either at the same

session or within 6 weeks for denite diagnosis and

treatment. At ERCP, if the cholangiogram looked suspicious

for choledocholithiasis or common bile duct sludge,

endoscopic sphincterotomy was done and a balloon

or basket extraction was applied to clear common bile

duct. e diagnosis was conrmed by two experienced

endoscopists.

All predictive factors, ndings of radial EUS and

ndings of ERCP were compared by statistical analysis.

All patients were classied by their predictors according to

the ASGE guideline 2010 into low, intermediate and high

likelihood groups to check the accuracy of this guideline

(Fig 1).

Descriptive data was presented as mean with

standard deviation (SD) or median with range. Factors

associated with choledocholithiaisis were analyzed by Chi-

square test or T-test. Statistical signicance was dened

as p-value < 0.05. e sensitivity and specicity values

for radial EUS for the detection of choledocholithiasis

and choledocholithiasis and/ or common bile duct sludge

were calculated, using the ERCP ndings as the gold

standard. All statistical analyses were performed using

STATA version 15.

Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

382

Kamalaporn et al.

RESULTS

Eighty patients were recruited into this study

from a total of 722 patients who underwent ERCP for

choledocholithiasis in the same time period. For the 80

patients with suspected choledocholithiasis and negative

choledocholithiasis on transabdominal ultrasonography,

the mean (±SD) of age was 65.2 (±14.9) years old and 46

patients were female. Patients presented with abdominal

pain 87.5% (70/80), fever 58.8% (47/80) and jaundice

71.3% (57/80). e median (range) for alkaline phosphatase

was 183 u/L (62-1309), serum glutamic-oxaloacetic

transaminase level 112 (16-4289) U/L, serum glutamic-

pyruvic transaminase level 142 (13-1782) U/L, total

bilirubin 2.5 (0.2-16) mg/dL and direct bilirubin 1.7 (0.1-

12) mg/dL. In 69 patients, radial EUS and ERCP were

done on the same day/ same setting. For nine patients

these two procedures were done with an interval of one

day, and in two patients, the interval was 35 days. No

complication was noted in all recruited patients.

Radial EUS showed choledocholithiasis in 37 patients

and ERCP conrmed choledocholithiasis in 41 patients.

e sensitivity and specicity of radial EUS for detecting

choledocholithiasis were 90.2% (37/41) and 97.4% (38/39),

respectively (Table 1). Baseline characteristics for patients

with and without choledocholithiasis, including symptoms,

blood chemistry and liver function tests, were similar

(Table 2). Only the mean age of patients was higher

in patients with choledocholithiasis, mean±SD: 68.66

(14.58) vs 61.54 (14.49), p=0.032.

When we included patients with choledocholithiasis

and/ or common bile duct sludge, radial EUS was positive

in 63 patients and ERCP was positive in 68 patients. e

sensitivity and specicity of radial EUS for detecting

choledocholithiasis and / or sludge in common bile

duct were 92.7% (63/68) and 100% (12/12), respectively

(Table 1). When we analyzed the performance of radial

EUS in detecting only common bile duct sludge, the

sensitivity was 72.5% (29/40) and the specicity was

95% (38/40), (Table 1).

When the predictors for choledocholithiasis were

used to classify the patients according to the ASGE

guideline 2010, 2 of our patients were in the low likelihood

group, 33 in the intermediate likelihood group and 45 in

the high likelihood group. e guideline recommended

further investigations for the intermediate group. In this

group, radial EUS was positive for choledocholithiasis

in 17 patients (51.5%, 17/33) and ERCP was positive

for choledocholithiasis in 18 patients (54.5%, 18/33).

When we included patients with choledocholithiasis and/

or common bile duct sludge, Radial EUS was positive

Very strong

Common bile duct stone on transabdominal ultrasonography

Clinical ascending cholangitis

Bilirubin ˃4 mg/dL

Strong

Dilated common bile duct on transabdominal ultrasonography (˃6 mm with gall bladder in situ)

Bilirubin level 1.8-4 mg/dL

Moderate

Abnormal liver biochemical test other than bilirubin

Age older than 55 year

Clinical gall stone pancreatitis

Assigning a likelihood of choledocholithiasis based on clinical predictors

Presence of any very strong predictor high

Presence of both strong predictor high

No predictor present low

All other patients intermediate

Fig 1. Predictors of choledocholithiasis

Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

383

Original Article

SMJ

in 25 patients (75.8%, 25/33) and ERCP was positive

in 28 patients (84.8%, 28/33). For the high likelihood

group, radial EUS was positive for choledocholithiasis

in 21 patients (46.7%, 21/45) and ERCP was positive for

choledocholithiasis in 22 patients (48.9%, 22/45). When we

included patients who were positive for choledocholithiasis

and/ or sludge in the analysis, radial EUS was positive

in 36 patients (80%, 36/45) and ERCP was positive in

38 patients (84.4%, 38/45). e accuracy of EUS in the

diagnosis of choledocholithiasis, and choledocholithiasis

and/ or common bile duct sludge was high in both groups

of patients and is shown in Table 1.

DISCUSSION

Endoscopic Ultrasound is a low risk endoscopic

procedure which can evaluate the common bile duct in

patients with suspected choledocholithiasis. Previous

studies have suggested that a strategy of EUS-directed ERCP

for choledocholithiasis allowed 50-70% of the patients

to avoid a diagnostic ERCP and reduced complications

by 4-7%. At 1-year follow-up, there was no dierence

in outcomes between patients in both groups.

12-14

Performance of EUS Sensitivity Specicity ROC area, (95% CI)

EUS/ERCP, n (%) EUS/ERCP, n (%)

For choledocholithiasis 37/41 (90.2%) 38/39 (97.4%) 0.94 (0.89-0.99)

For CBD sludge 29/40 (72.5%) 38/40 (95%) 0.86 (0.79-0.93)

For choledocholithiasis 63/68 (92.7%) 12/12 (100%) 0.96 (0.93-0.99)

and/or CBD sludge

Categorized by ASGE guideline:

Intermediate likelihood group

For choledocholithiasis 17/18 (94.4%) 15/15 (100%) 0.97 (0.92-1)

For choledocholithiasis 25/28 (89.3%) 5/5 (100%) 0.95 (0.89-1)

and/or CBD sludge

High likelihood group

For choledocholithiasis 21/22 (95.4%) 23/23 (100%) 0.97 (0.93-1)

For choledocholithiasis 36.38 (94.7%) 7/7 (100%) 0.97 (0.94-1)

and/or CBD sludge

TABLE 1. Performance of EUS (ERCP ndings as gold standard)

Abbreviations: CBD: common bile duct, ROC: Receiver operating characteristic

In this study, we recruited patients with suspected

choledocholithiasis from clinical symptoms and blood

chemistry, without visible choledocholithiasis on

transabdominal ultrasonography. e reason that our

study recruited this group of patients was that patients who

were positive for choledocholithiasis on transabdominal

ultrasonography would go straight for ERCP. ey would

not need to have another non-invasive imaging technique

such as EUS to conrm choledocholithiasis. Most of

the patients in this study (69/80) underwent radial EUS

and ERCP back-to-back, and 9/80 patients had both

procedures with an interval of only one day. As a result,

the results of both procedures, and the dierence between

them would unlikely be confounded by the passage of

stones in between the procedures. A long waiting time

between the two procedures would allow the passing

of choledocholithiasis and give a false negative ERCP

result. Conversely, cholelithiasis could also be passed

into the common bile duct during the waiting time. is

would produce a positive ERCP result, and decrease the

sensitivity of EUS.

Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

384

Kamalaporn et al.

TABLE 2. Comparison baseline characteristic between positive and negative choledocholithiasis (CBD stone) from

ERCP.

Characteristic

ERCP_CBD=negative ERCP_CBD=positive

(n=39) (n=41)

P-value

Age, mean (SD) 61.54 (14.49) 68.66 (14.58) 0.032

Sex, n (%)

male 14 (35.9) 20 (48.78) 0.244

female 25 (64.1) 21 (51.22)

Abdominal pain, n (%)

no 5 (12.82) 5 (12.2) 1.000

present 34 (87.18) 36 (87.8)

Fever, n (%)

no 18 (46.15) 15 (36.59) 0.385

present 21 (53.85) 26 (63.41)

Jaundice, n (%)

no 8 (20.51) 15 (36.59) 0.112

present 31 (79.49) 26 (63.41)

ALP, median (range) 174 (64, 1309) 196 (62, 503) 0.434

SGOT median (range) 118 (24, 4289) 99 (16, 1065) 0.713

SGPT median (range) 143 (23, 1782) 137 (13, 782) 0.906

GGT median (range) 353 (9, 2344) 462 (31, 1396) 0.201

TB median (range) 2.40 (0.210, 10.80) 2.70 (0.20, 16) 0.721

DB median (range) 1.70 (0.10, 8.60) 1.50 (0.10, 12) 0.743

Categorized by ASGE guideline

Low likelihood, n (%) 1 (2.56) 1 (2.43)

Intermediate likelihood, n (%) 15 (38.46) 18 (43.9)

High likelihood, n (%) 23 (58.97) 22 (53.65)

Common bile duct sludge is known to cause similar

complications to choledocholithiasis such as cholangitis,

common bile duct obstruction and gall stone pancreatitis.

So we also analyzed choledocholithiasis and/ or common

bile duct sludge in the study. Radial EUS was found to

have high sensitivity and specicity for the detection of

choledocholithiasis, at 90.2% and 97.4% respectively, and

also for choledocholithiasis and/or common bile duct

sludge, at 92.7% and 100%, respectively. Radial EUS

had a lower sensitivity, 72.5%, but still a high specicity,

95%, for detecting only common bile duct sludge. e

amount of sludge in the common bile duct varied for

each patient and may have aected the performance of

the radial EUS.

ere was no signicant dierence in the baseline

characteristics between patients with choledocholithiasis

Abbreviations: ALP: alkaline phosphatase, SGOT: serum glutamic-oxaloacetic transaminase, SGPT: serum glutamic-pyruvic transaminase,

GGT: gamma glutamyl transferase, TB: total bilirubin, DB: direct bilirubin

Volume 73, No.6: 2021 Siriraj Medical Journal

https://he02.tci-thaijo.org/index.php/sirirajmedj/index

385

Original Article

SMJ

and without choledocholithiasis, particularly in terms of

clinical symptoms (abdominal pain, fever and jaundice)

and blood chemistry including the liver function test.

It seems that the predictors were not so useful in this

study.

For patients with intermediate likelihood and high

likelihood of choledocholithiasis according to the ASGE

guideline 2010

classication, ERCP was positive for

choledocholithiasis 54.5% (18/33) and 48.9% (22/45),

respectively. e accuracy of the predictors from this

guideline was also low for our patients in both groups.

EUS ndings were more accurate than the predictors.

Although this study was prospective in design, it

only had a small number of patients, making it dicult to

suggest the applicability of the ndings with any strength. A

further large study may conrm the ndings of this study.

CONCLUSION

Endoscopic Ultrasound with a radial echoendoscope

was highly accurate and a safe endoscopic procedure for

the detection of choledocholithiasis when performed by

experienced endosonographers. For patients in both the

intermediate likelihood and the high likelihood groups, EUS

showed benet and allowed patients to avoid unnecessary

Endoscopic Retrograde cholangiopancreatography and

its related complications.

REFERENCES

1. Everhart JE, Khare M, Hill M, Maurer KR. Prevalence and ethnic

differences in gallbladder disease in the United States.

Gastroenterology 1999;117(3):632.

2. Freitas ML, Bell RL, Duy AJ. Choledocholithiasis: evolving

standards for diagnosis and management. World J Gastroenterol

2006;12:3162-7.

3. ASGE Standards of Practice Committee, Maple JT, Ben-

Menachem T, Anderson MA, Appalaneni V, Banerjee S, et al. e

role of endoscopy in the evaluation of suspected choledocholithiasis.

Gastrointest Endosc 2010;71:1-9.

4. Buscarini E, De Angelis C, Arcidiacono PG, Rocca R,Lupinacci

G,Manta R, et al. Multicentre retrospective study on endoscopic

ultrasound complications. Digestive and Liver Disease 2006;38:

762-7.

5. Ledro-Cano D. Suspected choledocholithiasis: endoscopic

ultrasound or magnetic resonance cholangio-pancreatography?

A systematic review. Eur J Gastroenterol Hepatol 2007;19:1007-

11.

6. Cotton PB, Garrow DA, Gallagher J, Romagnuolo J. Risk

factors for complications aer ERCP: a multivariate analysis of

11,497 procedures over 12 years. Gastrointest Endosc 2009;70:

80-8.

7. Petrov MS, Savides TJ. Systematic review of endoscopic

ultrasonography versus endoscopic retrograde cholangio-

pancreatography for suspected choledocholithiasis. Br J Surg

2009;96:967-74.

8. Kim JE, Lee JK, Lee KT, Park DI,Hyun JG,Paik SW, et al.

e clinical signicance of common bile-duct dilatation in patients

without biliary symptoms or causative lesions on ultrasonography.

Endoscopy 2001;33:495-500.

9. Venu RP, Geenen JE, Hogan W, Stone J, Johnson GK, Soergel

K. Idiopathic recurrent pancreatitis. An approach to diagnosis

and treatment. Dig Dis Sci 1989;34(1):56.

10. Lee SP, Maher K, Nicholls JF. Origin and fate of biliary sludge.

Gastroenterology 1988;94(1):170.

11. Neoptolemos JP, Davidson BR, Winder AF, Vallance D. Role

of duodenal bile crystal analysis in the investigation of ‘idiopathic’

pancreatitis. Br J Surg 1988;75(5):450.

12. Lee YT, Chan FKL, Leung WK, Chan HLY,Wu JCY,Yung MY,

et al. Comparison of EUS and ERCP in the investigation

with suspected biliary obstruction caused by choledocholithiasis:

a randomized study. Gastrointest Endosc2008;67:660-8.

13. Liu CL, Fan ST, Lo CM, Tso WK,Wong Y, Poon RTP, et al.

Comparison of early endoscopic ultrasonography and endoscopic

retrograde cholangiopancreatography in the management

of acute biliary pancreatitis: a prospective randomized study.

Clin Gastroenterol Hepatol2005;3:1238-44.

14. Polkowski M, Regula J, Tilszer A. Endoscopic ultrasound versus

endoscopic retrograde cholangiography for patients with

intermediate probability of bile duct stones: a randomized trial

comparing two management strategies. Endoscopy2007;39:

296-303.